Method for providing automatic selection and ordering services for image- and text-based tiles

Abstract

A device according to one embodiment receives request information for a required tile from a user's terminal, analyzes the request information to extract a feature value of the required tile, vectorizes the feature value to generate a feature vector, calculates the similarity between the tile information corresponding to each tile stored in the tile database and the feature vector, extracts tiles with a similarity greater than or equal to a reference similarity as similar tiles, and if the number of similar tiles is confirmed to be 1, provides a user interface to the user's terminal including tile information of the similar tile and input means for generating a purchase request signal for the similar tile, and if the number of similar tiles is confirmed to be 2 or more, calculates a priority score for each of the multiple similar tiles, and provides a user interface to the user's terminal including a list of similar tiles sorted according to the priority score and selection means for generating a selection and purchase request signal for the similar tile, and if the number of similar tiles is confirmed to be 0, generates a tile production request message including request information for a required tile and transmits it to a tile manufacturer's terminal.

Description

Technology Field

[0001] The following embodiments relate to a technology that converts a tile image or description received by a user into a feature vector to calculate similarity with a tile in a database, automatically links a purchase based on the existence of similar tiles, or, if no similar tiles exist, automatically matches the optimal manufacturer among multiple manufacturers to perform a production request. Background Technology

[0002] As the digital transformation of the architecture and interior design industries accelerates recently, there is an increasing demand to explore and purchase various tile products through online platforms.

[0003] However, it is difficult to specify the tiles users want through simple category selection, and they often seek similar products based on image captures or detailed descriptions.

[0004] Most existing tile search and recommendation systems are limited to simple recommendations based on keyword-based searches or past purchase history. They have limitations in that they cannot provide an integrated process that accurately determines whether a tile requested by a user based on specific images or detailed specifications is actually on sale, and either links to an immediate purchase if it is available or naturally transitions to the production stage if it is not.

[0005] In particular, tiles are materials in which multiple physical and visual elements such as color, pattern, material, size, and use interact in a complex manner; therefore, it is difficult to accurately search for similar products using only simple text matching, and even when production is required, a function to automatically select a suitable manufacturer from among multiple suppliers based on technical criteria has not been provided.

[0006] Therefore, there is a growing need for technology that can quantify and analyze user request information and automate the determination of sales status and conversion to production requests into a single workflow. Prior art literature

[0007] Korean Registered Patent No. 10-2708526 (Published Sep. 20, 2024) Korean Registered Patent No. 10-2716192 (Published Oct. 7, 2024) Korean Registered Patent No. 10-2719669 (Published Oct. 17, 2024) Korean Registered Patent No. 10-2655511 (Published Apr. 5, 2024) The problem to be solved

[0008] The embodiments analyze request information for an image or text-based tile entered by a user to search for tiles currently on sale (distributed) in the database, and if no tile corresponding to the request information exists, automatically switch to a production request procedure for the request.

[0009] The embodiments aim to evaluate multiple similar tiles from various angles by comprehensively reflecting request information and spatial information of the space to which the tile is to be applied, and to calculate a priority based on quantitative criteria according to the evaluation results.

[0010] The embodiments aim to comprehensively evaluate multiple tile manufacturers by utilizing tiles stored in a database and their associated historical information, and to execute production requests via user selection or automatic request methods based on the evaluation results.

[0011] The objectives of the present invention are not limited to those mentioned above, and other unmentioned objectives will be clearly understood from the description below. means of solving the problem

[0012] A method for providing an automatic selection and ordering service for image and text-based tiles comprises: receiving request information for a required tile from a user's terminal, the request information including at least one of image data or text data of the required tile; analyzing the request information to extract a feature value including at least one of the color, pattern, material, specifications, and use of the required tile, and vectorizing the feature value to generate a feature vector; calculating the similarity between tile information corresponding to each tile stored in a tile database and the feature vector, and extracting tiles with a similarity greater than or equal to a preset standard similarity as similar tiles; checking the number of extracted similar tiles; if the number of similar tiles is confirmed to be one, providing a user interface to the user's terminal that includes tile information of the similar tile and input means for generating a purchase request signal for the similar tile; if the number of similar tiles is confirmed to be two or more, calculating a priority score for each of the plurality of similar tiles, and providing a user interface to the user's terminal that includes a list of similar tiles sorted according to the priority score and selection means for generating a selection and purchase request signal for the similar tiles. And if the number of similar tiles is confirmed to be 0, the step of generating a tile production request message containing request information for the necessary tile and transmitting it to the tile manufacturer's terminal.

[0013] The step of calculating a priority score for each of the plurality of similar tiles comprises: for the similar tile, an operation of normalizing the similarity value between the tile information and the feature vector to calculate a similarity score; for the similar tile, an operation of receiving spatial information from the user's terminal including the area, shape, humidity, illuminance, intended use, and construction location of the space where the necessary tile is to be placed; for the similar tile, an operation of comparing the arrangement information of the similar tile with reference arrangement information including the pattern repetition cycle, joint spacing, and construction direction of the necessary tile, and for the similar tile, an operation of calculating an arrangement error value by additionally considering the area and shape included in the spatial information, and an operation of calculating an arrangement distortion minimization score which is converted to a higher value as the arrangement error value is smaller; for the similar tile, an operation of quantifying and calculating the probability of joint contamination occurrence based on the material of the similar tile and the humidity and intended use, and calculating a joint contamination prediction score which is converted to a higher value as the probability of joint contamination occurrence is lower; for the similar tile, an operation of calculating a reflectance reference value based on the illuminance and intended use, and calculating a difference value between the reflectance of the similar tile and the reflectance reference value, and a thermal reflection influence which is converted to a higher value as the difference value is smaller. The operation of calculating a score; for the similar tile, quantifying and calculating the frequency of use according to the intended use and the expected load according to the installation location; calculating the risk of damage by comparing the frequency of use and the expected load with the strength of the similar tile; and calculating a damage risk score that is converted to a higher value as the risk of damage decreases; for the similar tile, calculating a price ratio value by dividing the installation price of the similar tile by a standard price; calculating a period ratio value by dividing the expected installation period of the similar tile by a standard period; and calculating a price score and a delivery period score that are converted to a higher value as the price ratio value and the period ratio value decrease, and the similarity score.It includes an operation to calculate a priority score of similar tiles based on the array distortion minimization score, the joint contamination prediction score, the heat reflection influence score, the damage risk score, the price score, and the delivery time score.

[0014] The step of generating a tile production request message containing the request information for the required tile and transmitting it to the tile manufacturer's terminal comprises: verifying the tile manufacturer using manufacturer identification information stored corresponding to each tile stored in the tile database; selecting, among the tile manufacturers, a tile manufacturer that is matched with a number of tiles exceeding a preset target number as a candidate tile manufacturer; for each candidate tile manufacturer, collecting tile information of tiles produced by the candidate tile manufacturer stored in the tile database; decomposing the feature vector of the required tile into a plurality of attribute elements including material, color, pattern, size, and surface finish; for each candidate tile manufacturer, verifying whether an attribute corresponding to each of the plurality of attribute elements exists in the collected tile information and calculating an attribute implementation possibility score based on the ratio of implementable attribute elements; for each candidate tile manufacturer, verifying whether there is a past tile in which two or more of the plurality of attribute elements are simultaneously implemented and calculating an attribute combination implementation score based on the ratio of simultaneously implemented attribute combinations; and for each candidate tile manufacturer, the order fulfillment history of tiles produced by the candidate tile manufacturer stored in the tile database An operation to analyze and calculate the number of implementations, implementation compliance rate, recent implementation frequency, implementation period variability, and implementation gap period, and to calculate an implementation stability score based thereon; an operation to calculate a comprehensive score of the candidate tile manufacturer based on the attribute implementation feasibility score, the attribute combination implementation score, and the implementation stability score; an operation to select a candidate tile manufacturer whose comprehensive score is higher than a preset target score as a consideration tile manufacturer; an operation to check whether there are multiple consideration tile manufacturers; and if it is confirmed that there are multiple consideration tile manufacturers,The method includes generating a list of tile manufacturers by sorting the tile manufacturers of consideration according to the comprehensive score, providing a user interface to the user's terminal that includes a designation means for generating the list of tile manufacturers and the designation of the tile manufacturers of consideration, receiving a designation signal for the tile manufacturer subject to the production request request from the user's terminal, and transmitting the tile production request message to the terminal of the designated tile manufacturer subject to the production request request, and, if it is confirmed that there is only one tile manufacturer of consideration or no tile manufacturers exist, transmitting the tile production request message first to the terminal of the candidate tile manufacturer with the highest comprehensive score.

[0015] A method for providing automatic selection and ordering services for image and text-based tiles further comprises: a step of verifying the order form of the similar tile upon receiving a purchase request signal for the similar tile from the user's terminal; a step of analyzing the order form to extract order items including quantity, specifications, delivery time, and construction location; a step of analyzing text data and image data included in the request information to extract item values ​​corresponding to the order items; a step of automatically entering the extracted item values ​​to correspond to the order items to generate a draft order; a step of providing the generated draft order to the user's terminal; and a step of modifying or approving the draft order according to user input received from the user's terminal.

[0016] A method for providing automatic selection and ordering services for image and text-based tiles further comprises the steps of: receiving spatial information from the user's terminal upon receiving a purchase request signal for the similar tile; generating a spatial model in which the tile is to be installed based on the spatial information; identifying an application area in the spatial model, which is an area where the tile is to be applied; generating a first construction simulation by applying the necessary tile to the application area; generating a second construction simulation by applying the similar tile to the application area; dividing the application area into a plurality of subdivided areas according to a preset standard; calculating the difference size for each area by comparing the first construction simulation and the second construction simulation for each of the plurality of subdivided areas; identifying an area having a difference size higher than a preset standard difference using the difference size for each area as a difference area; determining display attributes based on the difference area; and providing the first construction simulation and the second construction simulation to the user's terminal so that the display attributes are reflected.

[0017] The step of determining the display attribute includes: checking the difference size of the difference area; displaying the difference area with a first saturation if the difference size of the difference area is greater than or equal to a preset threshold difference, and displaying the difference area with a second saturation lower than the first saturation if the difference size of the difference area is less than the threshold difference; checking the number of the difference area; and displaying the ratio of the second construction simulation as greater than the first construction simulation if the number of the difference area is greater than or equal to a preset threshold number, and displaying the first construction simulation and the second construction simulation at the same ratio if the number of the difference area is less than the threshold number.

[0018] A method for providing automatic selection and ordering services for image and text-based tiles comprises the steps of: receiving a purchase request signal for the similar tile from the user's terminal, verifying the requested quantity of the required tile included in the request information; querying the inventory based on the tile information of the similar tile; calculating an additional requested quantity by subtracting the inventory from the requested quantity based on the fact that the inventory of the similar tile is less than the requested quantity; searching the tile database and selecting the tile with the highest similarity to the similar tile as the additional similar tile among the tiles having inventory capable of corresponding to the additional requested quantity; receiving spatial information from the user's terminal and verifying the construction location where the tile is to be installed; determining whether the similar tile and the additional similar tile can be installed at different construction locations based on the construction location, the inventory of the similar tile, and the additional requested quantity; if it is determined that the similar tile and the additional similar tile can be installed at different construction locations, setting each of the separated construction locations as an independent construction zone; and matching the similar tile and the additional similar tile to each of the construction zones to generate matching information. A step of providing a user interface to the user's terminal, comprising, together with the matching information, tile information of the additional similar tile and input means for generating a purchase request signal for the additional similar tile; and a step of, if it is determined that the similar tile and the additional similar tile cannot be installed at different installation locations, generating a cross-arrangement pattern such that the similar tile and the additional similar tile are mixed and arranged based on the inventory of the similar tile and the additional request quantity.and further includes the step of providing a user interface to the user's terminal, comprising, together with the cross-arrangement pattern, tile information of the additional similar tile and input means for generating a purchase request signal for the additional similar tile.;

[0019] A device according to one embodiment may be combined with hardware and controlled by a computer program stored on a medium to execute the method of any one of the methods described above. Effects of the invention

[0020] The embodiments analyze request information for an image or text-based tile entered by a user to search for tiles currently on sale (distributed) in the database, and if no tile corresponding to the request information exists, can automatically switch to a production request procedure for the request.

[0021] The embodiments comprehensively reflect the request information and the spatial information of the space to which the tile is to be applied to evaluate multiple similar tiles from various angles, and can calculate a priority based on quantitative criteria according to the evaluation results.

[0022] The embodiments utilize tiles stored in a database and associated historical information to comprehensively evaluate multiple tile manufacturers, and based on the evaluation results, can perform production requests through user selection or automatic request methods.

[0023] Meanwhile, the effects according to the embodiments are not limited to those mentioned above, and other unmentioned effects will be clearly understood by those skilled in the art from the description below. Brief explanation of the drawing

[0024] FIG. 1 is a schematic diagram showing the configuration of a system according to one embodiment. FIG. 2 is a flowchart illustrating the process of providing an automatic selection and ordering service for image and text-based tiles according to one embodiment. FIG. 3 is a flowchart illustrating the process of calculating the priority score of similar tiles according to one embodiment. FIG. 4 is a flowchart illustrating the process of calculating a comprehensive score to select a tile manufacturer according to one embodiment. FIG. 5 is a flowchart for explaining the process of sending a tile production request message to a tile manufacturer, taking into account the comprehensive score according to one embodiment. FIG. 6 is a flowchart illustrating the process of automatically generating a purchase order according to one embodiment. FIG. 7 is a flowchart illustrating the process of applying and providing a simulation according to one embodiment. FIG. 8 is a flowchart illustrating the process of determining display attributes according to one embodiment. FIG. 9 is a flowchart illustrating the process of utilizing multiple tiles when the tile inventory is insufficient according to one embodiment. FIG. 10 is an example diagram of the configuration of a device according to one embodiment. Specific details for implementing the invention

[0025] Hereinafter, embodiments are described in detail with reference to the attached drawings. However, various modifications may be made to the embodiments, and thus the scope of the patent application is not limited or restricted by these embodiments. It should be understood that all modifications, equivalents, and substitutions to the embodiments are included within the scope of the rights.

[0026] Specific structural or functional descriptions of the embodiments are disclosed for illustrative purposes only and may be modified and implemented in various forms. Accordingly, the embodiments are not limited to the specific disclosed forms, and the scope of this specification includes modifications, equivalents, or substitutions that fall within the technical concept.

[0027] Terms such as "first" or "second" may be used to describe various components, but these terms should be interpreted solely for the purpose of distinguishing one component from another. For example, the first component may be named the second component, and similarly, the second component may be named the first component.

[0028] When it is stated that a component is "connected" to another component, it should be understood that it may be directly connected to or joined to that other component, or that there may be other components in between.

[0029] The terms used in the embodiments are for illustrative purposes only and should not be interpreted as intended to be limiting. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this specification, terms such as "comprising" or "having" are intended to indicate the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0030] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which the embodiments pertain. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this application.

[0031] In addition, when describing with reference to the attached drawings, identical components are assigned the same reference numeral regardless of drawing symbols, and redundant descriptions thereof are omitted. When describing embodiments, if it is determined that a detailed description of related prior art could unnecessarily obscure the essence of the embodiment, such detailed description is omitted.

[0032] The embodiments can be implemented in various forms of products such as personal computers, laptop computers, tablet computers, smartphones, televisions, smart home appliances, intelligent automobiles, kiosks, and wearable devices.

[0033] FIG. 1 is a schematic diagram showing the configuration of a system according to one embodiment.

[0034] Referring to FIG. 1, a system according to one embodiment may include a user terminal (100) and a device (200) capable of communicating with each other through a communication network.

[0035] First, the communication network can be configured regardless of the mode of communication, such as wired or wireless, and can be implemented in various forms to enable communication between servers and between servers and terminals.

[0036] The user's terminal (100) is a terminal used by a user who wishes to purchase or request the production of a desired tile, and can be implemented as a mobile phone, desktop PC, laptop PC, tablet PC, smartphone, etc., but is not limited thereto, and can be implemented as various types of communication devices that can be connected to an external server. For example, as shown in FIG. 1, the user's terminal (100) may be a smartphone, and may be adopted differently depending on the embodiment.

[0037] The user's terminal (100) may be configured to perform all or part of the computational functions, storage / reference functions, input / output functions, and control functions that a conventional computer has. The user's terminal (100) may be configured to communicate with the device (200) via wired or wireless means.

[0038] The user's terminal (100) may be connected to a website operated by a person or organization providing a service using the device (200), or may have an application developed and distributed by a person or organization providing a service using the device (200) installed. The user's terminal (100) may be linked with the device (200) through the website or application.

[0039] In FIG. 1 and the description below, for convenience of explanation, only one user terminal (100) is illustrated and described, but the number of terminals can vary depending on the embodiment. As long as the processing capacity of the device (200) allows, there is no particular limit to the number of terminals.

[0040] The device (200) may be a private server owned by a person or organization providing a service using the device (200), a cloud server, or a peer-to-peer (P2P) set of distributed nodes. The device (200) may be configured to perform all or part of the computational functions, storage / reference functions, input / output functions, and control functions that a conventional computer possesses. The device (200) may be configured to communicate with the user's terminal (100) via wired or wireless means.

[0041] Additionally, the device (200) can collect information by communicating wirelessly or via wired connection with external websites, including blogs, cafes, social media, or articles, and can utilize the collected information.

[0042] The device (200) receives request information for a required tile from a user's terminal (100), analyzes the request information to extract a feature value of the required tile, and can vectorize the feature value to generate a feature vector. The device (200) calculates the similarity between the tile information corresponding to each tile stored in the tile database and the feature vector, and can extract tiles with a similarity greater than or equal to a reference similarity as similar tiles. The device (200) checks the number of extracted similar tiles, and if the number of similar tiles is 1, it can provide a user interface to the user's terminal (100) that includes tile information of the similar tile and an input means for generating a purchase request signal for the similar tile; if the number of similar tiles is 2 or more, it can calculate a priority score for each of the multiple similar tiles and provide a user interface to the user's terminal (100) that includes a list of similar tiles sorted according to the priority score and a selection means; and if the number of similar tiles is 0, it can generate a tile production request message that includes request information for a required tile and transmit it to the tile manufacturer's terminal.

[0043] Furthermore, the device (200) may also communicate with the tile manufacturer's terminal via wired or wireless means.

[0044] At this time, the terminal of the tile manufacturer is a terminal used by a tile manufacturer that manufactures, sells, or distributes tiles, and may be implemented as a mobile phone, desktop PC, laptop PC, tablet PC, smartphone, etc., but is not limited thereto, and may be implemented as various electronic devices capable of communicating with an external server.

[0045] The terminal of the tile manufacturer may be configured to perform all or part of the computational functions, storage and reference functions, input / output functions, and control functions that a conventional computer has, and may receive a tile production request message transmitted from the device (200), generate a response including whether production is possible or related information, and transmit it to the device (200).

[0046] In addition, terminals of multiple tile manufacturers can communicate simultaneously with the device (200), and the number is not limited.

[0047] The device (200) can communicate with the tile database via wired or wireless means. The tile database is a data storage for storing various information related to tiles, and can be managed directly by the device (200) or distributed and stored on an external storage server.

[0048] The tile database may store tile information for each tile, and said tile information may include, but is not limited to, the tile's material, color, pattern, specifications, surface finish, reflectance, strength, construction cost, construction period, etc. Additionally, manufacturer identification information corresponding to each tile may be stored in conjunction.

[0049] Furthermore, the tile database may store the production history, delivery history, and production delay history of tiles produced by each tile manufacturer in a linked manner, either on a tile unit or a manufacturer unit. Accordingly, the device (200) can analyze the production capabilities and delivery history of a specific tile manufacturer by referring only to the tile database, without communicating with a separate database dedicated to the manufacturer.

[0050] The tile database can be implemented in various forms, such as a relational database, a NoSQL database, or a distributed data storage structure, and can be configured differently depending on the embodiment.

[0051] In the present invention, Artificial Intelligence (AI) refers to a technology that imitates human learning ability, reasoning ability, and perceptual ability, and implements them on a computer, and may include concepts such as machine learning and symbolic logic. Machine Learning (ML) is an algorithmic technology that classifies or learns the characteristics of input data on its own. AI technology can analyze input data as a machine learning algorithm, learn from the results of the analysis, and make judgments or predictions based on the results of the learning. Furthermore, technologies that mimic the functions of the human brain, such as cognition and judgment, by utilizing machine learning algorithms can also be understood as falling within the category of AI. For example, technological fields such as linguistic understanding, visual understanding, reasoning / prediction, knowledge representation, and motion control may be included.

[0052] Machine learning can refer to the process of training neural network models using experience in processing data. It implies that through machine learning, computer software improves its own data processing capabilities. A neural network model is constructed by modeling the correlations between data, and these correlations can be expressed by multiple parameters. A neural network model extracts and analyzes features from given data to derive correlations between them; machine learning can be defined as the process of optimizing the model's parameters by repeating this process. For example, a neural network model can learn the mapping (correlation) between inputs and outputs for data given as input-output pairs. Alternatively, even when only input data is provided, a neural network model can derive regularities between the given data and learn those relationships.

[0053] An artificial intelligence learning model or neural network model can be designed to implement the structure of the human brain on a computer and may include multiple network nodes that have weights and simulate neurons of a human neural network. The multiple network nodes may have interconnected relationships by simulating the synaptic activity of neurons, where neurons exchange signals through synapses. In an artificial intelligence learning model, multiple network nodes may be located in layers of different depths and exchange data according to convolutional connections. The artificial intelligence learning model may be, for example, an Artificial Neural Network (ANN) or a Convolutional Neural Network (CNN). As an embodiment, the artificial intelligence learning model may be machine learned according to methods such as supervised learning, unsupervised learning, and reinforcement learning. Machine learning algorithms for performing machine learning may include Decision Tree, Bayesian Network, Support Vector Machine, Artificial Neural Network, Ada-boost, Perceptron, Genetic Programming, and Clustering.

[0054] Among these, CNNs are a type of multilayer perceptron designed to use minimal preprocessing. CNNs consist of one or more convolutional layers and standard artificial neural network layers stacked on top, additionally utilizing weights and pooling layers. Thanks to this structure, CNNs can fully utilize two-dimensional input data. Compared to other deep learning architectures, CNNs demonstrate good performance in both image and audio fields. CNNs can also be trained using standard backpropagation. CNNs have the advantage of being easier to train than other feedforward artificial neural network techniques and using a small number of parameters.

[0055] Convolutional networks are neural networks comprising sets of nodes with bounded parameters. Many computer vision tasks have been significantly improved, driven by the increased size of available training data and the availability of computational power, combined with algorithmic advancements such as discriminative linear units and dropout training. In the case of massive datasets, such as those available for many tasks today, outfitting is not critical, and increasing the network size improves test accuracy. Optimal utilization of computing resources becomes a limiting factor. To address this, distributed, scalable implementations of deep neural networks can be employed.

[0056] FIG. 2 is a flowchart illustrating the process of providing an automatic selection and ordering service for image and text-based tiles according to one embodiment.

[0057] Referring to FIG. 2, first, in step S201, the device (200) can receive request information for a required tile from a user's terminal (100), which includes at least one of image data or text data of the required tile.

[0058] Specifically, the device (200) can receive a data packet containing request information from a user's terminal (100) via HTTP, HTTPS, WebSocket, API call, or other network protocols, and the received data can be stored in the database of the device (200).

[0059] Here, image data may include a photo taken by the user, a captured image of an existing tile, an interior reference image, a catalog image, or a scanned image, and may be received in the form of a still image file format such as JPEG, PNG, BMP, TIFF, etc., or a frame image extracted from a video. Additionally, image data may be transmitted by being captured in real-time through a camera provided in the user's terminal (100) and transmitted in a streaming manner, or by selecting and uploading an image file stored in the terminal.

[0060] The device (200) can perform a preprocessing process to verify the resolution, color space, file corruption, etc. of the received image data, and can perform image resizing, noise removal, color normalization, or decompression processes as needed.

[0061] Additionally, text data may include sentences, keywords, hashtags, specification information, color codes (RGB, CMYK, HEX, etc.), material names, or usage information entered directly by the user, and may be received in the form of natural language or structured input fields. For example, the text data may be a natural language sentence such as “beige matte marble pattern 600×600 floor tiles,” or it may be structured data in which color, specification, usage, etc. are entered separately in individual input fields.

[0062] The device (200) can perform string normalization, stop word removal, unit conversion (e.g., mm ↔ cm), special character removal, or morphological analysis on the received text data, and in the case of input field-based data, can separate values ​​for each field and map them to an internal data structure.

[0063] The request information may be received in the form of composite data containing both image data and text data. In this case, the device (200) may process the image data and text data independently and then combine them complementarily to form integrated request information. For example, color information extracted from an image and specification information extracted from text can be combined to create a single request object.

[0064] Additionally, the request information may be converted into JSON, XML, multipart / form-data, or other structured data formats and transmitted, and the device (200) may parse the received data and convert it into a key-value structure or an object structure.

[0065] The device (200) can perform an error handling procedure that verifies whether there are required items in the received request information and requests the user to re-enter if there is error or missing data.

[0066] In step S202, the device (200) can analyze the request information to extract a feature value including at least one of the color, pattern, material, specifications, and use of the required tile, and vectorize the feature value to generate a feature vector.

[0067] Specifically, the device (200) can extract feature values ​​by executing different analysis algorithms according to the type of request information based on request information including at least one of image data and text data received from the user's terminal (100) in step S201.

[0068] First, when image data is included in the request information, the device (200) can utilize a commonly used image processing algorithm to perform preprocessing on the image data, such as color space conversion, noise removal, and resolution correction, and then extract color features based on the pixel values ​​of the image data. At this time, the color features can be converted into RGB, HSV, or LAB color spaces and then quantified in the form of an average color value, a major color cluster center value, or a color histogram distribution.

[0069] Additionally, the device (200) can extract pattern features by performing edge detection, periodicity analysis, or texture analysis on image data, and can quantify the directionality, repetition period, and geometric arrangement information of the pattern using, for example, a Gabor filter, LBP (Local Binary Pattern), HOG (Histogram of Oriented Gradients), or a convolutional neural network-based feature extraction technique.

[0070] Additionally, the device (200) can extract material features by analyzing surface texture, brightness contrast, and reflective characteristics within image data, and can classify them into material types such as marble, ceramic, porcelain, and wood patterns using a pre-trained classification model or rule-based classification logic.

[0071] Additionally, the device (200) can recognize tile objects within image data and estimate specifications by calculating the width-to-height ratio of the tiles using the relative size of the objects or reference objects included in the image.

[0072] Meanwhile, if the request information includes text data, the device (200) can utilize a commonly used natural language processing algorithm to break down the text data into tokens and perform morphological analysis, part-of-speech tagging, named entity recognition, or keyword extraction to derive attribute candidates.

[0073] To this end, the device (200) may store an attribute classification dictionary or a learned text classification model corresponding to color, pattern, material, specification, surface finish, and use in a database provided in the device (200), and can determine which attribute category the token belongs to by comparing each extracted token with the attribute classification dictionary or classification model.

[0074] For example, if a token named “beige” is included in the text data, the device (200) can classify the token as a color attribute by comparing it with a list of color names stored in a color dictionary. Additionally, “matte” can be classified as a surface finish attribute by confirming that it corresponds to a surface finish attribute dictionary.

[0075] Additionally, when a composite expression such as “marble pattern” is extracted, the device (200) may decompose it into “marble” and “pattern,” or classify it into an attribute that includes both material attributes and pattern attributes by referring to a composite attribute dictionary.

[0076] Additionally, when an expression including numbers and separators, such as “600×600”, is extracted, the device (200) can recognize the dimension format using a regular expression and classify it as a specification attribute, and can apply unit correction logic as needed.

[0077] In addition, words indicating space or purpose of use, such as “floor,” “wall,” and “bathroom,” can be classified as use attributes by confirming that they correspond to the use attribute dictionary.

[0078] That is, the device (200) can identify attributes by using an attribute classification dictionary that is pre-stored in a database provided in the device (200).

[0079] Additionally, if the request information includes both image data and text data, the device (200) may execute an image processing algorithm and a natural language processing algorithm in parallel or sequentially to extract respective feature values, and then integrate the feature values. In this process, if specification information extracted from text data is applied prior to the image analysis result, or if the color distribution extracted from the image conflicts with the color keyword extracted from the text, a pre-set confidence-based weight can be applied to determine the final feature value.

[0080] The device (200) can normalize each feature value extracted in this way into a numeric, categorical, or binary value form, and can generate a multidimensional feature vector by converting the feature value into a coordinate value in a vector space.

[0081] For example, the device (200) can convert color values ​​into continuous numeric vectors, patterns and materials into one-hot encoding or embedding vector forms, and specifications into normalized length values.

[0082] Additionally, the device (200) may generate a weighted feature vector by assigning a preset importance weight to each feature element, and the feature vector may be stored in the memory of the device (200) or provided as an input value for calculating similarity with tile information stored in a tile database in a subsequent step.

[0083] In step S203, the device (200) calculates the similarity between the tile information and the feature vector corresponding to each tile stored in the tile database, and can extract tiles that are greater than or equal to a preset reference similarity as similar tiles.

[0084] Specifically, the device (200) can calculate the similarity with tile information corresponding to each of a plurality of tiles stored in a tile database by using the feature vector generated in step S202 as an input value.

[0085] To this end, the device (200) can communicate with a tile database via wired or wireless means to query tile information regarding tiles currently on sale (distribution). At this time, attribute values ​​corresponding to the color, pattern, material, specifications, and use of each tile may be stored in the tile information, but are not limited thereto. Additionally, the attribute values ​​may be stored in the form of numeric, categorical, or binary values.

[0086] The device (200) can generate or extract a comparison target feature vector having the same dimensional structure (attribute) as the feature vector generated in step S202 from the tile information.

[0087] At this time, the device (200) can perform attribute alignment so that each attribute dimension of the feature vector of the request tile and the feature vector to be compared have the same semantic system, and can perform unit conversion or normalization if the units are different.

[0088] The device (200) may apply a distance function or a similarity function to calculate the similarity between the feature vector and the comparison target feature vector. For example, a cosine similarity, Euclidean distance, Manhattan distance, dot product-based similarity calculation method, or a combination thereof may be applied, but is not limited thereto.

[0089] In the case of cosine similarity, the device (200) can calculate the similarity by dividing the inner product of two vectors by the product of the magnitudes (norms) of each vector. In the case of Euclidean distance, the distance value can be calculated by squaring the difference between each dimension value of two vectors, summing them, and taking the square root, and the device (200) can convert the distance value into a similarity value by inverse transformation or linear normalization.

[0090] Additionally, the device (200) may apply weights to reflect the importance of each attribute element corresponding to each dimension of the feature vector. For example, a first weight may be applied to a value corresponding to the color dimension and a second weight may be applied to the pattern dimension, and the weights may be pre-set or dynamically adjusted according to user settings or learning results.

[0091] Additionally, the device (200) can distinguish between categorical attributes and continuous attributes and apply different similarity calculation methods. For example, categorical attributes such as material or pattern may be assigned a value of 1 or 0 depending on whether they match or cosine similarity between embedding vectors may be calculated, and continuous attributes such as specification or color may have distance values ​​calculated based on numerical differences.

[0092] The similarity values ​​for each attribute calculated in this way can be integrated to calculate a final similarity score, and the device (200) can derive a final similarity score by weighting and summing or averaging multiple similarity values ​​for each attribute.

[0093] The device (200) can normalize the final similarity score to a range of 0 to 1 and compare it with a preset reference similarity value. The reference similarity value may be a preset fixed value, or it may be dynamically set based on the overall tile data distribution, average similarity, or a certain percentage of similarity values.

[0094] The device (200) can determine that a tile whose final similarity score is greater than or equal to the reference similarity score is a similar tile, and can generate a list of similar tiles by extracting identification information of the tile.

[0095] In step S204, the device (200) can check the number of extracted similar tiles.

[0096] Specifically, the device (200) can calculate the total number of similar tiles by counting the number of similar tiles whose similarity extracted in step S203 is greater than or equal to the reference similarity.

[0097] The device (200) can determine whether to perform a user interface provision method or a production request procedure in a subsequent step based on the number of final similar tiles calculated as above.

[0098] In step S205, if the device (200) confirms that the number of similar tiles is 1, it may provide a user interface to the user's terminal (100) that includes tile information of the similar tile and input means for generating a purchase request signal for the similar tile.

[0099] Specifically, when the device (200) checks the number of similar tiles in step S204 and confirms that the number of similar tiles is 1, it can look up the tile information of the similar tile by referring to the tile database.

[0100] At this time, the tile information may include at least one of a representative image, color, pattern, material, specifications, surface finish, price information, stock quantity information, delivery availability information, and manufacturer information of the tile, but is not limited thereto.

[0101] The device (200) can convert the queried tile information into screen display data to configure a user interface screen, and the user interface can be transmitted to a user's terminal (100) via a network and displayed on a display provided on the user's terminal (100).

[0102] The user interface may include an input means for generating a purchase request signal for the similar tile, along with a tile information area of ​​the similar tile. The input means may be implemented in the form of a button, a touch area, a selection box, or other graphic user interface elements, and may be selected through touch input, mouse click input, or key input of the user's terminal (100).

[0103] When a user selects the input means through the user's terminal (100), the user's terminal (100) can generate purchase request data corresponding to the selection event and transmit it to the device (200). The device (200) can receive the purchase request data and determine that the user has expressed an intention to purchase the similar tile.

[0104] At this time, the purchase request data may include at least one of identification information of the similar tile, selection quantity information, user identification information, and delivery address information, and the purchase request data may be used as input data for performing payment processing, order creation, or inventory deduction procedures.

[0105] Additionally, the device (200) may include an area in the user interface that visually highlights an attribute item matching the feature vector of the required tile, or displays the degree of matching with the request information in a numerical or graphical form.

[0106] In this way, the device (200) can provide a user interface configured to allow the user to immediately make a purchase decision for the tile when a similar tile is identified as a single item.

[0107] In step S206, if the device (200) confirms that the number of similar tiles is two or more, it may calculate a priority score for each of the multiple similar tiles and provide a user interface to the user's terminal (100) that includes a list of similar tiles sorted according to the priority score and a selection means for generating a selection and purchase request signal for the similar tiles.

[0108] Specifically, the device (200) can calculate a priority score based on a plurality of predefined judgment elements for each similar tile extracted in step S203 if, as a result of checking the number of similar tiles in step S204, the number of similar tiles is confirmed to be two or more.

[0109] The above priority score can be calculated by quantifying a plurality of factors including at least one of the relevance to the request information of each similar tile, spatial suitability, suitability for the usage environment, price conditions, and delivery conditions, and the scores for each factor can be integrated through weighted summation or average calculation.

[0110] Refer to Figure 3 for an explanation related to this.

[0111] The device (200) can generate a list of similar tiles by sorting a plurality of similar tiles in descending or ascending order according to the calculated priority score.

[0112] The above similar tile list may include tile information and priority scores for each similar tile, and the priority scores may be displayed in numerical form, grade form, or visual graph form.

[0113] Additionally, the device (200) may include a selection means corresponding to each of a plurality of similar tiles in the user interface. The selection means may be implemented in the form of a selection button, a comparison checkbox, or a purchase request button corresponding to each tile item, and may be selected through touch input or click input of the user's terminal (100).

[0114] When a user selects at least one of a plurality of similar tiles through the user's terminal (100), the user's terminal (100) may generate purchase request data including identification information of the selected tile and transmit it to the device (200). The device (200) may receive the purchase request data, determine that there is an intention to purchase the selected tile, and perform an order creation or payment procedure.

[0115] Additionally, the device (200) may further include a comparison display area configured to enable attribute comparison between a plurality of similar tiles in the user interface, and the comparison display area may display the degree of matching with the color, material, specifications, price, or request information of each tile in a table form or a graphic form in parallel.

[0116] In this way, the device (200) can provide a user interface configured to allow the user to select the optimal tile among multiple candidates by providing a priority-based sorting structure when multiple similar tiles are identified.

[0117] In step S207, if the device (200) confirms that the number of similar tiles is 0, it can generate a tile production request message containing request information for necessary tiles and transmit it to the terminal of the tile manufacturer.

[0118] Specifically, if the device (200) determines that there are no similar tiles after checking the number of similar tiles in step S204, it can generate a structured tile production request message for a production request based on the request information received in step S201 and the feature vector generated in step S202.

[0119] At this time, the tile production request message may include a feature vector and user identification information that is at least one of the color, pattern, material, specifications, surface finish, and use of the required tile confirmed through the request information, and the information may be generated by converting it into JSON, XML, or other structured data formats.

[0120] The device (200) can transmit the tile production request message generated as above to the tile manufacturer's terminal via a network.

[0121] At this time, the tile manufacturer may be selected based on manufacturer identification information stored in the tile database or may be automatically selected according to pre-set criteria.

[0122] Refer to FIGS. 4 and FIGS. 5 for an explanation related to this.

[0123] The terminal of the above tile manufacturer can receive the tile production request message from the device (200), and accordingly, can generate a response message including whether production is possible, the estimated production period, or quotation information and transmit it to the device (200).

[0124] The device (200) can receive the response message and provide it to the user, and the user can finally approve whether to proceed with the production request.

[0125] In this way, the device (200) can be configured to automatically perform a production request procedure based on the request information for the required tile, even if there are no similar tiles currently on sale.

[0126] As a result, the device (200) can implement an integrated tile selection and ordering process that can satisfy the user's request without interruption by analyzing the image or text-based tile request information entered by the user, prioritizing the search for tiles currently on sale, supporting purchase according to priority if similar tiles exist, and automatically switching to a production request procedure if similar tiles do not exist.

[0127] FIG. 3 is a flowchart illustrating the process of calculating the priority score of similar tiles according to one embodiment.

[0128] Referring to FIG. 3, first, in step S301, the device (200) can calculate a similarity score for similar tiles by normalizing the similarity value between tile information and feature vector.

[0129] Specifically, the device (200) can define tiles determined to be similar to or greater than a reference similarity in step S203 as similar tiles and calculate a similarity score for each similar tile.

[0130] To this end, the device (200) can check the similarity value calculated in step S203 for each similar tile and normalize the similarity value to enable comparison between different tiles to generate a similarity score.

[0131] The device (200) may perform scaling or normalization operations to convert the similarity value into a range of 0 or more and 1 or less.

[0132] For example, the device (200) can perform min-max normalization based on the maximum and minimum values ​​of similarity values ​​included in a set of similar tiles, and the similarity score can be calculated according to a formula such as 'similarity score = (current similarity value - minimum similarity value) / (max similarity value - minimum similarity value)'.

[0133] Additionally, the device (200) may apply various normalization techniques, such as a mean and standard deviation-based normalization method or a log scaling method, but is not limited thereto.

[0134] The device (200) can determine the normalized value as described above as a similarity score for the corresponding similar tile, and the higher the similarity, the higher the similarity score can be calculated.

[0135] The similarity score calculated in this way can be used as one of the judgment factors for calculating the priority score in the subsequent stage.

[0136] In step S302, the device (200) can receive spatial information from the user's terminal (100), including the area, shape, humidity, illuminance, intended use, and construction location of the space where the necessary tiles are to be placed.

[0137] Specifically, the device (200) can receive spatial information regarding a target space where a necessary tile is to be installed from a user's terminal (100). At this time, the spatial information is data representing the physical characteristics and environmental conditions of the target space, and may include area, shape, humidity, illuminance, intended use, and construction location.

[0138] Additionally, spatial information may include data directly input by the user or measured through a sensor equipped in the user's terminal (100) or a measuring device that communicates with the user's terminal (100) via wired or wireless means.

[0139] First, the device (200) can receive the area of ​​the target space. At this time, the area may be a value representing the floor area or wall area of ​​the space where the tile is to be installed, and may be entered in, for example, square meters (m^2), square feet (ft^2), or other area units. Additionally, the area may be a value entered directly by the user, or a value measured using a space measurement application or a distance measurement sensor equipped on the user's terminal (100).

[0140] Additionally, the device (200) can receive the shape of the target space. In this case, the shape may be information representing the planar structure of the space, and may be expressed as, for example, a rectangle, a square, a polygon, or an irregular structure. Additionally, the shape information may include dimensional data including length and width values ​​of the space, or may include space structure data generated based on a space image captured through the user's terminal (100) or space scan data.

[0141] Additionally, the device (200) can receive the humidity of the target space. At this time, the humidity may be a value representing the relative humidity of the target space, and may be expressed, for example, in a percentage (%) unit. The humidity may be a value measured through a humidity measuring device that communicates wirelessly or wiredly with the user's terminal (100).

[0142] Additionally, the device (200) can receive the illuminance of the target space. In this case, the illuminance may be a value representing the brightness of light incident on the target space, and may be expressed in units of lux, for example. The illuminance may be a value measured through an illuminance sensor, or a value estimated based on the window area of ​​the target space, the number of lighting fixtures, or natural lighting conditions.

[0143] Additionally, the device (200) can receive the intended use of the target space. At this time, the intended use may indicate whether the target space is used as a bathroom, kitchen, living room, bedroom, hallway, commercial space, or outdoor space, and the device (200) can estimate the frequency of water usage, frequency of movement, or environmental conditions that may occur in the space based on the intended use.

[0144] Additionally, the device (200) can receive a construction location where a tile is to be installed. At this time, the construction location may be information indicating the location of the structure where the tile is to be installed, and may be entered as, for example, a floor, a wall, a staircase, an object surface, or any other structural surface. Based on the construction location, the device (200) can estimate the expected load or usage conditions that the tile will receive.

[0145] The device (200) can receive the spatial information from the user's terminal (100) in JSON, XML, or other structured data formats, and can parse the received data and store it in an internal data structure.

[0146] The spatial information received as described above, including area, shape, humidity, illuminance, intended use, and construction location, can be stored by the device (200) or used as input data for evaluating similar tiles in a subsequent step.

[0147] In step S303, the device (200) compares the arrangement information of the similar tile with reference arrangement information including the pattern repetition cycle, joint spacing, and construction direction of the required tile for the similar tile, calculates an arrangement error value by additionally considering the area and shape included in the spatial information, and calculates an arrangement distortion minimization score which is converted to a higher value as the arrangement error value becomes smaller.

[0148] Specifically, the device (200) can derive reference arrangement information of the required tiles from the feature vector or request information generated in step S202.

[0149] At this time, the reference arrangement information may include at least one of the tile pattern repetition cycle, the joint spacing between tiles, and the tile installation direction.

[0150] For example, the pattern repetition period may refer to the length interval during which a tile surface pattern is repeated, the joint spacing may refer to the construction interval formed between adjacent tiles, and the construction direction may refer to the direction in which the tile pattern or arrangement is aligned according to a reference.

[0151] Additionally, the device (200) can check the arrangement information of the similar tiles by referring to the tile information of each similar tile stored in the tile database.

[0152] The above array information may include the pattern repetition cycle of similar tiles, recommended joint spacing, and recommended construction direction to correspond to the reference array information.

[0153] The device (200) can analyze the arrangement difference by comparing the arrangement information of the reference arrangement information and the arrangement information of the similar tile. For example, the device (200) can calculate a pattern repetition error based on the difference between the pattern repetition period included in the reference arrangement information and the pattern repetition period of the similar tile, the device (200) can calculate a joint spacing error based on the difference between the reference joint spacing and the recommended joint spacing of the similar tile, and the device (200) can calculate a construction direction error based on the angle difference between the reference construction direction and the recommended construction direction of the similar tile.

[0154] Additionally, the device (200) can analyze the impact of the pattern repetition error, joint spacing error, and construction direction error on the actual construction result by using the area and shape included in the spatial information received in step S302. In this process, the device (200) calculates the number of repetitions of the tile arrangement by comparing the length and width of the space with the pattern repetition cycle, and can analyze the possibility of pattern discontinuity or abnormally cut tiles occurring when the number of repetitions is not formed as an integer multiple.

[0155] For example, the device (200) calculates the number of array repetitions by dividing the length or width of the space included in the space information by the pattern repetition period, and determines that the larger the fractional part of the number of repetitions, the higher the probability that a pattern break will occur at the space boundary, and can correct the pattern repetition error by applying a pre-set weight to the pattern repetition error.

[0156] In addition, if the spatial shape is identified as polygonal or irregular, the device (200) can further correct the construction direction error by analyzing the deviation between the direction of the spatial boundary line and the reference construction direction.

[0157] Additionally, the device (200) determines that the larger the spatial area, the higher the likelihood that the arrangement mismatch will be visually perceived, calculates an influence coefficient based on the spatial area, and applies this as a weight to the pattern repetition error, joint spacing error, and construction direction error.

[0158] The device (200) can calculate a corrected arrangement error element by applying an influence coefficient based on spatial shape and area to the pattern repetition error, joint spacing error, and construction direction error calculated as above. Subsequently, the device (200) can calculate a final arrangement error value by performing a weighted summation or averaging operation on the corrected pattern repetition error, corrected joint spacing error, and corrected construction direction error.

[0159] The device (200) can calculate an array distortion minimization score by performing a score conversion operation such that the smaller the array error value, the higher the value is converted. For example, the device (200) can perform an inverse conversion or linear normalization operation of the array error value to calculate a higher score as the array error value becomes smaller.

[0160] The array distortion minimization score calculated in this way can be used as one of the judgment factors to calculate the priority score of similar tiles in a subsequent step.

[0161] In step S304, the device (200) can quantify and calculate the probability of joint contamination occurring for similar tiles based on the material, humidity, and intended use of the similar tiles, and can calculate a joint contamination prediction score that is converted to a higher value as the probability of joint contamination occurring is lower.

[0162] Specifically, the device (200) can identify the material of the similar tile by referring to the tile information of the similar tile stored in the tile database.

[0163] In this case, the material may include, but is not limited to, ceramic tiles, porcelain tiles, marble tiles, stone tiles, or other surface materials, and may additionally include properties indicating the surface absorption rate, surface roughness, or whether the surface is coated of the tile.

[0164] For example, the device (200) may determine that the more the material has many micropores, such as limestone-based stone tiles or unglazed ceramic tiles, that is, the higher the surface absorption rate, the higher the likelihood that water or contaminants will remain on the tile surface or joint area.

[0165] Additionally, the device (200) may be judged to be more likely to have contaminants attached to it if the material has a large surface roughness, such as stone tiles with a rough finish, natural stone tiles of the slate series, or embossing tiles with an emphasized surface texture.

[0166] Additionally, the device (200) may be judged to have relatively suppressed adhesion of contaminants in the case of tiles with surface coatings, such as ceramic tiles with glaze, porcelain tiles with nano-coating or water-repellent coating.

[0167] Additionally, the device (200) can check the humidity included in the spatial information received in step S302.

[0168] In this case, humidity may be a value representing the relative humidity of the target space, and it may be determined that the likelihood of grout contamination or mold growth is relatively high in high-humidity environments, such as bathrooms, laundry rooms, or outdoor spaces.

[0169] Additionally, the humidity may be the average value of humidity measured by a humidity measuring device that communicates wirelessly or wiredly with the user's terminal (100), and the humidity measuring device may be a sensor device installed inside the space.

[0170] Additionally, the device (200) can verify the usage included in the spatial information.

[0171] At this time, the intended use may indicate whether the target space is used as a bathroom, kitchen, living room, commercial space, or outdoor space, and the device (200) may determine that the possibility of joint contamination may vary depending on the intended use.

[0172] For example, spaces such as bathrooms or kitchens may be judged to have a high likelihood of grout contamination or discoloration due to the high frequency of water usage, while spaces such as bedrooms or studies may be judged to have a low likelihood of grout contamination or discoloration due to the relatively low frequency of water usage.

[0173] The device (200) can calculate the probability of joint contamination occurring using the above material, humidity, and use.

[0174] For example, the device (200) can calculate a contamination generation coefficient according to the material, a contamination generation coefficient according to humidity, and a contamination generation coefficient according to the intended use, respectively, and said coefficients can be stored in advance in a database provided in the device (200) to correspond to each attribute value.

[0175] At this time, the contamination generation coefficient according to the material can be set based on the surface absorption rate, surface roughness, or whether the surface is coated of the material, the contamination generation coefficient according to humidity can be set based on the relative humidity range of the space, and the contamination generation coefficient according to the intended use can be set based on the frequency of water use or the frequency of exposure to contaminants in the space.

[0176] The device (200) can calculate the probability of joint contamination by performing a weighted summation or multiplication operation of the contamination occurrence coefficient according to the material, the contamination occurrence coefficient according to humidity, and the contamination occurrence coefficient according to the use.

[0177] Additionally, the device (200) can normalize the probability value to convert it into a range of 0 or more and 1 or less so that the probability of joint contamination occurring can be compared between different similar tiles.

[0178] The device (200) can calculate a joint contamination prediction score by performing a score conversion operation based on the joint contamination occurrence probability calculated as above. For example, the device (200) can apply an inverse conversion or linear normalization operation to the joint contamination occurrence probability so that a higher score is calculated as the joint contamination occurrence probability decreases.

[0179] The joint contamination prediction score calculated in this way can be used as one of the judgment factors to calculate the priority score of similar tiles in a subsequent stage.

[0180] In step S305, the device (200) can calculate a reflectance reference value based on illumination and usage for similar tiles, calculate a difference value between the reflectance of the similar tile and the reflectance reference value, and calculate a heat reflection influence score that is converted to a higher value as the difference value becomes smaller.

[0181] Specifically, the device (200) can check the reflectance of the similar tile by referring to the tile information of the similar tile stored in the tile database.

[0182] In this case, reflectance may be a value indicating the degree to which the tile surface reflects incident light; for example, glossy tiles may have a relatively high reflectance, while matte tiles may have a relatively low reflectance. Additionally, reflectance may vary depending on the color brightness, surface finish condition, or whether a surface coating is applied to the tile surface, and this information may be included in the tile information.

[0183] Additionally, the device (200) can check the illuminance included in the spatial information received in step S302.

[0184] In this case, illuminance may be a value representing the intensity of light incident on the target space, and, for example, may be an illuminance value measured through an illuminance sensor, or an illuminance value estimated based on the window area of ​​the space, the arrangement of lighting fixtures, or natural lighting conditions.

[0185] Generally, in spaces with high illumination, it is considered that there is a high likelihood of glare occurring if highly reflective tiles are used.

[0186] Additionally, the device (200) can verify the usage included in the spatial information.

[0187] At this time, the intended use may indicate whether the target space is used as a living room, kitchen, bathroom, office space, or commercial space, and the device (200) may determine that the appropriate reflectance level may differ depending on the intended use.

[0188] For example, it may be determined that a relatively low reflectance is preferred to minimize glare in spaces where people stay for long periods, such as living rooms or office spaces, while a relatively high reflectance is acceptable in spaces where a bright atmosphere is required, such as commercial or exhibition spaces.

[0189] The device (200) can calculate a reference value for reflectance using the illuminance and usage. For example, the device (200) can calculate a reference coefficient for reflectance according to the illuminance level and a reference coefficient for reflectance according to the usage, respectively, and calculate a reference value for reflectance by performing a weighted summation or an average operation on the coefficients.

[0190] To this end, a database provided in the device (200) may have reflectance reference coefficients corresponding to each illuminance level and usage purpose stored in advance.

[0191] The device (200) can calculate the difference between the reflectance reference value calculated as above and the reflectance of a similar tile.

[0192] For example, the device (200) can calculate a reflection difference value by subtracting a reflection reference value from the reflection of a similar tile or by calculating the absolute difference between the reflection reference value and the reflection of a similar tile.

[0193] Additionally, the device (200) can convert the difference value to a range of 0 or more and 1 or less so that the difference value can be normalized to be comparable between different similar tiles.

[0194] The device (200) can calculate a heat reflection influence score by performing a score conversion operation based on the reflectance difference value calculated as above.

[0195] For example, the device (200) can apply an inverse transformation or linear normalization operation to the reflectance difference value so that a higher score is produced as the reflectance difference value becomes smaller.

[0196] The heat reflection influence score calculated in this way can be used as one of the judgment factors to calculate the priority score of similar tiles in a subsequent step.

[0197] In step S306, the device (200) can quantify and calculate the frequency of use and the expected load according to the installation location for similar tiles according to their intended use, calculate the risk of damage by comparing the frequency of use and the expected load with the strength of the similar tiles, and calculate a risk of damage score that is converted to a higher value as the risk of damage decreases.

[0198] Specifically, the device (200) can check the usage included in the spatial information received in step S302.

[0199] In this case, the intended use may indicate whether the target space is used as a bathroom, kitchen, living room, bedroom, hallway, commercial space, or outdoor space.

[0200] The device (200) can estimate the frequency of movement or usage that may occur in the space based on the above-mentioned use, and can quantify this as a frequency of use.

[0201] For example, the device (200) can determine that a space with high usage frequency is a space where people frequently move, such as a hallway, commercial space, or entrance, and can determine that a space with a medium level of usage frequency is a space where a certain level of movement occurs, such as a living room or kitchen. On the other hand, a space with relatively low movement frequency, such as a bedroom or decorative space, can be determined that a space with low usage frequency is a space.

[0202] That is, the device (200) can determine a usage frequency coefficient corresponding to the intended use of the space as described above, and calculate the usage frequency as a numerical value by referring to the usage frequency coefficient. To this end, a usage frequency coefficient matching the intended use may be stored in advance in a database provided in the device (200).

[0203] Additionally, the device (200) can verify the construction location included in the spatial information received in step S302. At this time, the construction location may be information indicating the location of the structure where the tile is to be installed, and may be input as, for example, any one of a floor, wall surface, stairs, object surface, or other structural surface.

[0204] Additionally, the device (200) can identify a corresponding load factor according to the construction location and calculate the expected load to be received by the tile as a numerical value by referring to the load factor. To this end, a load factor matching the construction location may be stored in advance in a database provided in the device (200).

[0205] For example, if the construction location is identified as a floor, the device (200) may assign a high expected load value by determining that there is a high likelihood of repeated action by human body weight, furniture load, or movement load. Additionally, if the construction location is identified as a staircase, it may assign a high expected load value by determining that repeated impact load may occur. On the other hand, if the construction location is identified as a wall, it may assign a low expected load value by determining that the load is relatively small. Furthermore, the device (200) may assign a very low expected load value for locations where almost no load occurs, such as object surfaces or decorative structures.

[0206] The device (200) can quantify the space usage intensity that the tile will receive in the target space using the usage frequency and expected load calculated as above. At this time, the device (200) can calculate the space usage intensity value by performing a weighted summation or multiplication operation of the usage frequency and expected load.

[0207] Additionally, the device (200) can check the strength of the similar tile by referring to the tile information of the similar tile stored in the tile database.

[0208] In this case, the strength of the tile may be a value representing the resistance characteristics to breakage or cracking of the tile, and may include, for example, at least one of bending strength, compressive strength, abrasion strength, or impact strength. Additionally, the strength value may be a value measured according to test data provided by the manufacturer or standard test criteria.

[0209] The device (200) can calculate the risk of damage by comparing the space usage strength with the strength of a similar tile. At this time, the device (200) can calculate the risk of damage by subtracting the strength value of the similar tile from the space usage strength value or by calculating the ratio of the two values.

[0210] For example, the device (200) can determine that there is a high probability that the tile will be damaged when the space usage intensity is high and the strength of the similar tile is low, and can calculate a high risk of damage. On the other hand, it can determine that there is a low probability that the tile will be damaged when the space usage intensity is low and the strength of the similar tile is high, and can calculate a low risk of damage.

[0211] Additionally, the device (200) can convert the risk of damage to a range of 0 or more and 1 or less so that the risk of damage can be normalized to be comparable between similar tiles with different risk levels.

[0212] The device (200) can calculate a damage risk score by performing a score conversion operation based on the damage risk calculated as above. For example, the device (200) can apply an inverse conversion or linear normalization operation of the damage risk to ensure that a higher score is calculated as the damage risk is lower.

[0213] The damage risk score calculated in this way can be used as one of the judgment factors to calculate the priority score of similar tiles in a subsequent step.

[0214] In step S307, the device (200) can calculate a price ratio value for a similar tile by dividing the construction price of the similar tile by a reference price, calculate a period ratio value by dividing the expected construction period by a reference period, and calculate a price score and a delivery period score that are converted to higher values ​​as the price ratio value and the period ratio value are smaller.

[0215] Specifically, the device (200) can check the construction price of a similar tile by referring to the tile information of a similar tile stored in a tile database.

[0216] At this time, the construction price may be a value representing the cost required to perform construction work using similar tiles, and may include, for example, at least one of the tile material unit price, construction labor cost, auxiliary material cost, or other construction-related costs. Additionally, the construction price may be expressed as a price per unit area or a total construction price, and the device (200) may convert the price per unit area into a total construction price using the area included in the spatial information received in step S302.

[0217] Additionally, the device (200) can determine the estimated construction period of the similar tile. In this case, the estimated construction period may represent the period expected to be required to complete the construction work using the similar tile, and may include, for example, at least one of the tile preparation period, the delivery period, and the actual construction work period. Additionally, the estimated construction period may be expressed in days, hours, or other time units.

[0218] In addition, the device (200) can check the reference price and reference period.

[0219] At this time, the standard price may be a reference value for comparing tile installation costs, and may be, for example, either a budget value entered by a user or a standard price preset in the device (200). Additionally, the device (200) may generate a standard price by checking the installation prices of multiple similar tiles and calculating the average value of the installation prices of multiple similar tiles.

[0220] Additionally, the reference period may be a value representing the standard period allowed until the completion of the tile installation work, for example, a construction completion period desired by the user or an average construction period preset in the device (200). Additionally, the device (200) may generate a reference period by checking the estimated construction periods of multiple similar tiles and calculating the average value of the estimated construction periods of multiple similar tiles.

[0221] The device (200) can calculate a price ratio value by dividing the construction price of a similar tile identified as above by the standard price. At this time, the price ratio value may be a value indicating the level of the construction price of the similar tile compared to the standard price, and the smaller the price ratio value, the more likely it is to be determined that the tile has a lower construction price compared to the standard price.

[0222] Additionally, the device (200) can calculate a period ratio value by dividing the expected construction period of a similar tile by a reference period. At this time, the period ratio value may be a value indicating the level of the expected construction period of the similar tile compared to the reference period, and the smaller the period ratio value, the more likely it is to be determined that the tile has a shorter construction period compared to the reference period.

[0223] Additionally, the device (200) can normalize each ratio value to convert it into a range of 0 or more and 1 or less so that the price ratio value and the period ratio value can be compared between similar tiles that are different.

[0224] The device (200) can calculate a price score by performing a score transformation operation based on the price ratio value calculated as above. For example, the device (200) can apply an inverse transformation or linear normalization operation to the price ratio value so that a higher price score is calculated as the price ratio value is smaller.

[0225] Additionally, the device (200) can calculate a delivery period score by performing a score transformation operation based on the period ratio value. For example, the device (200) can apply an inverse transformation or linear normalization operation to the period ratio value so that a higher delivery period score is calculated as the period ratio value becomes smaller.

[0226] The price score and delivery time score calculated in this way can be used as one of the judgment factors to calculate the priority score of similar tiles in a subsequent stage.

[0227] In step S308, the device (200) can calculate a priority score of similar tiles based on a similarity score, a score for minimizing array distortion, a score for predicting joint contamination, a score for the impact of heat reflection, a score for the risk of damage, a price score, and a score for the delivery time.

[0228] Specifically, the device (200) can check the similarity score calculated in step S301, the array distortion minimization score calculated in step S303, the joint contamination prediction score calculated in step S304, the heat reflection effect score calculated in step S305, the damage risk score calculated in step S306, and the price score and delivery time score calculated in step S307.

[0229] The device (200) can calculate the priority score of similar tiles based on each of the above scores.

[0230] At this time, the device (200) may calculate the priority score of similar tiles by simply summing each score, or it may calculate the priority score of similar tiles through weighted summation by applying weights corresponding to each score to reflect the importance of each evaluation element.

[0231] For example, the device (200) may apply a first weight to the similarity score, a second weight to the array distortion minimization score, a third weight to the joint contamination prediction score, a fourth weight to the heat reflection impact score, a fifth weight to the damage risk score, a sixth weight to the price score, and a seventh weight to the delivery time score, respectively. In this case, the weights may be pre-set values, or they may be dynamically adjusted based on user-set values ​​or training data.

[0232] Additionally, the device (200) can calculate the priority score of similar tiles through (1st weight × similarity score) + (2nd weight × array distortion minimization score) + (3rd weight × joint contamination prediction score) + (4th weight × heat reflection influence score) + (5th weight × damage risk score) + (6th weight × price score) + (7th weight × delivery time score).

[0233] The device (200) can sort a plurality of similar tiles based on priority scores calculated as above, and can determine the recommended order of similar tiles according to the order of priority scores.

[0234] The priority score calculated in this way can be used as a criterion value to generate a list of similar tiles or provide recommended tiles to the user in subsequent steps.

[0235] As a result, the device (200) calculates scores for multiple evaluation factors based on the request information for the required tile, the tile information of the similar tile, and the spatial information of the space where the tile is to be installed, and determines the priority of the similar tile by combining the scores, thereby enabling tile recommendations that comprehensively consider the environmental conditions, construction conditions, and economic feasibility of the target space.

[0236] FIG. 4 is a flowchart illustrating the process of calculating a comprehensive score to select a tile manufacturer according to one embodiment.

[0237] Referring to FIG. 4, first, in step S401, the device (200) can identify the tile manufacturer using manufacturer identification information stored in the tile database corresponding to each tile.

[0238] Specifically, the device (200) can query tile information stored in the tile database.

[0239] At this time, the tile database may store tile information for multiple tiles currently being sold or distributed, and each tile information may include attribute values ​​representing the tile's color, pattern, material, specifications, surface finish, reflectivity, strength, or other tile characteristics. Additionally, each tile information may also store manufacturer identification information to identify the tile manufacturer performing the order fulfillment for the corresponding tile. Here, the order fulfillment may include at least one of the manufacturing, production, supply, or delivery of the tile, but is not limited thereto.

[0240] In this case, the manufacturer identification information may be information for uniquely identifying a tile manufacturer, and may include, for example, a manufacturer identification code, a company name, a company code, or other identifiable identifiers, but is not limited thereto. Additionally, company information such as the manufacturer's name, contact information, or location information may be stored together in correspondence with the above manufacturer identification information.

[0241] Accordingly, the device (200) can verify the manufacturer identification information stored in the tile database corresponding to each tile information stored in the tile database in order to send a tile production request message to the tile manufacturer's terminal because no similar tile exists in step S207, and can identify the tile manufacturer performing the order fulfillment of the tile using the manufacturer identification information.

[0242] Additionally, the device (200) can generate a list of tile manufacturers by collecting manufacturer identification information corresponding to multiple tile information.

[0243] At this time, if the same manufacturer identification information is stored corresponding to multiple tiles, the device (200) can generate a list of manufacturers by removing duplicate manufacturer identification information.

[0244] The device (200) can identify a tile manufacturer using the list of manufacturers generated as above, and the identified tile manufacturer can be utilized as a candidate manufacturer to perform a tile manufacturing request in a subsequent step.

[0245] The tile manufacturers identified in this way can be used as basic data to select candidate tile manufacturers or send tile production request messages in subsequent stages.

[0246] In step S402, the device (200) can select a tile manufacturer that is matched with more than the target number of tiles as a candidate tile manufacturer.

[0247] Specifically, the device (200) can determine the number of tiles matched with each tile manufacturer for the multiple tile manufacturers identified in step S401.

[0248] At this time, the tile matched with the tile manufacturer can be verified based on the manufacturer identification information stored corresponding to each tile information stored in the tile database.

[0249] For example, the device (200) can calculate the number of tiles matched with a tile manufacturer by aggregating the number of tiles that have manufacturer identification information identical to that of a specific manufacturer identification information among the multiple tile information stored in the tile database.

[0250] Additionally, the device (200) can check a preset target number. In this case, the target number may be a reference value for judging the production experience or product diversity of a tile manufacturer based on the number of tiles the tile manufacturer has fulfilled orders for. That is, the target number may be a value preset in the device (200), or a value set according to a user setting value or a system operation policy. The target number may vary depending on the embodiment.

[0251] The device (200) can compare the number of tiles matched with each tile manufacturer identified as above with the target number.

[0252] At this time, if the device (200) confirms that the number of tiles matched with a specific tile manufacturer is greater than or equal to the target number, it can select the tile manufacturer as a candidate tile manufacturer.

[0253] On the other hand, if the number of tiles matched with a specific tile manufacturer is found to be less than the target number, the device (200) may exclude the tile manufacturer from the candidate tile manufacturers.

[0254] In addition, the device (200) can perform the above comparison process for each of the multiple tile manufacturers to select multiple candidate tile manufacturers.

[0255] The device (200) can store the candidate tile manufacturers selected as above as a list of candidate manufacturers, and the list of candidate manufacturers can be used as target companies for analyzing the production history or delivery history of each candidate tile manufacturer in a subsequent step.

[0256] In step S403, the device (200) can collect tile information of tiles produced by a candidate tile manufacturer stored in a tile database for each candidate tile manufacturer.

[0257] Specifically, the device (200) can identify candidate tile manufacturers that are matched with more than the target number of tiles according to step S402, and the device (200) can search the tile database using manufacturer identification information corresponding to the candidate tile manufacturers.

[0258] At this time, the device (200) can check the tile information of the tile for which the order fulfillment was performed by the corresponding candidate tile manufacturer by querying the tile information in which the same manufacturer identification information as the above manufacturer identification information is stored.

[0259] That is, the device (200) can verify the tile information of the tile for which an order has been fulfilled by the candidate tile manufacturer by extracting from the tile database tile information in which manufacturer identification information matching the manufacturer identification information of the specific candidate tile manufacturer is stored.

[0260] In this case, the tile information may include, but is not limited to, attribute values ​​representing the color, pattern, material, specifications, surface finish, reflectivity, strength, or other tile characteristics of the tile. Additionally, the tile information may also store order fulfillment history for each fulfillment detail of the tile, including the scheduled time of fulfillment, the actual time of fulfillment, and compliance status. Here, fulfillment may refer to the entire order execution process, including manufacturing, production, delivery, and supply, but is not limited to.

[0261] The device (200) can collect the tile information retrieved as described above and generate a set of tile information for each candidate tile manufacturer.

[0262] For example, the device (200) can organize and store multiple tile information corresponding to a specific candidate tile manufacturer into a single data set, and can generate sets of tile information corresponding to other candidate tile manufacturers in the same way.

[0263] In addition, the device (200) can store or load the set of tile information for each candidate tile manufacturer generated as above into memory and use it as analysis data in a subsequent step.

[0264] The tile information collected in this way can be used as basic data in subsequent stages to determine whether candidate tile manufacturers can implement specific attributes or to evaluate the feasibility of implementing attribute combinations.

[0265] In step S404, the device (200) can break down the feature vector of the required tile into a plurality of attribute elements including material, color, pattern, size and surface finish.

[0266] Specifically, the device (200) can check the feature vector of the required tile generated in step S202.

[0267] At this time, the feature vector may be vector data generated by analyzing request information including at least one of image data and text data of a required tile received from a user terminal (100), and may include a plurality of attribute values ​​representing the color, pattern, material, specifications, surface finish, or other tile characteristics of the tile in a vector form.

[0268] The device (200) can decompose into a plurality of attribute elements based on each attribute value included in the feature vector and separate them into each attribute element.

[0269] In this case, attribute elements may refer to individual elements that constitute the characteristics of the tile, and can be classified into attribute items such as material, color, pattern, size, and surface finish.

[0270] For example, the device (200) can classify a value representing the material of a tile among the attribute values ​​included in the feature vector as a material attribute element, and a value representing the color of a tile as a color attribute element.

[0271] Additionally, the device (200) can classify values ​​representing the surface pattern or pattern type of the tile as pattern attribute elements, and can classify dimension information including the width and height of the tile as standard attribute elements.

[0272] Additionally, the device (200) can classify a value representing the surface treatment state of the tile as a surface finish attribute element.

[0273] Additionally, the device (200) can store the attribute elements by separating them into independent attribute items in a data structure. For example, the device (200) can create a set of attribute elements by configuring material, color, pattern, specifications, and surface finish as separate attribute fields.

[0274] The device (200) can generate a set of attribute elements for a required tile using a plurality of attribute elements decomposed as described above, and the set of attribute elements can be used as reference data to determine whether a candidate tile manufacturer can implement the corresponding attribute in a subsequent step or to evaluate the possibility of implementing the attribute combination.

[0275] In step S405, the device (200) can check for each candidate tile manufacturer whether an attribute corresponding to each of the multiple attribute elements exists in the collected tile information, and calculate an attribute implementation possibility score based on the ratio of implementable attribute elements.

[0276] Specifically, the device (200) can check the set of attribute elements of the required tile generated in step S404.

[0277] At this time, the set of attribute elements may include multiple attribute elements such as material, color, pattern, specifications, and surface finish.

[0278] Additionally, the device (200) can verify the set of tile information by candidate tile manufacturer collected in step S403.

[0279] At this time, the above tile information set may include tile information for multiple tiles for which orders have been fulfilled by the corresponding candidate tile manufacturer, and each tile information may include attribute values ​​representing the color, pattern, material, specifications, surface finish, or other tile characteristics of the tile.

[0280] The device (200) can check whether, for each attribute element included in the set of attribute elements of the required tile, there exists an attribute corresponding to the set of tile information of the candidate tile manufacturer.

[0281] At this time, the device (200) can determine whether an attribute can be implemented by comparing the attribute value corresponding to each attribute element with the attribute value included in the tile information.

[0282] For example, if the material attribute element of the required tile is identified as “porcelain,” the device (200) can check whether a tile made of porcelain material exists in the tile information set of the candidate tile manufacturer. Additionally, if the color attribute element of the required tile is identified as “beige,” the device (200) can check whether a tile in the beige color family exists in the tile information set of the candidate tile manufacturer. Additionally, the device (200) can check whether a specific pattern type exists in the case of the pattern attribute element, and in the case of the specification attribute element, check whether there is a history of producing tiles of the same or similar specifications. Additionally, the device (200) can check whether a tile with a matte, glossy, or other surface finish method exists in the case of the surface finish attribute element.

[0283] The device (200) can determine whether each attribute element is implementable based on the comparison results as described above, for example, if the attribute exists in the tile information set of a candidate tile manufacturer, it can be determined as an implementable attribute element, and if it does not exist, it can be determined as an unimplementable attribute element.

[0284] The device (200) can determine the number of implementable attribute elements determined as above and can calculate the ratio of implementable attribute elements to the total number of attribute elements. For example, if there are 5 total attribute elements and 4 of them are confirmed to be implementable, the device (200) can calculate the implementable ratio as 4 / 5.

[0285] The device (200) can calculate an attribute implementation possibility score based on the above implementation possibility ratio (ratio of implementable attribute elements), and can calculate a higher attribute implementation possibility score as the implementation possibility ratio is higher, and can calculate a lower attribute implementation possibility score as the implementation possibility ratio is lower.

[0286] That is, the device (200) can calculate an attribute implementation possibility score in proportion to the implementation possibility ratio.

[0287] The attribute implementation feasibility score calculated in this way can be used as one of the judgment factors to evaluate the manufacturing capability or suitability of attribute implementation of candidate tile manufacturers in subsequent stages.

[0288] In step S406, the device (200) can check for each candidate tile manufacturer whether there is a past tile in which two or more of the multiple attribute elements are simultaneously implemented, and calculate an attribute combination implementation score based on the ratio of the simultaneously implemented attribute combinations.

[0289] Specifically, the device (200) can check the set of attribute elements of the required tile generated in step S404.

[0290] At this time, the set of attribute elements may include multiple attribute elements such as material, color, pattern, specifications, and surface finish.

[0291] The device (200) can generate multiple attribute combinations by combining two or more attributes among the multiple attribute elements included in the set of attribute elements.

[0292] In this case, an attribute combination may refer to a set of attributes in which two or more attribute elements are simultaneously satisfied, for example, a combination of material and color, a combination of color and pattern, a combination of pattern and specification, or a combination including material, color, and surface finish simultaneously may be generated.

[0293] Additionally, the device (200) can verify the set of tile information by candidate tile manufacturer collected in step S403.

[0294] At this time, the above tile information set may include tile information for multiple past tiles for which orders were fulfilled by the corresponding candidate tile manufacturer, and each tile information may include attribute values ​​representing the color, pattern, material, specifications, surface finish, or other tile characteristics of the tile.

[0295] For each of the generated attribute combinations, the device (200) can check whether a past tile in which the corresponding attribute combination was simultaneously implemented exists in the tile information set of the candidate tile manufacturer.

[0296] For example, the device (200) can determine whether there exists a tile that is porcelain material and simultaneously has a beige color in the tile information set of candidate tile manufacturers when a specific combination of attributes consists of a combination of “porcelain material” and “beige color”.

[0297] Additionally, the device (200) may check whether there is a tile in which “beige color” and “marble pattern” are implemented simultaneously, or whether there is a tile in which “porcelain material”, “beige color” and “matte surface finish” are implemented simultaneously.

[0298] The device (200) can determine whether to simultaneously implement each attribute combination based on the above-mentioned verification results, and if the attribute combination exists in the past tile information of the candidate tile manufacturer, it can determine that it is an attribute combination that can be implemented, and if it does not exist, it can determine that it is an attribute combination that cannot be implemented.

[0299] The device (200) can determine the number of implementable attribute combinations determined as above and can calculate the ratio of implementable attribute combinations to the total number of attribute combinations. For example, if the total number of attribute combinations is 10 and it is confirmed that 6 of these attribute combinations were simultaneously implemented in past tiles, the device (200) can calculate the attribute combination implementation ratio as 6 / 10.

[0300] The device (200) can calculate an attribute combination implementation score based on the attribute combination implementation ratio (ratio of simultaneously implemented attribute combinations), and can calculate a higher attribute combination implementation score as the attribute combination implementation ratio is higher, and can calculate a lower attribute combination implementation score as the attribute combination implementation ratio is lower.

[0301] That is, the device (200) can calculate an attribute combination implementation score in proportion to the attribute combination implementation ratio.

[0302] The attribute combination implementation score calculated in this way can be used as one of the judgment factors to evaluate the production experience or attribute combination implementation ability of candidate tile manufacturers in subsequent stages.

[0303] In step S407, the device (200) can calculate the number of executions, the execution compliance rate, the frequency of recent executions, the variability of the execution period, and the execution gap period by analyzing the order execution history of tiles produced by the candidate tile manufacturer stored in the tile database for each candidate tile manufacturer, and calculate an execution stability score based thereon.

[0304] Specifically, the device (200) can verify the set of tile information by candidate tile manufacturer collected in step S403.

[0305] In this case, the tile information may include, but is not limited to, attribute values ​​representing the color, pattern, material, specifications, surface finish, reflectivity, strength, or other tile characteristics of the tile. Additionally, the tile information may also store order fulfillment history for each fulfillment detail of the tile, including the scheduled time of fulfillment, the actual time of fulfillment, and compliance status. Here, fulfillment may refer to the entire order execution process, including manufacturing, production, delivery, and supply, but is not limited to.

[0306] The device (200) can calculate the number of executions of a candidate tile manufacturer using the order execution history. In this case, the number of executions may refer to the total number of execution records recorded for the candidate tile manufacturer, and the more executions there are, the more likely it is to be judged that the company possesses diverse experience in tile manufacturing and supply.

[0307] Additionally, the device (200) can calculate the compliance rate of a candidate tile manufacturer using the order fulfillment history. In this case, the compliance rate can be calculated by dividing the number of fulfillment records in which compliance is recorded as 'compliant' by the total number of fulfillment records, and the higher the compliance rate, the more likely the manufacturer is to be judged to be a company that reliably fulfills orders.

[0308] Additionally, the device (200) can calculate the recent fulfillment frequency using the order fulfillment history. In this case, the recent fulfillment frequency may refer to the number of fulfillment details performed during a predetermined reference period based on the current point in time, and may be calculated based on, for example, the number of order fulfillments performed during the last 1 month, the last 3 months, or the last 6 months. The device (200) may determine that the company is not active in production activities if the recent fulfillment frequency is low, and conversely, may determine that the production facilities are likely to be in an overloaded state if the recent fulfillment frequency is high. Accordingly, the device (200) may evaluate that the recent fulfillment frequency has a higher value as it approaches a predetermined appropriate range.

[0309] Additionally, the device (200) can calculate the variability of the execution period based on the difference between the scheduled execution time and the actual execution time using the order execution history. At this time, the device (200) can calculate the difference value for a plurality of execution details and calculate the variability of the execution period using the average or deviation of the difference value. The smaller the difference between the scheduled execution time and the actual execution time, the more likely it is to be judged that the company has a certain execution capability, and the larger the difference, the more likely it is to be judged that the company has low stability of the execution schedule.

[0310] Additionally, the device (200) can calculate a performance gap period by using the order performance history and the time between the actual performance time of the last performance record and the current time. In this case, the performance gap period may refer to the period during which the candidate tile manufacturer has not recently performed performance for tiles, and the longer the performance gap period, the more likely it is to be determined that the manufacturer is not active in performance activities.

[0311] The device (200) can calculate an implementation stability score using the number of implementations, implementation compliance rate, recent implementation frequency, implementation period variability, and implementation gap period calculated as above. For example, the device (200) can calculate an implementation stability score by applying weights corresponding to each of the above elements to perform weighted summation, or by normalizing each of the above element values ​​and then performing an average operation.

[0312] The performance stability score calculated in this way can be used as one of the judgment factors to evaluate the order fulfillment capability and supply reliability of candidate tile manufacturers in subsequent stages.

[0313] In step S408, the device (200) can calculate a comprehensive score of candidate tile makers based on an attribute implementation possibility score, an attribute combination implementation score, and an implementation stability score.

[0314] Specifically, the device (200) can check the attribute implementation possibility score calculated in step S405, the attribute combination implementation score calculated in step S406, and the implementation stability score calculated in step S407.

[0315] In this case, the attribute implementation feasibility score may be a value representing the candidate tile manufacturer's ability to implement individual attributes of the required tile, and the attribute combination implementation score may be a value representing experience in manufacturing tiles where multiple attributes are implemented simultaneously. Additionally, the implementation stability score may be a value representing implementation reliability and stability calculated based on the candidate tile manufacturer's order fulfillment history.

[0316] The device (200) can calculate a comprehensive score of a candidate tile manufacturer using the attribute implementation possibility score, attribute combination implementation score, and implementation stability score.

[0317] At this time, the device (200) may calculate a total score by simply summing each of the above scores, or may calculate a total score by performing a weighted summation by applying weights corresponding to each score to reflect the importance of each evaluation element.

[0318] For example, the device (200) may apply a weight a to the attribute implementation possibility score, a weight b to the attribute combination implementation score, and a weight c to the implementation stability score, respectively, and said weights may be pre-set values, or may be dynamically adjusted based on user-set values ​​or training data.

[0319] Additionally, the device (200) can calculate the total score of a candidate tile manufacturer using (a weight × attribute implementation possibility score) + (b weight × attribute combination implementation score) + (c weight × implementation stability score).

[0320] The device (200) can sort multiple candidate tile manufacturers based on the total score calculated as above, and can determine the priority of candidate tile manufacturers to perform tile manufacturing requests in order of highest total score.

[0321] The comprehensive score calculated in this way can be used as a criterion value to select the target tile manufacturer to whom the tile production request message will be sent in the subsequent stage.

[0322] As a result, the device (200) can reasonably select a manufacturer suitable for a tile production request by comprehensively evaluating the feasibility of implementing the attributes of the required tile, the experience of implementing attribute combinations, and the stability of order fulfillment.

[0323] FIG. 5 is a flowchart for explaining the process of sending a tile production request message to a tile manufacturer, taking into account the comprehensive score according to one embodiment.

[0324] Referring to FIG. 5, first, in step S501, the device (200) can select a candidate tile manufacturer with a total score higher than the target score as a consideration tile manufacturer.

[0325] Specifically, the device (200) can check the total score of each candidate tile manufacturer calculated through the process of FIG. 4.

[0326] At this time, the above-mentioned total score may be a value calculated based on the attribute implementation feasibility score, the attribute combination implementation score, and the implementation stability score, and may be a value that comprehensively represents the possibility of a candidate tile manufacturer meeting the production requirements of the necessary tiles and the order implementation stability.

[0327] Additionally, the device (200) can check a preset target score. At this time, the target score may be a reference value for determining whether a candidate tile manufacturer is suitable for performing a tile manufacturing request, and may be, for example, a reference value preset in the device (200). Additionally, the target score may be a value dynamically adjusted based on past tile manufacturing request history or system operation data, and may vary depending on the embodiment.

[0328] The device (200) can compare the total score and target score of the candidate tile manufacturer identified as above.

[0329] For example, the device (200) can determine whether the overall score of a candidate tile manufacturer is higher than the target score, and if the overall score is higher than the target score, it can determine that the candidate tile manufacturer is a suitable company to perform the production of the necessary tiles.

[0330] The device (200) can select a candidate tile manufacturer with a total score higher than the target score as a consideration tile manufacturer based on the above judgment result.

[0331] In this case, the Goryeo tile manufacturer may refer to a candidate target company to which a tile production request message is sent.

[0332] In step S502, the device (200) can check whether there are multiple tile manufacturers.

[0333] Specifically, the device (200) can verify the number of Goryeo tile manufacturers selected in step S501.

[0334] At this time, a Goryeo tile manufacturer may refer to a company among the candidate tile manufacturers whose overall score is higher than the target score, and the device (200) can calculate the total number of Goryeo tile manufacturers by counting the number of the above Goryeo tile manufacturers.

[0335] In addition, the device (200) can determine whether there are multiple Goryeo tile manufacturers based on the number of Goryeo tile manufacturers calculated as above.

[0336] If multiple Goryeo tile manufacturers are identified in step S502, in step S503, the device (200) may sort the Goryeo tile manufacturers according to a total score to generate a list of tile manufacturers, and provide a user interface to the user's terminal (100) that includes a designation means for generating a list of tile manufacturers and a designation for the Goryeo tile manufacturers.

[0337] Specifically, if multiple tile manufacturers are identified in step S502, the device (200) can sort the tile manufacturers based on the total score calculated through the process of FIG. 4 so that the user can directly select the tile manufacturer to send the tile manufacturing request message to.

[0338] At this time, the device (200) can sort the tile manufacturers in order of highest overall score, and the sorted order may represent a priority indicating the degree of suitability for performing a tile manufacturing request.

[0339] The device (200) can generate a list of tile manufacturers based on the sorted tile manufacturers.

[0340] At this time, the above list of tile manufacturers may include manufacturer identification information including the name of each Goryeo tile manufacturer, company information including location and contact information, a comprehensive score corresponding to each Goryeo tile manufacturer, and list data including tile information of tiles associated with the manufacturer, but is not limited thereto.

[0341] Additionally, the device (200) can convert the tile manufacturer list into screen display data to configure a user interface screen, and the user interface can be transmitted to the user's terminal (100) and displayed through a display provided on the user's terminal (100).

[0342] Additionally, the device (200) may include a designation means for designating one of the aforementioned tile manufacturers in the user interface. In this case, the designation means may be a button, a checkbox, a touch area, or other selectable graphical user interface element, and may be selected via touch input, mouse click input, or other input device of the user's terminal (100).

[0343] When a user selects the above-mentioned designation means by the user's terminal (100), the user's terminal (100) can generate a designation signal that designates a specific consideration tile manufacturer based on the selection event and can transmit the designation signal to the device (200).

[0344] The device (200) can identify a tile manufacturer selected by the user based on the designated signal, and the identified tile manufacturer can be used as a target company to send a tile manufacturing request message in a subsequent step.

[0345] In this way, the device (200) can provide a user interface configured to allow the user to compare and select a candidate company when there are multiple tile manufacturers.

[0346] In step S504, the device (200) receives a designation signal for a tile manufacturer to be requested for production from the user's terminal (100) and can transmit a tile production request message to the terminal of the designated tile manufacturer to be requested for production.

[0347] Specifically, the device (200) can receive a designation signal corresponding to a tile manufacturer selected by the user from the user's terminal (100) through the user interface provided in step S503.

[0348] In this case, the designation signal may be a signal generated as a user selects a designation means included in the user interface, and the designation signal may include manufacturer identification information for identifying the Goryeo tile manufacturer selected by the user. Additionally, the designation signal may further include at least one of user identification information, selection time information, or user terminal identification information.

[0349] The device (200) can identify the tile manufacturer subject to the production request using the manufacturer identification information included in the designated signal. At this time, the device (200) can check the contact information of the tile manufacturer corresponding to the manufacturer identification information by querying the tile database.

[0350] Here, the above contact information may be information for communicating with the terminal of the tile manufacturer and may include, but is not limited to, commonly used contact information and network addresses (IP addresses), server addresses (URLs), communication account information, message reception identifiers, or other communication identification information.

[0351] Additionally, the device (200) can generate a tile production request message to be transmitted to the tile manufacturer that is the target of the production request.

[0352] At this time, the tile production request message may include data containing request information for the required tile, for example, attribute information indicating the color, pattern, material, specifications, surface finish, or other tile characteristics of the required tile. Additionally, the message may include a feature vector of the required tile generated in step S202, and the feature vector may be data representing the attribute information of the required tile in vector form.

[0353] Additionally, the above message may further include data containing information about the space where the tile is to be installed. For example, the spatial information may include at least one of the area, shape, humidity, illuminance, intended use, and installation location of the space where the tile is to be installed, but is not limited thereto.

[0354] In addition, the above message may further include the requested quantity, desired production schedule, desired delivery date, or other information related to the production request.

[0355] The device (200) can transmit the generated tile production request message to the terminal of the tile manufacturer that is the target of the production request. At this time, the message can be transmitted via a wired or wireless communication network, for example, using an internet network, a mobile communication network, or other data communication network. To this end, the device (200) can communicate with the tile manufacturer's terminal via wired or wireless means.

[0356] The device (200) can receive a response message to the tile production request message transmitted as above from a terminal of the tile manufacturer to which the production request was made. At this time, the response message may include response information indicating whether the tile manufacturer accepts the tile production request.

[0357] The device (200) can determine the type of response by analyzing the response message. For example, the device (200) can classify the response message into a positive response or a negative response using a preset response classification criterion or a natural language processing algorithm.

[0358] When the response message is confirmed as a positive response, the device (200) can match the user with the tile manufacturer that is the target of the production request. At this time, the device (200) can transmit the matching result to the user's terminal (100) and can support the user in producing and purchasing tiles through the manufacturer.

[0359] Meanwhile, if the device (200) confirms that the response message is a negative response, it may send a message to the user's terminal (100) instructing it to select another tile manufacturer. For example, the device (200) may provide the list of tile manufacturers generated in step S503 again, or automatically recommend another tile manufacturer.

[0360] In addition, the device (200) can repeat the above process to match the user with the tile manufacturer selected by the user.

[0361] In this way, the device (200) can proceed with the procedure for producing the necessary tiles by sending a tile production request message to a tile manufacturer selected by the user and processing the response from the manufacturer.

[0362] If it is confirmed in step S502 that there is only one tile manufacturer to consider or that no one exists, in step S505, the device (200) may first send a tile manufacturing request message to the terminal of the candidate tile manufacturer with the highest overall score.

[0363] Specifically, the device (200) checks the number of tile manufacturers calculated in step S502, and if the number of tile manufacturers is confirmed to be one, it may be determined that there is no need for the user to compare and select, and if there are no tile manufacturers, the company with the highest overall score among the candidate tile manufacturers may be selected as the candidate for production request.

[0364] If the device (200) confirms that there is only one or no tile manufacturer to consider as described above, it can identify the company with the highest overall score among the candidate tile manufacturers based on the overall score calculated in step S408.

[0365] At this time, the device (200) can compare the total score corresponding to each candidate tile manufacturer and select the manufacturer with the maximum value.

[0366] Additionally, the device (200) can verify the contact information of a tile manufacturer by querying a tile database using manufacturer identification information corresponding to the candidate tile manufacturer with the highest overall score. At this time, the contact information may include, but is not limited to, contact information, network address, server address, account information, or other communication identification information for communicating with the manufacturer.

[0367] The device (200) can generate a tile production request message to be sent to the candidate tile manufacturer with the highest overall score.

[0368] At this time, the tile production request message may include data containing request information for the required tile, for example, attribute information indicating the color, pattern, material, specifications, surface finish, or other tile characteristics of the required tile.

[0369] Additionally, the above message may include a feature vector of the required tile generated in step S202, and the feature vector may be data representing the attributes of the required tile in vector form.

[0370] Additionally, the above message may further include data containing information about the space where the tile is to be installed. For example, the space information may include at least one of the area, shape, humidity, illuminance, intended use, and installation location of the space where the tile is to be installed, but is not limited thereto.

[0371] In addition, the above message may include additional information related to the requested quantity, desired production schedule, or other production requests.

[0372] The device (200) can transmit the generated tile production request message to the terminal of the candidate tile manufacturer with the highest overall score. At this time, the message can be transmitted via a wired or wireless communication network, for example, using an internet network, a mobile communication network, or other data communication network. To this end, the device (200) can communicate with the tile manufacturer's terminal via wired or wireless means.

[0373] Additionally, the device (200) can receive a response message from a terminal of a tile manufacturer that is the subject of a production request after the transmission of the above message, and can determine whether to accept the production request by analyzing the response message.

[0374] The device (200) can match the user with the manufacturer when the response message is confirmed as a positive response, thereby supporting the user to proceed with tile production through the manufacturer.

[0375] Meanwhile, if the above response message is confirmed to be a negative response, the device (200) can identify the next-ranked candidate tile manufacturer based on the total score and resend a tile manufacturing request message to the terminal of the manufacturer.

[0376] The device (200) can repeat the above procedure to match the user with a tile manufacturer suitable for the user.

[0377] In this way, the device (200) can proceed with the procedure for producing necessary tiles by sending a production request message to the most suitable candidate tile manufacturer based on the comprehensive score, even if there are not enough tile manufacturers to consider.

[0378] As a result, the device (200) can automatically select a tile manufacturer based on a comprehensive score or allow the user to select one, and can efficiently perform the production of necessary tiles by sending a tile production request message to the terminal of the selected tile manufacturer.

[0379] FIG. 6 is a flowchart illustrating the process of automatically generating a purchase order according to one embodiment.

[0380] Referring to FIG. 6, first, in step S601, when the device (200) receives a purchase request signal for a similar tile from the user's terminal (100), it can check the order form of the similar tile.

[0381] Specifically, the device (200) can receive a purchase request signal corresponding to a similar tile selected by the user from the user's terminal (100) through a user interface provided to the user's terminal (100) in step S205 or S206.

[0382] At this time, the purchase request signal may be a signal generated as a user requests the purchase or production of a specific similar tile through an input means or selection means included in the user interface, and the purchase request signal may include tile identification information for identifying the selected similar tile.

[0383] In addition, the purchase request signal may further include request information indicating the quantity requested by the user, the time of the request, the construction location, or other purchase conditions.

[0384] The device (200) can identify tile information corresponding to the similar tile by searching the tile database using the tile identification information.

[0385] At this time, the tile database may store tile information for each of the plurality of tiles by matching it with tile identification information, and the tile information may include attribute values ​​representing the tile's color, pattern, material, size, surface finish, reflectance, intensity, or other tile characteristics.

[0386] Additionally, the tile database may store a purchase order form matched with each tile information, and the device (200) can check the purchase order form corresponding to the tile by querying the tile information corresponding to the tile identification information.

[0387] At this time, the above-mentioned order form may be a data structure in which order items necessary to perform tile purchase or production request are defined, and may be structured data including, for example, quantity, specifications, delivery time, construction location, unit price, delivery method, payment conditions, supplier identification information, or other items necessary for performing the order.

[0388] Additionally, the above order form may be composed of JSON, XML, CSV, tabular data, or other structured data formats, and the device (200) may receive or load the order form and perform parsing to separate and store each order item as a field of an internal data structure.

[0389] The device (200) can store the order form confirmed as above in memory or a database, and the order form can be used as reference data for extracting order items, generating an order draft, or automatically entering values ​​corresponding to the order in a subsequent step.

[0390] In step S602, the device (200) can analyze the order form to extract order items including quantity, specifications, delivery time, and construction location.

[0391] Specifically, the device (200) can load the order form confirmed in step S601 and analyze the order item definition information included in the order form.

[0392] At this time, the above purchase order form may be structured data in which items required for order execution are predefined, and an item identifier, item name, or field name corresponding to each order item may be stored together.

[0393] The above purchase order form may include, for example, quantity, specification, delivery date, construction location, unit price, delivery method, payment terms, or other items necessary for the execution of the order, but is not limited thereto.

[0394] Additionally, each order item can be defined in the form of a data field, and each field may include attributes indicating the item name, data type, input method, or whether it is required.

[0395] The device (200) can parse the order form to identify fields corresponding to each order item and classify the types of order items based on the item names of the fields. At this time, the process of parsing the order form to identify fields and classify types can be performed in the same way as a commonly used natural language processing algorithm.

[0396] For example, the device (200) can recognize a field defined as a “quantity”, “quantity”, “qty” or a corresponding code value as a quantity item.

[0397] In addition, a field defined by a field name or identifier as “specification,” “size,” “dimension,” or a corresponding code value can be recognized as a specification item.

[0398] Additionally, the device (200) can recognize a field defined with a field name or identifier as “delivery date,” “delivery date,” or a corresponding code value as a delivery date item, and can recognize a field defined with a field name or identifier as “installation location,” “installation location,” or a corresponding code value as an installation location item.

[0399] The device (200) can store the order items identified as above by separating them into attribute fields of an internal data structure, and can generate order item structure data including, for example, a quantity field, a specification field, a delivery time field, and a construction location field.

[0400] Additionally, the device (200) may store the order item structure data in memory or a database, and the order item structure data may be used as reference data for automatically entering values ​​extracted from request information into each order item in a subsequent step.

[0401] In step S603, the device (200) can analyze the text data and image data included in the request information to extract item values ​​corresponding to the order items.

[0402] Specifically, the device (200) can check the request information received in step S201, and the request information may include text data, image data, or a combination thereof input from the user's terminal (100).

[0403] The above request information may include information indicating the quantity, specifications, installation location, delivery time, or other order conditions of the tiles desired by the user.

[0404] The device (200) can analyze the text data included in the request information and extract a value corresponding to the order item.

[0405] To this end, the device (200) can analyze words or sentences included in text data using commonly used morphological analysis, keyword extraction, or natural language processing algorithms.

[0406] For example, the device (200) can extract a value corresponding to a quantity by analyzing a numerical representation included in text data, and can extract a value corresponding to a specification by analyzing a string containing a unit representing a length, width, or specification.

[0407] Additionally, the device (200) can extract a value corresponding to the delivery time by analyzing a string containing a date or time expression, and can extract a value corresponding to the construction location by analyzing a word or phrase indicating a location.

[0408] Additionally, the device (200) can analyze the image data included in the request information and extract a value corresponding to the order item.

[0409] To this end, the device (200) can analyze the shape, pattern, color, or dimensional information of the tiles included in the image using a commonly used image recognition algorithm or object recognition algorithm.

[0410] For example, the device (200) can recognize dimension markings indicating the width and height of a tile in an image and extract a value corresponding to the standard, and can analyze the shape of the construction target space included in the image and extract a value corresponding to the construction location.

[0411] Additionally, the device (200) can recognize a text area included in an image using a commonly used optical character recognition (OCR) algorithm to additionally extract a value corresponding to the quantity, specifications, or delivery time.

[0412] The device (200) can match values ​​corresponding to each order item by comparing the text data analysis results and image data analysis results with order item structure data.

[0413] For example, the device (200) can assign corresponding values ​​to each of the quantity field, specification field, delivery time field, and construction location field included in the order item structure data generated in step S602.

[0414] Additionally, the device (200) can store the item values ​​extracted as described above in each field of the order item structure data, and the stored item values ​​can be used as input data to generate a draft order or automatically enter it into an order form in a subsequent step.

[0415] In step S604, the device (200) can automatically enter the extracted item values ​​to correspond to the order items to generate a draft order.

[0416] Specifically, the device (200) can verify the order item extracted in step S602 and the item value extracted in step S603. At this time, the order item may include a quantity field, a specification field, a delivery time field, a construction location field, or other fields corresponding to the order item, and the item value may be a value extracted to correspond to each of the fields.

[0417] The device (200) can automatically match and enter item values ​​corresponding to each field of the order item. To this end, the device (200) can verify the correspondence between the item value and the order item based on the field identifier or field name included in the order item structure data, and can automatically enter the item value into the corresponding field according to the verified correspondence.

[0418] For example, the device (200) can enter the extracted quantity value into the quantity field and the extracted specification value into the specification field. Additionally, the device (200) can enter the extracted delivery time value into the delivery time field and the extracted construction location value into the construction location field.

[0419] Additionally, the device (200) may perform a value verification or data normalization process before entering the item value. For example, if the specification value includes a length unit, the device (200) may convert the unit to a unified standard unit, and if the delivery time is extracted in a string format rather than a date format, it may convert it to a date data format.

[0420] Additionally, the device (200) may leave the corresponding field value blank or automatically enter a preset default value into the field if the value corresponding to a specific order item is not extracted from the request information. For example, if the quantity is not extracted, the field may be kept blank, and if the delivery date is not extracted, the delivery date may be set to a date after a preset reference period based on the current time. Additionally, if the delivery method is not specified, the default delivery method may be automatically entered.

[0421] Additionally, the device (200) may calculate and enter some item values ​​by considering the relationship between the ordered items. For example, the device (200) may calculate the total quantity using the tile usage per unit area and the construction area and enter it in the quantity field, and may calculate the total amount using the unit price and quantity and enter it in the amount field.

[0422] The device (200) can generate a draft order based on order item structure data in which values ​​corresponding to each order item are entered as described above. At this time, the draft order may be structured data in which each order item value is filled into an order form, and may be generated in JSON, XML, tabular data, or other data formats.

[0423] Additionally, the device (200) may store the draft order in memory or a database, and the draft order may be provided to a user in a subsequent step and used as data for receiving modification or approval input.

[0424] In step S605, the device (200) can provide the generated draft order to the user's terminal (100).

[0425] Specifically, the device (200) can convert the draft order generated in step S604 into screen display data to configure a user interface, and can transmit the user interface to the user's terminal (100).

[0426] At this time, the above draft order may be structured data in which quantity, specifications, delivery time, construction location, unit price, delivery method, payment conditions, or other order items are entered, and the above user interface may include input fields corresponding to each of the above order items.

[0427] The user's terminal (100) can display the user interface through a display, and the user can check or modify the values ​​entered through input fields corresponding to each of the order items. For example, the user can change the values ​​entered in the quantity field, specification field, delivery time field, or construction location field, and can also enter new values.

[0428] Additionally, the user interface may include an approval input means for approving a draft order and a modification input means for modifying a draft order. The approval input means may be a button, a checkbox, a touch area, or other selectable user interface element, and may be selected via touch input, mouse click input, or other input device of the user's terminal (100).

[0429] Additionally, the user interface may include a regeneration input means for requesting the regeneration of a draft order, and if the user selects the regeneration input means, the device (200) may return to the previous step and perform the process of extracting order item values ​​or automatic entry again.

[0430] The device (200) may receive a response signal including a modification input or an approval input for a draft order from the user's terminal (100), and the response signal may be used as input data to generate a final order in a subsequent step or to send an order request message to the tile manufacturer's terminal.

[0431] In step S606, the device (200) can modify or approve the draft order based on user input received from the user's terminal (100).

[0432] Specifically, the device (200) can receive a user response signal including a modification input or an approval input from the user's terminal (100) regarding the draft order provided to the user's terminal (100) in step S605.

[0433] At this time, the user response signal may be a signal generated as the user changes the order item value through an input field included in the user interface or selects an approval input means or a modification input means, and the user response signal may include information indicating the modified order item value or whether it is approved.

[0434] The device (200) can update each order item of the draft order using the information included in the user response signal.

[0435] For example, the device (200) can modify the draft order by reflecting the quantity value, specification value, delivery time value, or construction location value modified by the user in the corresponding order item field.

[0436] Additionally, the device (200) can confirm the modified draft order as the final order if the user's input is confirmed as an approval input.

[0437] At this time, the final purchase order may be structured data with fixed purchase item values, and may be generated in JSON, XML, tabular data, or other structured data formats.

[0438] Additionally, if the device (200) confirms that the user's input is a modification request input, it can return to the order item value extraction step or automatic entry step to regenerate the order draft.

[0439] Additionally, when the device (200) confirms that the user's input is an approved input and a final order is generated, that is, when a signal is received from the user's terminal (100) indicating that the order is no longer necessary to modify, the final order can be stored in a database and used as order data to send an order request message to the tile manufacturer's terminal in a subsequent step.

[0440] Additionally, after the final order form is generated, the device (200) may provide the final order form back to the user's terminal (100) to verify the final order details, and the final order form may be included in a production request message and transmitted to the tile manufacturer's terminal.

[0441] As a result, when a purchase request for similar tiles is received, the device (200) can automatically check the order form and automatically extract and enter order item values ​​based on the request information to create a draft order, and then modify or approve it according to user input, thereby automating the order creation process, reducing the burden of user input, and improving order accuracy and processing efficiency.

[0442] FIG. 7 is a flowchart illustrating the process of applying and providing a simulation according to one embodiment.

[0443] Referring to FIG. 7, first, in step S701, when the device (200) receives a purchase request signal for a similar tile from the user's terminal (100), it can receive spatial information from the user's terminal (100).

[0444] Specifically, the device (200) may request spatial information regarding a target space where the similar tile is to be installed after receiving a purchase request signal for a specific similar tile from the user's terminal (100), and may receive the spatial information from the user's terminal (100).

[0445] At this time, the spatial information may be data indicating the structure, size, and construction conditions of the target space where the tile is to be installed, and may include information on area, shape, humidity, illuminance, intended use, construction location, or spatial zone. Additionally, the spatial information may be data directly entered by the user, or data obtained through communication with a sensor, camera, distance measuring device, or external measuring device equipped on the user's terminal (100).

[0446] First, the device (200) can receive area information of the target space. At this time, the area may be a value representing the floor area or wall area where tiles are to be installed, and may be entered in square meters (m^2), square feet (ft^2), or other area units. Additionally, the area may be a value measured using a space measurement application or a distance measurement sensor equipped on the user's terminal (100).

[0447] Additionally, the device (200) may receive shape information of the target space. In this case, the shape may be information representing the planar structure of the space and may be expressed as any one of a rectangular, square, polygonal, or irregular structure. Additionally, the shape information may include dimensional data including length, width, or dimensions of each side, or it may be spatial structure data generated based on image data of the space or spatial scan data.

[0448] Additionally, the device (200) can receive information on the installation location where the tile is to be installed. At this time, the installation location may be information indicating the location of the structure where the tile is to be installed, and may be entered as any one of the floor, wall surface, ceiling, stairs, object surface, or other structural surface. Additionally, the installation location may be a single location or may be entered as multiple installation locations.

[0449] Additionally, the device (200) may receive spatial zone information. In this case, the spatial zone information may be information indicating whether the target space is divided into multiple construction zones, and, for example, if construction is performed at different locations such as a first zone, a second zone, and a third zone, the area, shape, or construction location of each zone may be entered. Additionally, the spatial zone information may include coordinate information or relative location information of each zone.

[0450] Additionally, the device (200) may receive humidity, illuminance, or usage indicating environmental conditions of the target space, and said usage may be any one of a bathroom, kitchen, living room, hallway, commercial space, or outdoor space.

[0451] The device (200) can receive spatial information received as described above in JSON, XML, or other structured data formats, and can parse the received data and store it in an internal data structure.

[0452] The spatial information received as described above can be used as input data to generate a spatial model, perform construction simulations, or divide construction zones in subsequent steps.

[0453] In step S702, the device (200) can generate a space model where tiles will be installed based on spatial information.

[0454] Specifically, the device (200) can verify the spatial information received in step S701 and can generate a spatial model for the target space where the tile is to be installed using the area, shape, construction location, or spatial zone information included in the spatial information.

[0455] At this time, the space model may be a data structure representing the structure of the space where the tile is to be installed, and may be created as either a 2D planar model or a 3D space model.

[0456] For example, if the shape is input as a rectangle, the device (200) can generate a rectangular spatial model using length and width values, and if the shape is input as a polygon or irregular structure, it can generate a polygonal spatial model using each vertex coordinate or dimension information.

[0457] Additionally, if the spatial information includes multiple construction locations or multiple spatial zone information, the device (200) can generate a spatial model including multiple regions corresponding to each construction location or each spatial zone.

[0458] The spatial model generated as described above can be used as reference data to identify areas to be tiled in subsequent steps or to perform construction simulations.

[0459] At this time, the process of generating a spatial model using spatial information can be performed through commonly used modeling algorithms.

[0460] In step S703, the device (200) can identify an application area, which is an area in the spatial model to which the tile is to be applied.

[0461] Specifically, the device (200) can verify the spatial model generated in step S702 and can identify the area where tiles are to be installed within the spatial model using the construction location or spatial zone information included in the spatial information received in step S701.

[0462] At this time, the application area may refer to a partial area within the space model where the tile is actually to be installed, and may be set as at least one of a floor area, a wall area, a ceiling area, or a specific structural surface.

[0463] For example, if the construction location is entered as a floor, the device (200) can identify an area corresponding to the floor of the space model as an application area, and if the construction location is entered as a wall, it can identify an area corresponding to the wall of the space model as an application area.

[0464] Additionally, if the spatial information includes multiple construction locations or multiple spatial zone information, the device (200) can identify multiple application areas within the spatial model.

[0465] At this time, each application area can be set as an independent construction zone, and each application area can be set to correspond to a different location or a different structural surface.

[0466] Additionally, the device (200) can define the location and range of the application area using coordinate information or area boundary information included in the spatial model, and the application area can be set as a polygonal area, a grid area, or an area of ​​arbitrary shape.

[0467] The device (200) can use the application area identified as above as a reference area to generate a construction simulation with tiles applied in a subsequent step or to divide the application area into a plurality of divided areas.

[0468] In step S704, the device (200) can generate a first construction simulation by applying necessary tiles to the application area.

[0469] Specifically, the device (200) can verify the application area identified in step S703 and can verify the feature vector of the required tile generated in step S202 or the attribute information of the required tile included in the request information.

[0470] At this time, the attribute information of the required tile may include values ​​representing the tile's color, pattern, material, size, surface finish, or other tile characteristics, and the feature vector may be data representing the attribute information in vector form.

[0471] The device (200) can generate example images of necessary tiles based on the feature vector or attribute information. To this end, the device (200) may include a generative artificial intelligence model configured to generate an image corresponding to the input attribute information, or may communicate with the generative artificial intelligence model via wired or wireless means.

[0472] Here, the generative artificial intelligence model may be a model capable of generating images, text, or other data based on given input data, and may include deep learning-based generative models, diffusion models, generative adversarial networks (GANs), large-scale language models (LLMs), or similar artificial intelligence models, and may be implemented in the same way as commonly used generative artificial intelligence models.

[0473] That is, the device (200) can generate an example image assuming that the required tile actually exists by combining visual elements corresponding to the color, pattern, material, size, and surface finish of the required tile.

[0474] For example, the device (200) can generate a tile image by combining a color value corresponding to a color attribute, pattern information corresponding to a pattern attribute, and a glossy or matte expression corresponding to a surface finish attribute.

[0475] Additionally, the device (200) can set multiple tile placement positions within the application area according to the specifications of the required tiles, and can generate a first construction simulation by repeatedly applying the generated example image to each placement position.

[0476] Additionally, the device (200) can determine the arrangement of tiles within the application area using the pattern repetition cycle, joint spacing, or construction direction information of the required tiles, and can place example images according to the arrangement.

[0477] At this time, the first construction simulation may be data that visually represents the construction result when the necessary tile is actually constructed, and may be generated in a 2D image data, 3D rendering data, or other visualization data format.

[0478] The device (200) can use the first construction simulation generated as described above as a reference simulation for comparison with a second construction simulation with similar tiles applied in a subsequent step.

[0479] In step S705, the device (200) can generate a second construction simulation by applying similar tiles to the application area.

[0480] Specifically, the device (200) can verify the application area identified in step S703 and can verify the tile information of the similar tile identified in step S205 or S206.

[0481] At this time, the tile information may include attribute values ​​representing the color, pattern, material, specifications, surface finish, reflectance, or other tile characteristics of the similar tile, and the tile information may also include actual image or texture data of the similar tile.

[0482] The device (200) can place similar tiles in the application area using image or texture data of the similar tiles stored in the tile database.

[0483] For example, the device (200) can set multiple tile placement locations within an application area using specification information of similar tiles, and can visually represent the construction result by repeatedly applying the actual image or texture data to each placement location.

[0484] Additionally, the device (200) can determine the arrangement of tiles within the application area using information on the pattern repetition cycle, joint spacing, or construction direction of similar tiles, and can place images of similar tiles according to the arrangement.

[0485] Additionally, if the application area is separated into multiple construction zones, the device (200) can generate a second construction simulation by applying the same similar tile to each construction zone or by applying the similar tile only to a specific construction zone according to the request information.

[0486] At this time, the second construction simulation may be data that visually represents the construction result when similar tiles are actually constructed, and may be generated in a 2D image data, 3D rendering data, or other visualization data format.

[0487] The device (200) can use the second construction simulation generated as described above as a comparison target simulation to compare with the first construction simulation generated in step S704.

[0488] In step S706, the device (200) can divide the application area into multiple divided areas according to preset criteria.

[0489] Specifically, the device (200) can identify the application area identified in step S703 and can divide the application area into multiple division areas according to a certain size or a certain coordinate standard.

[0490] At this time, the above division criteria may be a criterion for comparing and analyzing the application area, and may be set according to at least one of a grid unit, a coordinate unit, a tile placement unit, or a preset division size unit.

[0491] For example, the device (200) can divide the application area into rectangular areas of the same size, or it can set the divided area to a size corresponding to the dimensions of the tile.

[0492] In addition, the device (200) can generate multiple divided regions using coordinate-based division criteria even when the shape of the application area is an irregular structure.

[0493] Additionally, the device (200) can create a division area for each construction area when the application area consists of multiple construction areas, and can also apply division criteria independently within each construction area.

[0494] Additionally, the device (200) may assign a unique area identification value or coordinate information to each divided area, and the area identification value may be used as a reference value for comparing area-specific construction simulations in a subsequent step.

[0495] The device (200) can use a plurality of divided regions created as above as a reference region for comparing the first construction simulation and the second construction simulation by region in a subsequent step.

[0496] In step S707, the device (200) can calculate the magnitude of the difference between regions by comparing the first construction simulation and the second construction simulation for each of the plurality of divided regions.

[0497] Specifically, the device (200) can verify a plurality of divided regions generated in step S706 and can extract region data of the first construction simulation and region data of the second construction simulation corresponding to each divided region.

[0498] At this time, the first construction simulation may be a simulation showing the construction result with the necessary tiles applied, and the second construction simulation may be a simulation showing the construction result with similar tiles applied.

[0499] The device (200) can compare the visual characteristics of the first construction simulation and the second construction simulation for each divided area.

[0500] At this time, the visual characteristic may include at least one of color value, pattern information, brightness value, texture information, reflectance information, or other data corresponding to a visual representation.

[0501] For example, the device (200) can calculate a color difference, a pattern difference, or a brightness difference by comparing pixel values ​​or texture data included in each divided area, and can convert said difference into a numeric value.

[0502] Additionally, the device (200) can calculate a difference value representing the difference between the first construction simulation and the second construction simulation for each divided area, and the difference value may be a numerical value representing the degree of similarity or difference between the two simulations.

[0503] Additionally, the device (200) can normalize the difference value to convert it into a range of 0 or more and 1 or less so that comparison between different divided regions is possible, and the value can be used as an indicator representing the magnitude of the difference between regions.

[0504] The device (200) can use the difference size per region calculated as above as a reference value for identifying the difference region or determining display attributes in a subsequent step.

[0505] In step S708, the device (200) can identify a region having a difference size higher than a reference difference as a difference region by using the difference size for each region.

[0506] Specifically, the device (200) can check the difference size for each divided area calculated in step S707 and compare the difference size with a reference difference.

[0507] At this time, the reference difference may be a reference value for determining the difference between the first construction simulation and the second construction simulation, and may be set based on any one of a preset fixed value, an average difference value of the entire area, a ratio value to the maximum difference value, or a user-set value. For example, the device (200) may calculate an average value of the difference size of the entire divided area and set the value obtained by multiplying the average value by a preset weight as the reference difference, but is not limited thereto.

[0508] Additionally, the device (200) may set an area having a difference of a certain ratio or more based on the maximum difference value among the entire divided area as a difference area.

[0509] The device (200) can compare the difference size of each divided area with a reference difference and identify a divided area where the difference size is greater than the reference difference as a difference area.

[0510] Additionally, the device (200) can identify all areas exceeding a reference difference among a plurality of divided areas as difference areas, and can store area identification values ​​or coordinate information for each of the identified difference areas.

[0511] Additionally, the device (200) can calculate the number of difference regions or the distribution of difference regions together, and the information can be used as a judgment criterion for determining display attributes in a subsequent step.

[0512] The device (200) can determine the display method of the first construction simulation and the second construction simulation in a subsequent step using the difference area information identified as above.

[0513] In step S709, the device (200) can determine display attributes based on the difference area.

[0514] Specifically, the device (200) can verify the difference regions identified in step S708, and can verify the number of difference regions and the difference size of each difference region.

[0515] At this time, the display attribute may be a value representing a display method for visually emphasizing the difference when providing the first construction simulation and the second construction simulation to the user, and may include at least one of a color display attribute, a saturation attribute, a brightness attribute, a display size attribute, a highlight display attribute, or a display position attribute.

[0516] The device (200) can determine display attributes based on the difference size of each difference area.

[0517] For example, the device (200) can be set to display the area with a first saturation if the difference size of the difference area is greater than or equal to a preset threshold difference, and can be set to display the area with a second saturation lower than the first saturation if the difference size is less than the threshold difference.

[0518] Additionally, the device (200) can determine display attributes based on the number of difference regions.

[0519] For example, the device (200) can be set to display the second construction simulation larger than the first construction simulation when the number of difference areas is greater than or equal to a preset threshold number, and can be set to display the first construction simulation and the second construction simulation at the same size when the number of difference areas is less than the threshold number.

[0520] Additionally, the device (200) can be configured to highlight a specific area based on the location of the difference area, and the highlighting may include at least one of a border display, color highlighting, transparency control, or pattern display.

[0521] The device (200) can use the display attributes determined as above as display control information to apply to the first construction simulation and the second construction simulation in subsequent steps and provide them to the user.

[0522] In step S710, the device (200) can provide a first construction simulation and a second construction simulation to the user's terminal (100) so that display attributes are reflected.

[0523] Specifically, the device (200) can check the display attributes determined in step S709 and can set the display form of the first construction simulation and the second construction simulation according to the display attributes.

[0524] At this time, the display attributes may include at least one of the display saturation of the difference area, whether color is emphasized, display size, transparency, border display, or emphasis pattern, and the device (200) may modify or render the image data of the first construction simulation and the second construction simulation so that the display attributes are reflected.

[0525] For example, the device (200) can be set to display the area corresponding to the difference area with a first color or a second color, and depending on the number of difference areas, the display size of the second construction simulation can be set larger than that of the first construction simulation.

[0526] Additionally, the device (200) can create a screen configuration to display the first construction simulation and the second construction simulation together on the same screen, and can apply any one of the display methods of left-right arrangement, up-down arrangement, or overlapping arrangement.

[0527] Additionally, the device (200) can generate a first construction simulation and a second construction simulation reflecting the display attributes as user interface screen data, and the user interface screen data can be transmitted to the user's terminal (100).

[0528] The user's terminal (100) can receive the user interface screen data and display it through a display, and the user can visually check the application result of similar tiles by comparing the first construction simulation and the second construction simulation.

[0529] Additionally, the device (200) may receive additional input from the user's terminal (100) and may regenerate or re-display a construction simulation for another similar tile according to the input.

[0530] In this way, the device (200) can support intuitively checking the difference between the required tile and the similar tile by providing the user with a construction simulation that reflects the display attributes.

[0531] As a result, the device (200) can improve the accuracy and reliability of tile selection by generating a construction result based on a spatial model when the required tile and similar tile are actually constructed, dividing the application area to analyze the difference between the areas, and then determining and providing display attributes based on the difference area, thereby allowing the user to intuitively compare the difference in construction results between the two tiles.

[0532] FIG. 8 is a flowchart illustrating the process of determining display attributes according to one embodiment.

[0533] Referring to FIG. 8, first in step S801, the device (200) can check the difference size of the difference area.

[0534] Specifically, the device (200) can identify the difference region identified in step S708 and can check the difference size for each region stored corresponding to each of the difference regions.

[0535] In this case, the difference region may refer to a segmented region having a difference greater than the reference difference as a result of comparing the first construction simulation and the second construction simulation, and the region identification value, coordinate information, and difference magnitude value may be stored together in the difference region.

[0536] Additionally, the device (200) can determine the difference size corresponding to each difference area by referring to the difference size for each area calculated in step S707, and the difference size may be a numerical value of the color difference, pattern difference, brightness difference, texture difference, or other visual difference between the first construction simulation and the second construction simulation.

[0537] In addition, the difference size may be a normalized value that allows comparison between different regions, for example, a value converted into a range of 0 to 1.

[0538] The device (200) can determine whether the difference in each difference area is large or small by using the difference size of each difference area, and the result of the determination can be used as a reference value for determining the display saturation, display size, or highlighting in a subsequent step.

[0539] Additionally, the device (200) can calculate at least one of the distribution of the difference size over the entire difference area or the maximum value, minimum value, and average value together, and said value can be used as an additional judgment criterion for determining display attributes in a subsequent step.

[0540] In this way, the device (200) can generate reference data for determining display attributes in a subsequent step by checking the difference size of each difference area.

[0541] In step S802, the device (200) can display the difference region as a first saturation if the difference size of the difference region is greater than or equal to a threshold difference.

[0542] Specifically, the device (200) can determine the difference size of each difference region identified in step S801 and compare the difference size with a preset threshold difference.

[0543] At this time, the threshold difference may be a reference value for determining whether to visually emphasize the difference area, and may be set based on any one of a preset fixed value, an average value of the difference size, a ratio value to the maximum value, or a user-set value. For example, the device (200) may set the threshold difference as a value obtained by multiplying the average value of the difference size of the entire difference area by a preset coefficient, or may set the threshold difference as a value greater than a certain ratio of the maximum difference value in the entire area, but is not limited thereto.

[0544] The device (200) can determine whether the difference size of each difference area is greater than or equal to the threshold difference, and can determine the difference area where the difference size is greater than or equal to the threshold difference as the area to be highlighted.

[0545] Additionally, the device (200) can set display attributes to apply a first saturation to the highlighted target area.

[0546] At this time, the first saturation may be a saturation value for visually emphasizing a difference area and may be set to have a higher saturation value than the second saturation. For example, the device (200) may set the first saturation to a display attribute having a high saturation color, a high brightness value, or a high contrast value. That is, the first saturation value and the second saturation value may each be preset values ​​and may vary depending on the embodiment. Additionally, the first saturation and the second saturation may have the same color value and the same brightness value, and only the saturation values ​​may be set to different values.

[0547] Additionally, the device (200) can be configured to apply at least one of color enhancement, saturation increase, border display, or pattern display so that the difference area with the first saturation applied is visually prominent in the first construction simulation or the second construction simulation.

[0548] The device (200) can use the first saturation information set as above as display control information for applying display attributes in a subsequent step.

[0549] In step S803, the device (200) may display the difference area with a second saturation lower than the first saturation if the difference size of the difference area is less than the threshold difference.

[0550] Specifically, the device (200) can determine the difference size of each difference region identified in step S801 and compare the difference size with a preset threshold difference.

[0551] In this case, the threshold difference may be a reference value for distinguishing the degree of visual emphasis of the difference area, and may be set based on any one of a preset fixed value, an average value of the difference size, a ratio value to the maximum value, or a user-set value.

[0552] The device (200) can determine whether the difference size of each difference region is less than the threshold difference, and can determine the difference region with a difference size less than the threshold difference as a region having a relatively small difference.

[0553] Additionally, the device (200) can set display attributes to apply a second saturation lower than the first saturation to the difference area determined as above.

[0554] At this time, the second saturation may be a display attribute having a lower saturation value compared to the first saturation, and may be a saturation value intended to prevent areas with relatively small differences from being excessively emphasized. For example, the device (200) may set the second saturation to a display attribute having a low saturation color, a low brightness value, or a low contrast value. That is, the first saturation value and the second saturation value may each be preset values ​​and may vary depending on the embodiment.

[0555] Additionally, the device (200) can adjust at least one of the color emphasis level, saturation value, transparency value, or border display so that the difference area with the second saturation applied is visually weaker than the difference area with the first saturation applied.

[0556] The device (200) can use the second saturation information set as above as display control information for applying display attributes in a subsequent step, thereby enabling visual distinction between areas with relatively small and large differences.

[0557] In step S804, the device (200) can check the number of difference regions.

[0558] Specifically, the device (200) can verify the difference region identified in step S708 and calculate the number of segmented regions identified as the difference region.

[0559] In this case, the difference region may refer to a subdivided region having a difference size greater than the reference difference as a result of comparing the first construction simulation and the second construction simulation, and a region identification value or coordinate information may be stored in each difference region.

[0560] The device (200) can count the total number of difference regions using the region identification value or coordinate information, and the number can be used as a value representing the degree of distribution of the difference regions.

[0561] Additionally, the device (200) can determine that the difference between the first construction simulation and the second construction simulation occurs over a wide range as the number of difference regions increases, and can determine that the difference occurs in a limited area as the number of difference regions decreases.

[0562] Additionally, the device (200) may store the number of difference regions as a reference value for comparison with a preset threshold number in a subsequent step, and the comparison result may be used as a judgment criterion for determining the display size or display method of the first construction simulation and the second construction simulation.

[0563] In this way, the device (200) can generate reference data for determining display attributes in a subsequent step by checking the number of difference regions.

[0564] In step S805, the device (200) can display the ratio of the second construction simulation as greater than the first construction simulation if the number of difference regions is greater than or equal to a threshold number.

[0565] Specifically, the device (200) can check the number of difference regions calculated in step S804 and compare the number of difference regions with a preset threshold number.

[0566] At this time, the threshold number may be a reference value for determining whether the difference between the first construction simulation and the second construction simulation occurs over a wide range, and may be set based on any one of a preset fixed value, a ratio value to the total number of divided areas, a user-set value, or a system-set value.

[0567] The device (200) can determine that the difference between the first construction simulation and the second construction simulation occurs over a wide range when the number of difference regions is greater than or equal to the threshold number.

[0568] Additionally, the device (200) can set the display ratio of the second construction simulation to be larger than that of the first construction simulation based on the above judgment result.

[0569] At this time, setting the display ratio to be large may mean increasing at least one of the screen layout ratio, magnification ratio, or display area size so that the second construction simulation is displayed larger to the user.

[0570] For example, when the device (200) displays the first construction simulation and the second construction simulation on the same screen, it may allocate a wider display area for the second construction simulation or apply a magnification scale to the second construction simulation to display it larger than the first construction simulation.

[0571] Additionally, the device (200) can store the display ratio information set as above as a display attribute, and the display attribute can be applied when providing the first construction simulation and the second construction simulation to the user in a subsequent step.

[0572] In this way, the device (200) can provide the user with a clearer view of construction results with large differences by adjusting the display ratio of the construction simulation according to the number of difference areas.

[0573] In step S806, the device (200) can display the first construction simulation and the second construction simulation at the same rate if the number of difference regions is less than the threshold number.

[0574] Specifically, the device (200) can check the number of difference regions calculated in step S804 and compare the number of difference regions with a preset threshold number.

[0575] At this time, the threshold number may be a reference value for determining whether the difference between the first construction simulation and the second construction simulation occurs over a wide range, and may be set based on any one of a preset fixed value, a ratio value to the total number of divided areas, a user-set value, or a system-set value.

[0576] If the number of difference regions is less than the threshold number, the device (200) can determine that the difference between the first construction simulation and the second construction simulation occurs within a limited range.

[0577] Additionally, the device (200) can set the display ratio of the first construction simulation and the second construction simulation to be the same based on the above judgment result.

[0578] In this case, displaying at the same ratio may mean that the first construction simulation and the second construction simulation are provided to the user with the same screen layout ratio, the same magnification scale, or the same display area size.

[0579] For example, the device (200) may place the first construction simulation and the second construction simulation in display areas of the same size that are divided left and right or up and down, and may also display both simulations with the same magnification ratio applied.

[0580] Additionally, the device (200) can store the same ratio information set as above as a display attribute, and the display attribute can be applied when providing the first construction simulation and the second construction simulation to the user in a subsequent step.

[0581] In this way, the device (200) can provide the user with the ability to compare construction results that are not significantly different in a balanced manner by adjusting the display ratio of the construction simulation according to the number of difference areas.

[0582] As a result, the device (200) determines the saturation and display ratio based on the difference size and the number of difference areas of the difference area, and provides the first construction simulation and the second construction simulation in different display ways, thereby visually highlighting the difference in construction results between the required tile and the similar tile, and providing intuitive comparison information to the user.

[0583] FIG. 9 is a flowchart illustrating the process of utilizing multiple tiles when the tile inventory is insufficient according to one embodiment.

[0584] Referring to FIG. 9, first, in step S901, when the device (200) receives a purchase request signal for similar tiles from the user's terminal (100), it can check the requested quantity of necessary tiles included in the request information.

[0585] Specifically, the device (200) can receive a purchase request signal for a specific similar tile from the user's terminal (100) through a user interface provided to the user's terminal (100) in step S205 or S206.

[0586] At this time, the purchase request signal may be a signal generated as a user requests to purchase or commission the production of a similar tile selected by the user, and the purchase request signal may include request information along with tile identification information for identifying the selected similar tile.

[0587] In addition, the above request information may include request quantity information indicating the quantity of required tiles, and the request quantity information may be a value directly entered by the user, a value automatically calculated based on the construction area, or a value extracted by analyzing text data or image data included in the request information.

[0588] For example, the device (200) can calculate the total quantity of tiles required based on the tile usage per unit area using space information or construction area information included in the request information, and can set the calculated value as the requested quantity.

[0589] Additionally, the device (200) may extract a number or unit representing the quantity from text data included in the request information, or analyze a design drawing or construction drawing included in the image data to calculate the quantity of necessary tiles.

[0590] Additionally, the requested quantity may be expressed in the number of tiles, box unit quantity, area unit quantity, or other units, and the device (200) can verify the requested quantity by converting the quantity entered in different units into the same standard unit.

[0591] The device (200) can store the requested quantity confirmed as above, and the requested quantity can be used as a reference value for comparing with the inventory of similar tiles in a subsequent step or for calculating an additional requested quantity.

[0592] In this way, the device (200) can generate reference data for determining whether there is a shortage of stock in a subsequent step by checking the requested quantity of necessary tiles included in the request information.

[0593] In step S902, the device (200) can look up inventory based on tile information of similar tiles.

[0594] Specifically, the device (200) can query a tile database using tile identification information included in a purchase request signal in step S901, and can check tile information of a similar tile corresponding to the tile identification information.

[0595] At this time, the tile information may include attribute values ​​representing the color, pattern, material, specifications, surface finish, reflectance, strength, or other tile characteristics of the corresponding similar tile, and the tile information may also include inventory data representing the inventory information of the corresponding tile.

[0596] Additionally, the above inventory information may be a value representing the quantity of tiles currently held, and may be stored in units of tile count, box unit quantity, area unit quantity, or other units.

[0597] Additionally, the device (200) can directly query inventory information stored in the tile database, or it can query current inventory information in real time by performing wired or wireless communication with the tile manufacturer's terminal, logistics server, or external inventory management system.

[0598] Additionally, if the inventory information is stored in different units, the device (200) can convert the inventory quantity into the same unit using a preset standard unit. For example, the inventory quantity stored in box units can be converted into the number of tiles or area units.

[0599] The device (200) can check the inventory quantity retrieved as described above, and the inventory quantity can be used as a reference value to calculate an additional request quantity by comparing it with the requested quantity in a subsequent step.

[0600] In this way, the device (200) can generate reference data to determine whether the requested quantity can be satisfied by checking the inventory of similar tiles.

[0601] In step S903, the device (200) can calculate an additional request amount by subtracting the inventory from the requested quantity, based on the fact that the inventory of similar tiles is less than the requested quantity.

[0602] Specifically, the device (200) can check the requested quantity confirmed in step S901 and the inventory of similar tiles queried in step S902.

[0603] At this time, the requested quantity and inventory may be expressed in tile count units, box units, area units, or other units, and the device (200) may convert to a preset standard unit and then perform a comparison if expressed in different units.

[0604] The device (200) can determine whether the inventory is less than the requested quantity by comparing the requested quantity with the inventory.

[0605] If the inventory is greater than the requested quantity, the device (200) may determine that no additional requested quantity is needed and may set the additional requested quantity to 0.

[0606] On the other hand, if the stock is less than the requested quantity, the device (200) can calculate the insufficient quantity by performing an operation to subtract the stock quantity from the requested quantity.

[0607] In this case, the additional request quantity can be calculated by 'Additional request quantity = Request quantity stock quantity'.

[0608] Additionally, the device (200) can correct the additional request amount to 0 when the calculation result is less than 0 in order to prevent the calculation result from becoming negative.

[0609] Additionally, if the requested quantity or stock quantity is expressed in area units, the device (200) may calculate the additional requested quantity by converting it into tile count units using the tile usage per unit area.

[0610] The device (200) can store the additional request amount calculated as above, and the additional request amount can be used as a reference value for selecting additional similar tiles or performing multiple manufacturer placement in a subsequent step.

[0611] In this way, the device (200) can determine cases where the requested quantity cannot be satisfied with a single tile by calculating an additional requested quantity using the difference between the requested quantity and the stock quantity, and can generate reference data for searching for a replacement tile in a subsequent step.

[0612] In this process, the device (200) may determine that no additional request is made if, based on the comparison result, the inventory of similar tiles is greater than or equal to the requested quantity.

[0613] That is, if the additional request amount is 0 or negative, the device (200) may not perform subsequent steps for selecting additional similar tiles, and may determine that the requested amount can be satisfied with only the similar tiles, and accordingly, the device (200) may not perform steps after step S903.

[0614] Additionally, if the device (200) determines that the requested quantity can be satisfied as described above, it can proceed with the ordering procedure based on similar tiles by performing the process of FIG. 6 or FIG. 7.

[0615] In this way, the device (200) can prevent unnecessary additional tile search or placement calculations and improve processing efficiency by selectively determining whether to perform a subsequent step based on the result of comparing the requested quantity and the inventory.

[0616] In step S904, the device (200) can search the tile database and select the tile with the highest similarity to the similar tile among the tiles that have inventory capable of responding to the additional request amount as the additional similar tile.

[0617] Specifically, the device (200) can check the additional request amount calculated in step S903 and can search the tile database to search for tiles that have an additional request amount or a stock quantity that corresponds to the additional request amount.

[0618] At this time, the device (200) can query tile information containing inventory information among a plurality of tile information stored in a tile database, and can determine whether the inventory quantity of each tile is greater than or equal to the additional request quantity.

[0619] Additionally, the device (200) can filter only tiles whose stock quantity is greater than or equal to the additional request quantity as candidate tiles, and can calculate the similarity with the similar tile for each of the candidate tiles.

[0620] At this time, the similarity can be calculated using attribute values ​​representing the color, pattern, material, size, surface finish, or other tile characteristics of the tile, and the similarity can be calculated in the same manner as the similarity calculation method performed in step S203 or step S301.

[0621] Additionally, the device (200) can compare the similarity calculated for each of the candidate tiles and select the tile with the highest similarity.

[0622] At this time, if multiple tiles have the same degree of similarity, the device (200) may select a tile with a larger stock quantity, a tile with a more stable supply history, or a tile that satisfies a preset priority criterion as an additional similar tile.

[0623] The device (200) can set the tile selected as above as an additional similar tile, and the additional similar tile can be used as a substitute tile to supplement the insufficient quantity when the requested quantity cannot be met with the inventory of similar tiles.

[0624] In this way, the device (200) can provide tiles that are as similar in appearance and characteristics as possible while satisfying the requested quantity by having an inventory capable of responding to the additional requested quantity and selecting tiles that have a high degree of similarity to similar tiles as additional similar tiles.

[0625] In step S905, the device (200) receives spatial information from the user's terminal (100) and can confirm the construction location where the tile will be installed.

[0626] Specifically, the device (200) can receive spatial information from the user's terminal (100), and the spatial information may include the area, shape, zone information, coordinate information, or construction location information of the target space where the tile is to be installed.

[0627] At this time, the above construction location information may be information indicating the location of the structure where the tile is to be installed, and may include, for example, at least one of a floor, wall, ceiling, stairs, exterior space, or other structural surface.

[0628] In addition, the above construction location information may include only one location or may include multiple different locations, and may include, for example, construction locations for different rooms, different wall surfaces, or separated spatial zones.

[0629] Additionally, the device (200) can divide the location where the tile is to be installed into a plurality of installation zones using coordinate information or zone identification information included in the spatial information, and can set each installation zone as an independent installation location.

[0630] Additionally, the device (200) may use distance between each construction location, zone division information, or spatial structure information to determine whether the construction locations correspond to separate spaces.

[0631] The device (200) can store the construction location identified as above, and the construction location information can be used as reference data to determine whether similar tiles and additional similar tiles can be placed at different construction locations in a subsequent step or to generate a mixed placement pattern.

[0632] In this way, the device (200) can generate spatial reference information to determine whether multiple tiles can be distributed by verifying the construction location using spatial information.

[0633] In step S906, the device (200) can determine whether similar tiles and additional similar tiles can be installed at different construction locations based on the construction location, the inventory of similar tiles, and the additional requested amount.

[0634] Specifically, the device (200) can verify the construction location identified in step S905 and determine whether the construction zone or construction location included in the construction location corresponds to an independent space.

[0635] Additionally, the device (200) can check the stock quantity of similar tiles retrieved in step S902 and the additional request quantity calculated in step S903, and can also check the stock quantity of additional similar tiles selected in step S904.

[0636] At this time, the device (200) can calculate the quantity of tiles required for each construction location, and the quantity required for each construction location can be calculated based on area information or construction zone information included in the spatial information.

[0637] Additionally, the device (200) can determine the construction area that can be satisfied with the stock quantity of similar tiles and the construction area that can be satisfied with the stock quantity of additional similar tiles, respectively.

[0638] Additionally, the device (200) may determine that when different construction zones are physically separated, the likelihood of visual or functional problems occurring is low even if different tiles are installed, and in such cases, it may determine that it is possible to place similar tiles and additional similar tiles at different construction locations.

[0639] On the other hand, the device (200) may determine that it is difficult to install different tiles separately when the construction location consists of a single continuous space or adjacent areas.

[0640] Additionally, the device (200) can determine whether the required quantity for each construction location can be met by considering the stock quantity of similar tiles and the stock quantity of additional similar tiles, and if it is determined that construction with different tiles is possible for each construction location, it can determine that the similar tiles and additional similar tiles can be constructed at different construction locations.

[0641] The device (200) can store the above judgment result, and the above judgment result can be used as reference information to determine whether to separate and arrange tiles by construction location or to generate a mixed arrangement pattern in a subsequent step.

[0642] In this way, the device (200) can determine a tile arrangement method suitable for the spatial structure by determining whether multiple tiles can be distributed and installed by considering the installation location and the stock quantity together.

[0643] If it is determined in step S906 that similar tiles and additional similar tiles can be installed at different installation locations, in step S907, the device (200) can set each of the separated installation locations as an independent installation zone.

[0644] Specifically, the device (200) corresponds to a space where the construction locations are separated from each other in step S906, and can confirm the result that independent construction is possible for each construction location using the inventory quantity of similar tiles and additional similar tiles.

[0645] At this time, the device (200) can identify multiple locations where tiles are to be installed using construction location information included in the spatial information, and can divide each construction location into independent construction zones.

[0646] Additionally, the device (200) can set the range of each construction zone using area information, coordinate information, or zone identification information corresponding to each construction location, and can define each construction zone as an independent area where different tiles can be applied.

[0647] For example, if there are different rooms, different wall surfaces, or separate floor areas, the device (200) can set each area as a first construction area, a second construction area, or other construction area.

[0648] Additionally, the device (200) can calculate the quantity of tiles required for each construction zone, and the quantity required for each construction zone can be calculated based on the area or construction scope of the zone.

[0649] Additionally, the device (200) can generate and store construction zone information set as above as a data structure, and the construction zone information can be used as reference information for determining tiles to be placed in each zone.

[0650] The device (200) can use the independent construction zones set as above as a standard for matching similar tiles and additional similar tiles to each construction zone in a subsequent step.

[0651] In this way, the device (200) can divide the spatial structure so that different tiles can be placed at different locations by setting the separated construction locations as independent construction zones.

[0652] In step S908, the device (200) can generate matching information by matching similar tiles and additional similar tiles to each construction zone.

[0653] Specifically, the device (200) can verify a plurality of construction zones set in step S907, and can verify the area, location information, and required quantity information corresponding to each construction zone.

[0654] Additionally, the device (200) can check the stock quantity of similar tiles retrieved in step S902 and the stock quantity of additional similar tiles selected in step S904.

[0655] At this time, the tile information for each of the similar tiles and additional similar tiles may include attribute values ​​representing color, pattern, material, dimensions, surface finish, or other tile characteristics.

[0656] The device (200) can determine which tile to place in which construction zone by comparing the required quantity of each construction zone with the stock quantity of similar tiles and additional similar tiles.

[0657] For example, the device (200) may first place similar tiles in construction zones that can be satisfied with the stock quantity of similar tiles, and place additional similar tiles in the remaining construction zones.

[0658] Additionally, the device (200) can place similar tiles and additional similar tiles considering the location, area, or importance of the construction area.

[0659] For example, similar tiles can be placed first in construction areas that are more exposed to the user's field of vision, larger construction areas, or main spaces, while additional similar tiles can be placed in construction areas with relatively lower exposure or auxiliary areas.

[0660] Additionally, the device (200) can generate matching information indicating the corresponding relationship between each construction zone and the tiles placed therein.

[0661] At this time, the matching information may include construction area identification information, tile placement information by construction area, and placement quantity or placement location information of the corresponding tiles.

[0662] Additionally, the device (200) may store the matching information in a data structure, and the matching information may be provided through a user interface in a subsequent step or used as reference information to induce the purchase of additional similar tiles.

[0663] In this way, the device (200) can generate tile placement information suitable for the spatial structure while satisfying the requested quantity by matching similar tiles and additional similar tiles to each construction zone.

[0664] In step S909, the device (200) may provide a user interface to the user's terminal (100) that includes, along with matching information, tile information of additional similar tiles and input means for generating a purchase request signal for additional similar tiles.

[0665] Specifically, the device (200) can check the matching information generated in step S908, and the matching information may include the type of tile placed in each construction zone, the quantity of tiles placed, the placement location, or the construction zone identification information.

[0666] Additionally, the device (200) can check tile information of an additional similar tile selected in step S904, and said tile information may include attribute values ​​indicating the color, pattern, material, specifications, surface finish, stock quantity, or other tile characteristics of the additional similar tile.

[0667] The device (200) can configure a user interface screen using the matching information and additional similar tile information.

[0668] At this time, the user interface may visually display information about each construction zone and the tiles placed in the corresponding construction zone, and the zone where similar tiles are placed and the zone where additional similar tiles are placed may be displayed separately.

[0669] Additionally, the device (200) may include an input means for generating a purchase request signal for additional similar tiles in the user interface.

[0670] At this time, the input means may be a button, a checkbox, a selection area, or other selectable graphic user interface element, and may be selected through touch input, mouse input, or other input device of the user's terminal (100).

[0671] When a user selects the input means through the user's terminal (100), the user's terminal (100) may generate a purchase request signal for additional similar tiles and transmit it to the device (200), and the purchase request signal may include tile identification information of the additional similar tiles and purchase request quantity information.

[0672] Additionally, the device (200) may include a display area in the user interface that allows checking the placement results by construction zone, and the user can check the placement results in advance when purchasing additional similar tiles through the display area.

[0673] The device (200) can transmit a user interface configured as above to a user's terminal (100), and the user interface can be displayed through a display provided on the user's terminal (100).

[0674] In this way, the device (200) can induce the user to purchase additional tiles to satisfy the requested quantity by providing a user interface for purchasing additional similar tiles along with matching information.

[0675] If it is determined in step S906 that similar tiles and additional similar tiles cannot be installed at different installation locations, in step S910, the device (200) can generate a cross arrangement pattern so that similar tiles and additional similar tiles are mixed and arranged based on the inventory of similar tiles and the additional requested amount.

[0676] Specifically, the device (200) can confirm the result that it is difficult to install similar tiles and additional similar tiles separately at different (i.e., separated) installation locations in step S906.

[0677] In this case, the above judgment result may be a case where it is determined that there is a high possibility of visual dissonance or construction problems occurring if different tiles are placed separately, as the construction location is formed as a single continuous space or consists of adjacent areas.

[0678] Additionally, the device (200) can check the inventory of similar tiles retrieved in step S902 and the additional request amount calculated in step S903, and can determine the placement ratio to which similar tiles and additional similar tiles are to be applied based on the inventory quantity of similar tiles and the additional request amount.

[0679] For example, the device (200) can calculate the ratio of the number of similar tiles and additional similar tiles to be applied or the ratio of the area based on the ratio of the inventory of similar tiles and the additional request amount.

[0680] Additionally, the device (200) can adjust the placement ratio so that the above ratio is uniformly reflected throughout the entire construction area.

[0681] Additionally, the device (200) can generate an intersecting arrangement pattern such that similar tiles and additional similar tiles are mixed and arranged according to the above arrangement ratio.

[0682] In this case, the cross arrangement pattern may be a pattern in which different tiles are arranged alternately according to a certain rule, and may include at least one of a checkerboard pattern, a stripe pattern, a mosaic pattern, or other repeating arrangement patterns, but is not limited thereto.

[0683] For example, the device (200) can generate a checkerboard-shaped cross arrangement pattern when the ratio of similar tiles to additional similar tiles is close to 1:1, and can generate a stripe pattern that repeats in rows or columns when the ratio of specific tiles is higher. Additionally, the device (200) can adjust the repeating direction or spacing of the pattern according to the shape or construction direction of the application area.

[0684] Additionally, the device (200) can generate placement information that corresponds the cross arrangement pattern to an application area or a divided area, and the placement information may include location-specific tile identification information indicating which tile is applied to each placement location.

[0685] Additionally, the device (200) may store the cross-arrangement pattern information, and the cross-arrangement pattern information may be provided through a user interface in a subsequent step or used as reference information to induce a purchase request for additional similar tiles.

[0686] In this way, the device (200) can meet the requested quantity while maintaining construction quality and visual balance by generating an inter-arrangement pattern in which similar tiles and additional similar tiles are mixed and arranged even when the construction location cannot be separated and tiles cannot be placed.

[0687] In step S911, the device (200) may provide a user interface to the user's terminal (100) that includes an input means for generating tile information of additional similar tiles and a purchase request signal for additional similar tiles, along with a cross arrangement pattern.

[0688] Specifically, the device (200) can verify the cross arrangement pattern generated in step S910, and the cross arrangement pattern may include arrangement information indicating a tile type, arrangement order, or arrangement rule corresponding to each placement position within the application area.

[0689] Additionally, the device (200) can generate construction simulation data to visually display the cross arrangement pattern, and the construction simulation data may include a state in which similar tiles and additional similar tiles are mixed and arranged in the form of an image or graphic.

[0690] Additionally, the device (200) can check tile information of an additional similar tile selected in step S904, and said tile information may include attribute values ​​indicating the color, pattern, material, specifications, surface finish, stock quantity, or other tile characteristics of the additional similar tile.

[0691] The device (200) can generate a user interface screen including a construction simulation according to the cross arrangement pattern and tile information of additional similar tiles. At this time, the user interface may display information of additional similar tiles along with a display area where the mixed arrangement result is displayed.

[0692] Additionally, the device (200) may include an input means for generating a purchase request signal for additional similar tiles in the user interface. In this case, the input means may be a button, a checkbox, a selection area, or other selectable graphic user interface element, and may be selected via touch input, mouse input, or other input device of the user's terminal (100).

[0693] When a user selects the input means through the user's terminal (100), the user's terminal (100) may generate a purchase request signal for additional similar tiles and transmit it to the device (200), and the purchase request signal may include tile identification information of the additional similar tiles and purchase request quantity information.

[0694] Additionally, the device (200) can transmit the user interface to the user's terminal (100), and the user interface can be displayed through a display provided on the user's terminal (100).

[0695] In this way, the device (200) can present a tile arrangement to the user that can maintain construction quality while satisfying the requested quantity by providing a user interface for purchasing additional similar tiles along with the mixed arrangement result according to the cross arrangement pattern.

[0696] As a result, the device (200) can automatically select additional similar tiles by comparing the requested quantity of necessary tiles with the inventory, determine separate placement or mixed placement by considering the spatial structure according to the construction location, and provide the user with placement results and additional purchase information, thereby supporting the maintenance of construction quality and visual consistency while satisfying the requested quantity.

[0697] FIG. 10 is an example diagram of the configuration of a device according to one embodiment.

[0698] A device (200) according to one embodiment includes a processor (210) and a memory (220). The processor (210) may include at least one device described with reference to FIGS. 1 through 9 or may perform at least one method described with reference to FIGS. 1 through 9. A person or organization using the device (200) may provide services related to some or all of the methods described with reference to FIGS. 1 through 9.

[0699] The memory (220) may store information related to the methods described above or store a program in which the methods described below are implemented. The memory (220) may be volatile memory or non-volatile memory.

[0700] The processor (210) can execute a program and control the device (200). The code of the program executed by the processor (210) can be stored in memory (220). The device (200) can be connected to an external device (e.g., a personal computer or a network) through an input / output device (not shown in the drawing) and can exchange data via wired or wireless communication. The processor (210) can be operatively connected to the components of the device (200). The processor (210) can load commands or data received from other components of the device (200) into memory (220), process the commands or data stored in memory (220), and store the resulting data.

[0701] Additionally, the device (200) may further include a communication circuit. The communication circuit may establish a communication channel with an external device (e.g., a user's terminal (100)) and transmit and receive various data with the external device. According to various embodiments, the communication circuit may include a cellular communication module and be configured to be connected to a cellular network (e.g., 3G, LTE, 5G, Wibro, or Wimax). According to various embodiments, the communication circuit may include a short-range communication module and transmit and receive data with the external device using short-range communication (e.g., Wi-Fi, Bluetooth, Bluetooth Low Energy (BLE), UWB), but is not limited thereto.

[0702] Additionally, the device (200) may be used to train an artificial neural network or to use a trained artificial neural network. The memory (220) may contain an artificial neural network that is being trained or has been trained. The processor (210) may train or execute an artificial neural network algorithm stored in the memory (220). The device (200) for training the artificial neural network and the device (200) for using the trained artificial neural network may be the same or separate.

[0703] The embodiments described above may be implemented as hardware components, software components, and / or combinations of hardware and software components. For example, the devices, methods, and components described in the embodiments may be implemented using one or more general-purpose or special-purpose computers, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing unit may execute an operating system (OS) and one or more software applications executed on said operating system. Additionally, the processing unit may access, store, manipulate, process, and generate data in response to the execution of the software. For ease of understanding, the processing unit may be described as being used as a single unit, but those skilled in the art will understand that the processing unit may include multiple processing elements and / or multiple types of processing elements. For example, the processing unit may include multiple processors or one processor and one controller. Additionally, other processing configurations, such as parallel processors, are also possible.

[0704] The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc., either alone or in combination. The program instructions recorded on the medium may be those specifically designed and configured for the embodiment, or they may be those known and available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, and flash memory. Examples of program instructions include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. The hardware devices described above may be configured to operate as one or more software modules to perform the operation of the embodiment, and vice versa.

[0705] Software may include computer programs, code, instructions, or a combination of one or more of these, and may configure a processing unit to operate as desired or command the processing unit independently or collectively. Software and / or data may be permanently or temporarily embodied in any type of machine, component, physical device, virtual equipment, computer storage medium or device, or transmitted signal wave so as to be interpreted by the processing unit or to provide instructions or data to the processing unit. Software may be distributed over networked computer systems and may be stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

[0706] Although the embodiments have been described above with reference to the limited drawings, those skilled in the art can apply various technical modifications and variations based on the above. For example, suitable results may be achieved even if the described techniques are performed in a different order than described, and / or if the components of the described system, structure, device, circuit, etc. are combined or assembled in a form different from described, or replaced or substituted by other components or equivalents.

[0707] Therefore, other implementations, other embodiments, and equivalents to the claims also fall within the scope of the claims set forth below.

Claims

Claim 1 A method for providing an automatic selection and ordering service for image and text-based tiles, performed by a device, comprising: receiving request information for a required tile from a user's terminal, the request information including at least one of image data or text data of the required tile; analyzing the request information to extract a feature value including at least one of the color, pattern, material, specifications, and use of the required tile, and vectorizing the feature value to generate a feature vector; calculating a similarity between tile information corresponding to each tile stored in a tile database and the feature vector, and extracting tiles with a similarity greater than or equal to a preset reference similarity as similar tiles; checking the number of extracted similar tiles; if the number of similar tiles is confirmed to be one, providing a user interface to the user's terminal that includes tile information of the similar tile and input means for generating a purchase request signal for the similar tile; if the number of similar tiles is confirmed to be two or more, calculating a priority score for each of the similar tiles and providing a user interface to the user's terminal that includes a list of similar tiles sorted according to the priority score and selection means for generating a selection and purchase request signal for the similar tiles. and if the number of similar tiles is confirmed to be zero, the step of generating a tile production request message including request information for the necessary tile and transmitting it to the tile manufacturer's terminal is included, and the step of generating a tile production request message including request information for the necessary tile and transmitting it to the tile manufacturer's terminal includes: an operation of identifying a tile manufacturer using manufacturer identification information stored corresponding to each tile stored in the tile database; an operation of selecting a tile manufacturer among the tile manufacturers that is matched with a number of tiles greater than or equal to a preset target number as a candidate tile manufacturer; and for each of the candidate tile manufacturers,The operation of collecting tile information of tiles produced by the candidate tile makers stored in the tile database; the operation of decomposing the feature vector of the required tile into multiple attribute elements including material, color, pattern, specifications, and surface finish; for each of the candidate tile makers, the operation of verifying whether an attribute corresponding to each of the multiple attribute elements exists in the collected tile information and calculating an attribute implementation possibility score based on the ratio of implementable attribute elements; for each of the candidate tile makers, the operation of verifying whether there is a past tile in which two or more of the multiple attribute elements are simultaneously implemented and calculating an attribute combination implementation score based on the ratio of simultaneously implemented attribute combinations; for each of the candidate tile makers, the operation of analyzing the order fulfillment history of tiles produced by the candidate tile makers stored in the tile database to calculate the number of fulfillments, fulfillment compliance rate, recent fulfillment frequency, fulfillment period variability, and fulfillment gap period, and calculating a fulfillment stability score based thereon; the operation of calculating a comprehensive score of the candidate tile makers based on the attribute implementation possibility score, the attribute combination implementation score, and the fulfillment stability score; and selecting a candidate tile maker whose comprehensive score is higher than a preset target score as a consideration tile maker. A selection operation; an operation to check whether there are multiple tile manufacturers in consideration; if it is confirmed that there are multiple tile manufacturers in consideration, a user interface including a designation means for generating a tile manufacturer list by sorting the tile manufacturers in consideration according to the comprehensive score and generating a designation for the tile manufacturer list and the tile manufacturers in consideration is provided to the user's terminal; a designation signal for a tile manufacturer subject to a production request request is received from the user's terminal; and a tile production request message is transmitted to the terminal of the designated tile manufacturer subject to the production request request.A method for providing an automatic selection and ordering service for image and text-based tiles, comprising the operation of, if it is confirmed that there is only one or no tile manufacturer to be considered, sending the tile production request message first to the terminal of the candidate tile manufacturer with the highest overall score. Claim 2 In claim 1, the step of calculating a priority score for each of the similar tiles comprises: for the similar tile, an operation of normalizing the similarity value between the tile information and the feature vector to calculate a similarity score; for the similar tile, an operation of receiving spatial information from the user's terminal including the area, shape, humidity, illuminance, intended use, and construction location of the space where the necessary tile is to be placed; for the similar tile, an operation of comparing the arrangement information of the similar tile with reference arrangement information including the pattern repetition cycle, joint spacing, and construction direction of the necessary tile, and for the similar tile, an operation of calculating an arrangement error value by additionally considering the area and shape included in the spatial information, and an operation of calculating an arrangement distortion minimization score which is converted to a higher value as the arrangement error value is smaller; for the similar tile, an operation of quantifying and calculating the probability of joint contamination occurrence based on the material of the similar tile and the humidity and intended use, and calculating a joint contamination prediction score which is converted to a higher value as the probability of joint contamination occurrence is lower; for the similar tile, an operation of calculating a reflectance reference value based on the illuminance and intended use, and calculating a difference value between the reflectance of the similar tile and the reflectance reference value, and for the similar tile, a higher value as the difference value is smaller The operation of calculating a converted thermal reflection influence score; for the similar tile, quantifying and calculating the usage frequency according to the intended use and the expected load according to the construction location; calculating the risk of damage by comparing the usage frequency and the expected load with the strength of the similar tile; and calculating a damage risk score that is converted to a higher value as the risk of damage decreases; for the similar tile, calculating a price ratio value by dividing the construction price of the similar tile by a standard price; calculating a period ratio value by dividing the expected construction period of the similar tile by a standard period; and calculating a price score and a delivery period score that are converted to a higher value as the price ratio value and the period ratio value decrease.A method for providing an automatic selection and ordering service for image and text-based tiles, comprising the operation of calculating a priority score of similar tiles based on the similarity score, the array distortion minimization score, the joint contamination prediction score, the heat reflection influence score, the damage risk score, the price score, and the delivery time score. Claim 3 delete

Images