Multi-dimensional fusion recommendation and selection method and system for material taking of material management intelligent warehouse

Abstract

The invention belongs to the technical field of intelligent material management warehouses, and particularly discloses a multi-dimensional fusion recommendation and selection method and system for material taking of an intelligent material management warehouse, and the method comprises the following steps: collecting each-time taking record data in past set time, and carrying out the preprocessing of the data, and obtaining a material recommendation combination; obtaining work task data and external factors of the day, and predicting the type and quantity of materials which need to be taken by the installation and maintenance personnel in the day in combination with a material recommendation combination; inventory conditions of a warehouse and a personal library are monitored in real time, it is ensured that recommended materials do not exceed inventory limitation, and recommendation is dynamically adjusted according to inventory data; and generating a material recommendation list based on the historical receiving data, the demand prediction of the day and the inventory data. According to the technical scheme, personalized recommendation is realized through multi-dimensional data fusion, and the material receiving accuracy is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent warehouse for material management, and relates to a method and system for multi-dimensional fusion recommendation and selection of materials for intelligent warehouse for material management. Background Art

[0002] As home users' demands and requirements for the Internet gradually increase, more and more communication devices are needed. In order to meet the needs of different users, front-line installation and maintenance personnel need to carry a variety of equipment for on-site work, and they need to collect more than a dozen types of equipment and accessories every day.

[0003] At present, the collection of materials is entirely determined by the experience of front-line installation and maintenance personnel. The type, manufacturer, model, quantity and combination of materials collected need to be completed based on experience. There is a lack of automated and intelligent systems to assist in completing the collection of materials.

[0004] The existing material management intelligent warehouse material collection process mainly relies on manual operation. The overall process includes the following steps:

[0005] Frontline installation and maintenance personnel fill out the demand report form (paper or electronic form) based on their experience. The report content includes the type, model and quantity of materials.

[0006] Warehouse managers manually check inventory, allocate materials and record transfer requirements based on the reported content, lacking intelligent support;

[0007] After arriving at the warehouse, the installation and maintenance personnel check and receive the materials, and register them using a paper list or barcode scanning equipment;

[0008] The warehouse manager then manually updates the inventory records, including adjusting the inventory status and recording the transfer;

[0009] If the inventory is insufficient, the administrator needs to manually coordinate with other warehouses to transfer goods, and the driver needs to plan the delivery route based on experience.

[0010] The entire process relies entirely on manual judgment, which is inefficient and lacks the ability to respond to dynamic demands, making it difficult to meet the needs for accurate and efficient material collection. Summary of the invention

[0011] The purpose of the present invention is to provide a multi-dimensional fusion recommendation and selection method and system for material delivery in a material management intelligent warehouse, so as to realize intelligent recommendation of materials.

[0012] In order to achieve the above-mentioned purpose, the basic scheme of the present invention is: a multi-dimensional fusion recommendation and selection method for material delivery in a material management intelligent warehouse, comprising the following steps:

[0013] Collect the data of each collection record within the past set time, and pre-process it to obtain the recommended combination of materials;

[0014] Obtain the day's work task data and external factors, and combine them with the recommended material combination to predict the types and quantities of materials that installation and maintenance personnel need to collect on the day;

[0015] Monitor the inventory status of warehouses and personal libraries in real time to ensure that recommended materials do not exceed inventory limits, and dynamically adjust recommendations based on inventory data;

[0016] Generate a recommended list of materials based on historical requisition data, current demand forecast and inventory data.

[0017] The working principle and beneficial effects of this basic solution are: This technical solution completes integrated recommendations based on data such as daily user needs, historical collection of front-line installation and maintenance, warehouse inventory quantity, personal warehouse inventory quantity, daily predicted usage, material combination, accessory configuration, etc., thereby reducing manual dependence and realizing intelligent collection.

[0018] Dynamically adjust the recommended results during the material recommendation process to ensure that they match real-time needs. Reduce the manual dependence of installation and maintenance personnel and achieve efficient and accurate material collection.

[0019] Furthermore, the method for preprocessing the data of each collection record within the past set time is as follows:

[0020] Remove duplication, complete and standardize the collected data, and fill in missing values;

[0021] Use cluster analysis to identify common material combinations and matching patterns;

[0022] Count the usage frequency of each material and identify high-frequency materials;

[0023] Identify the material collection patterns of each installation and maintenance personnel in specific scenarios in historical data, generate personalized material collection combinations and recommendations, and output recommended material combinations.

[0024] Analyze and mine the historical records of material use by installation and maintenance personnel to extract the patterns of material use to assist in generating accurate material recommendations.

[0025] Furthermore, the day's operation task data and external factors are obtained, and combined with the recommended material combination, the types and quantities of materials that the installation and maintenance personnel need to collect on the day are predicted. The specific method is as follows:

[0026] Collect the day's task data: including task type, user needs, operation location, and equipment installation scale;

[0027] Obtain the user demand information of the day in real time through external systems;

[0028] Use regression analysis or machine learning models to combine historical data, current demand, and device types to make predictions:

[0029] D f =αD history +βD current +γD external

[0030] Among them, D f is the forecast demand for the day; D history Indicates historical usage data for a period of time; D current Indicates the task demand data for the day; D external Represents the impact of external factors on demand; α, β, γ are weight parameters;

[0031] Through cluster analysis, similar task requirements are classified;

[0032] Select the corresponding material combination according to the task type;

[0033] Predict the types and quantities of materials that installation and maintenance personnel will need to collect on the day.

[0034] Based on current mission requirements, historical data and external factors, predict the types and quantities of materials to be collected for subsequent use.

[0035] Furthermore, the inventory status of the warehouse and personal library is monitored in real time to ensure that the recommended materials do not exceed the inventory limit, and the recommendations are adjusted dynamically based on the inventory data. The specific steps are as follows:

[0036] Through RFID tags, barcode scanning or IoT sensors, real-time inventory data of materials in the warehouse is collected, including the total inventory of the warehouse, the quantity of each material, and its location;

[0037] Set inventory thresholds. When the inventory of a certain material is lower than the set value, an alert will be automatically issued, and when recommending materials, avoid recommending low-stock materials.

[0038] If the inventory is insufficient, provide emergency transfer suggestions or recommend alternative materials;

[0039] Update inventory data in real time and automatically record each material collection and return information;

[0040] Based on the forecast of the types and quantities of materials that the installation and maintenance personnel will need on the day, they are matched according to the inventory situation to ensure that the recommended materials are within the sufficient inventory range.

[0041] Monitor the inventory status of warehouses and personal libraries in real time to ensure that recommended materials do not exceed inventory limits, and dynamically adjust recommendations based on inventory data.

[0042] Furthermore, based on historical collection data, current demand forecast and inventory data, a recommended material list is generated. The specific steps are as follows:

[0043] The intelligent recommendation engine combines historical collection data, demand forecasts, inventory data, and accessory configuration requirements to evaluate the recommendation priority of each material through weighted scoring:

[0044] Using a weighted scoring algorithm:

[0045]

[0046] Among them, R i is the recommendation score of material i, ω j is the weight of data source j, X ij is the score of material i corresponding to data source j; n is the number of data sources;

[0047] According to the score R of each material i Sort the materials and recommend the materials with high scores first;

[0048] After sorting by score, a recommended material list is generated, showing the recommended materials, quantity, inventory status, and accessory configuration information;

[0049] If inventory is insufficient or demand changes, dynamically adjust recommended materials to ensure that the recommended materials meet actual operational needs.

[0050] Based on historical collection data, current demand forecasts and inventory data, a personalized list of recommended materials is generated to support installation and maintenance personnel in collecting materials accurately and efficiently.

[0051] The present invention also provides a multi-dimensional fusion recommendation and selection system for material withdrawal of intelligent warehouse for material management based on the method of the present invention, comprising a historical withdrawal data analysis module, a daily demand forecasting module, an inventory data analysis module, an intelligent recommendation engine, and a visualization interface and interaction module connected in sequence;

[0052] The historical requisition data analysis module is used to analyze the historical requisition records of installation and maintenance personnel and extract the rules of material use;

[0053] The daily demand forecasting module predicts the types and quantities of materials that the installation and maintenance personnel may need to collect on the same day based on the current task requirements, historical data and external factors;

[0054] The inventory data analysis module monitors the inventory status of warehouses and personal libraries in real time to ensure that the recommended materials do not exceed the inventory limit and dynamically adjusts the recommendations based on the inventory data;

[0055] The intelligent recommendation engine generates a recommended list of materials based on historical collection data, current demand forecasts and inventory data;

[0056] The visualization interface and interactive module are used to display the materials recommended for the day, including material type, quantity, inventory status and accessory configuration information, and output and display prompt signals when the inventory is insufficient or the material combination is inappropriate, automatically record the material collection status and generate material collection documents.

[0057] This system uses the cooperation of various modules to realize intelligent collection and provides a display interface so that recommended materials can be quickly viewed and confirmed. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a flow chart of the multi-dimensional fusion recommendation and selection method for material delivery in the material management intelligent warehouse of the present invention. DETAILED DESCRIPTION

[0059] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0060] In the description of the present invention, it is necessary to understand that the terms "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0061] In the description of the present invention, unless otherwise specified and limited, it should be noted that the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a mechanical connection or an electrical connection, or it can be the internal connection between two components. It can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to the specific circumstances.

[0062] The present invention discloses a multi-dimensional fusion recommendation and selection method for material delivery in a material management intelligent warehouse, such as Figure 1 As shown, the following steps are included:

[0063] Collect the data of each collection record within the past set time (such as the past three months or one year) (including material type; material model and quantity; collection frequency and matching (collection of materials collected)), and pre-process it to obtain the recommended combination of materials; store all collection data in the database and update it regularly.

[0064] Obtain the day's work task data and external factors, and combine them with the recommended material combination to predict the types and quantities of materials that installation and maintenance personnel need to collect on the day;

[0065] Monitor the inventory status of warehouses and personal libraries in real time to ensure that recommended materials do not exceed inventory limits, and dynamically adjust recommendations based on inventory data;

[0066] Generate a recommended list of materials based on historical requisition data, current demand forecast and inventory data.

[0067] In a preferred embodiment of the present invention, the method for preprocessing the data of each use record within the past set time is:

[0068] De-duplicate, complete and standardize the collected data to ensure data quality, and handle missing values ​​(such as missing material models or quantities) and fill them in;

[0069] Use cluster analysis (count the material collection records for the past month, and then convert the data into a structure suitable for cluster analysis, similar to the data structure of a table, where each row represents a collection, each column represents a material, and the value of each cell represents the number or frequency of occurrence; then use K-Means clustering and use the scikit-learn library to implement clustering. The elbow method helps determine the K value by calculating the total sum of squared errors (SSE) under different numbers of clusters, and then perform cluster analysis to calculate the clustering results, the cluster id to which each transaction belongs, and then calculate the mean of each material in each cluster, and finally output the material combination or matching rules under each cluster) to identify commonly used material combinations and matching rules; for example, if a certain material (such as a router) is often collected together with another material (such as a network cable), the system will mark the combination as "commonly matched"; in cluster 1, the combination of materials A and B is usually used in scenario 1; in cluster 2, the combination of materials C and D is used in scenario 2.

[0070] Count the usage frequency of each material and identify high-frequency materials;

[0071] Analyze historical data to identify the material collection patterns of each installation and maintenance personnel in specific scenarios (such as new user installation, replacement of old equipment, corporate customers, household customers, etc.). For example, when installing broadband, a certain installation and maintenance personnel often collects "router + network cable + power adapter", while other installation and maintenance personnel may have different collection patterns. Generate personalized material collection combinations and recommendations, output material recommendation combinations (specific materials are common to installation and maintenance personnel, such as set-top boxes, gateways, IPTV, FTTR, etc.), and support the system to refer to these historical data in subsequent intelligent recommendations.

[0072] In a preferred embodiment of the present invention, the operation task data and external factors of the day are obtained, and the types and quantities of materials that the installation and maintenance personnel need to collect on the day are predicted in combination with the recommended combination of materials. The specific method is as follows:

[0073] Collect the day's task data: including task type, user needs, operation location, and equipment installation scale;

[0074] Obtain the user demand information of the day in real time through external systems (such as customer service platform);

[0075] Use regression analysis or machine learning models (linear regression, support vector machine) to combine historical data, current demand and device types to make predictions:

[0076] D f =αD history +βD current +γD external

[0077] Among them, D f is the forecast demand for the day; D history Indicates historical usage data for a period of time; D current Indicates the task demand data for the day; D external Indicates the impact of external factors (such as weather, holidays, etc.) on demand; α, β, γ are weight parameters, which can be optimized based on historical data and demand characteristics;

[0078] Through cluster analysis, similar task requirements are classified;

[0079] Select the corresponding material combination according to the task type (such as broadband installation, equipment maintenance, etc.);

[0080] Predict the types and quantities of materials that installation and maintenance personnel will need to collect on the day.

[0081] In a preferred embodiment of the present invention, the inventory status of the warehouse and the personal library is monitored in real time to ensure that the recommended materials do not exceed the inventory limit, and the recommendation is dynamically adjusted according to the inventory data. The specific steps are:

[0082] Through RFID tags, barcode scanning or IoT sensors, real-time inventory data of materials in the warehouse is collected, including the total inventory of the warehouse, the quantity of each material, and its location;

[0083] Set inventory thresholds. When the inventory of a certain material is lower than the set value, an alert will be automatically issued, and when recommending materials, avoid recommending low-stock materials.

[0084] If the inventory is insufficient, provide emergency transfer suggestions or recommend alternative materials (if the inventory is insufficient, no recommendation will be made and the insufficient inventory will be directly prompted);

[0085] Update inventory data in real time, automatically record each material collection and return information, and ensure the accuracy of inventory data;

[0086] Based on the forecast of the types and quantities of materials that the installation and maintenance personnel will need on the day, they are matched according to the inventory situation to ensure that the recommended materials are within the sufficient inventory range.

[0087] In a preferred embodiment of the present invention, a recommended material list is generated based on historical requisition data, current day demand forecast and inventory data. The specific steps are as follows:

[0088] Using the intelligent recommendation engine, combined with historical collection data, demand forecasts, inventory data, and accessory configuration requirements (such as accessories that may be needed for network installation, such as power cords, screwdrivers, etc.), the recommendation priority of each material is evaluated through weighted scoring:

[0089] Using a weighted scoring algorithm:

[0090]

[0091] Among them, R i is the recommendation score of material i, ω j is the weight of data source j, X ij is the score of material i corresponding to data source j; n is the number of data sources (data sources generally refer to data in different time periods. For example, data from 24 years and 25 years can be used as two different data sources. Data sources of different branches can also be processed. Each branch has multiple sub-bureaus, and the usage data of each sub-bureau can be used as a data source);

[0092] According to the score R of each material i Sort the materials and recommend the materials with high scores first;

[0093] After sorting by score, a recommended material list is generated, showing the recommended materials, quantity, inventory status, and accessory configuration information;

[0094] If the inventory is insufficient or the demand changes, the recommended materials will be adjusted dynamically (a prompt will be given if the inventory is insufficient, and priority will be given to ensuring that the recommended materials meet the actual operational needs) to ensure that the recommended materials meet the actual operational needs.

[0095] The present invention realizes intelligent material recommendation and improves the accuracy of material collection through multi-dimensional integration of historical collection data, daily demand forecast and inventory status. In the process of material recommendation, the recommendation results are dynamically adjusted, and the recommendation list is automatically generated to reduce the time of manual selection and ensure matching of real-time demand. The recommendation results are adjusted in real time to avoid insufficient inventory or wrong recommendations, reduce the manual dependence of installation and maintenance personnel, reduce material waste, reduce inventory pressure, and realize efficient and accurate material collection.

[0096] The present invention also provides a multi-dimensional fusion recommendation and selection system for material withdrawal of intelligent warehouse for material management based on the method described in the present invention, comprising a historical withdrawal data analysis module, a daily demand forecasting module, an inventory data analysis module, an intelligent recommendation engine and a visualization interface and interaction module connected in sequence.

[0097] The historical collection data analysis module is used to analyze the historical collection records of installation and maintenance personnel, extract the rules of material use, and assist in generating accurate material recommendations. The daily demand forecasting module predicts the types and quantities of materials that installation and maintenance personnel may need to collect on the same day based on current task requirements, historical data, and external factors.

[0098] The inventory data analysis module monitors the inventory status of warehouses and personal libraries in real time to ensure that the recommended materials do not exceed the inventory limit, and dynamically adjusts the recommendations based on the inventory data. The intelligent recommendation engine generates a list of recommended materials based on historical collection data, daily demand forecasts and inventory data, supporting installation and maintenance personnel to collect materials accurately and efficiently.

[0099] The visual interface and interactive module are used to display the recommended materials for the day, including material type, quantity, inventory status, and accessory configuration information, etc. When the inventory is insufficient or the material combination is inappropriate, a prompt signal is output and displayed, and the material collection status is automatically recorded and a material collection document is generated. A "Confirm Collection" button is provided on the visual interface and interactive module, and the installation and maintenance personnel can click to confirm the collection of materials. The material adjustment function is provided, allowing the installation and maintenance personnel to manually modify the recommended list according to needs (for example, adjust the quantity, replace materials, etc.).

[0100] The system needs to complete integrated recommendations based on data such as daily user needs, historical collection of front-line installation and maintenance, warehouse inventory quantities, personal warehouse inventory quantities, daily forecast usage, material combinations, accessory configurations, etc., to reduce manual dependence and achieve intelligent collection.

[0101] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0102] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

Claims

1. A multi-dimensional fusion recommendation and selection method for material delivery in a material management intelligent warehouse, characterized in that: The steps include: Collect the data of each collection record within the past set time, and pre-process it to obtain the recommended combination of materials; Obtain the day's work task data and external factors, and combine them with the recommended material combination to predict the types and quantities of materials that installation and maintenance personnel need to collect on the day; Monitor the inventory status of warehouses and personal libraries in real time to ensure that recommended materials do not exceed inventory limits, and dynamically adjust recommendations based on inventory data; Generate a recommended list of materials based on historical requisition data, current demand forecast and inventory data.

2. The multi-dimensional fusion recommendation and selection method for material withdrawal in a material management intelligent warehouse as claimed in claim 1 is characterized in that: The method for preprocessing the data of each collection record within the past set time is: Remove duplication, complete and standardize the collected data, and fill in missing values; Use cluster analysis to identify common material combinations and matching patterns; Count the usage frequency of each material and identify high-frequency materials; Identify the material collection patterns of each installation and maintenance personnel in specific scenarios in historical data, generate personalized material collection combinations and recommendations, and output recommended material combinations.

3. The multi-dimensional fusion recommendation and selection method for material withdrawal in a material management intelligent warehouse as claimed in claim 1 is characterized in that: Obtain the day's task data and external factors, and combine them with the recommended material combination to predict the types and quantities of materials that the installation and maintenance personnel need to collect on the day. The specific method is as follows: Collect the day's task data: including task type, user needs, operation location, and equipment installation scale; Obtain the user demand information of the day in real time through external systems; Use regression analysis or machine learning models to combine historical data, current demand, and device types to make predictions: D f =αD history +βD current +γD external Among them, D f is the forecast demand for the day; D history Indicates historical usage data for a period of time; D current Indicates the task demand data for the day; D external Represents the impact of external factors on demand; α, β, γ are weight parameters; Through cluster analysis, similar task requirements are classified; Select the corresponding material combination according to the task type; Predict the types and quantities of materials that installation and maintenance personnel will need to collect on the day.

4. The multi-dimensional fusion recommendation and selection method for material withdrawal in a material management intelligent warehouse as claimed in claim 1 is characterized in that: Monitor the inventory status of warehouses and personal libraries in real time to ensure that the recommended materials do not exceed the inventory limit, and dynamically adjust the recommendations based on inventory data. The specific steps are as follows: Through RFID tags, barcode scanning or IoT sensors, real-time inventory data of materials in the warehouse is collected, including the total inventory of the warehouse, the quantity of each material, and its location; Set inventory thresholds. When the inventory of a certain material is lower than the set value, an alert will be automatically issued, and when recommending materials, avoid recommending low-stock materials. If the inventory is insufficient, provide emergency transfer suggestions or recommend alternative materials; Update inventory data in real time and automatically record each material collection and return information; Based on the forecast of the types and quantities of materials that the installation and maintenance personnel will need on the day, they are matched according to the inventory situation to ensure that the recommended materials are within the sufficient inventory range.

5. The multi-dimensional fusion recommendation and selection method for material withdrawal in a material management intelligent warehouse as claimed in claim 1 is characterized in that: Generate a recommended material list based on historical requisition data, current demand forecast, and inventory data. The specific steps are as follows: The intelligent recommendation engine combines historical collection data, demand forecasts, inventory data, and accessory configuration requirements to evaluate the recommendation priority of each material through weighted scoring: Using a weighted scoring algorithm: Among them, R i is the recommendation score of material i, ω j is the weight of data source j, X ij is the score of material i corresponding to data source j; n is the number of data sources; According to the score R of each material i Sort the materials and recommend the materials with high scores first; After sorting by score, a recommended material list is generated, showing the recommended materials, quantity, inventory status, and accessory configuration information; If inventory is insufficient or demand changes, dynamically adjust recommended materials to ensure that the recommended materials meet actual operational needs.

6. A material management intelligent warehouse material delivery multi-dimensional fusion recommendation and selection system based on the method of any one of claims 1-5, characterized in that: It includes a historical use data analysis module, a daily demand forecasting module, an inventory data analysis module, an intelligent recommendation engine, and a visualization interface and interaction module that are connected in sequence; The historical requisition data analysis module is used to analyze the historical requisition records of installation and maintenance personnel and extract the rules of material use; The daily demand forecasting module predicts the types and quantities of materials that the installation and maintenance personnel may need to collect on the same day based on the current task requirements, historical data and external factors; The inventory data analysis module monitors the inventory status of warehouses and personal libraries in real time to ensure that the recommended materials do not exceed the inventory limit and dynamically adjusts the recommendations based on the inventory data; The intelligent recommendation engine generates a recommended list of materials based on historical collection data, current demand forecasts and inventory data; The visualization interface and interactive module are used to display the materials recommended for the day, including material type, quantity, inventory status and accessory configuration information, and output and display prompt signals when the inventory is insufficient or the material combination is inappropriate, automatically record the material collection status and generate material collection documents.

Images