Generate spatial and geometric models using machine learning systems with multi-platform interfaces.

Abstract

The computer-implemented invention receives building information (705) including geometric and budgetary data, and uses machine learning (770) to generate an architectural space model (735) including the building's floor plan and furniture arrangement (790). The generated design is scored, ranked (745), and sent to the user (750).

Description

[0001] Cross-references to related applications

[0002] This application claims the benefit and priority of the following applications: U.S. Provisional Patent Application No. 63 / 041,535, filed June 19, 2020, entitled "Generating Space Models and Geometry Models Using a Machine Learning System with Multi-Platform Interfaces"; U.S. Patent Application No. 17 / 350,538, filed June 17, 2021, entitled "Generating Space Models and Geometry Models Using a Machine Learning System with Multi-Platform Interfaces"; and U.S. Patent Application No. 63 / 041,535, filed June 17, 2021, entitled "Generating Space Models and Geometry Models Using a Machine Learning System with Multi-Platform Interfaces". U.S. Patent Application No. 17 / 350,582, entitled “Multi-Platform Interfaces”; and U.S. Patent Application No. 17 / 350,641, filed June 17, 2021, entitled “Generating Space Models and Geometry Models Using a Machine Learning System with Multi-Platform Interfaces”. Each of the foregoing applications is incorporated herein by reference in its entirety. Technical Field

[0003] Various aspects of this disclosure relate to digital data processing systems, data processing methods, and machine learning systems. Specifically, one or more aspects of this disclosure relate to digital data processing systems that use machine learning components to generate spatial and geometric models, and these digital data processing systems include multi-platform interfaces to achieve interoperability. Background Technology

[0004] In some cases, office floor plans and other space models or configurations may be created, updated, and / or otherwise modified when new spaces are created, occupancy changes occur, and / or tastes or other preferences change. In many cases, creating, updating, and / or otherwise modifying space configurations may require a manual and labor-intensive process involving selecting design details from a large number of design options. While attempts have been made to automate this labor-intensive process using computer systems to automatically generate floor plans, these traditional systems have largely failed to produce usable results because, among other reasons, a large number of variables need to be considered simultaneously when creating a working space configuration, and there are many different ways to document floor plans or layouts, each of which may be desirable and / or necessary in a given situation. These traditional systems also implement inefficient software and hardware, resulting in delayed processing times, increased processing loads, and other technical challenges. Summary of the Invention

[0005] Various aspects of this disclosure provide technical solutions to overcome one or more of the aforementioned technical problems and / or other technical challenges. For example, one or more aspects of this disclosure relate to using machine learning techniques in conjunction with generative design algorithms to create and output spatial models and provide other functionalities.

[0006] According to one or more embodiments, a computing platform having at least one processor, a communication interface, and a memory can receive first spatial program data identifying one or more parameters of a first physical space from a first user computing device via the communication interface. The computing platform can load a first geometric model from a database storing one or more geometric models, the first geometric model including information defining a first plurality of design rules. The computing platform can generate a first plurality of spatial models for the first physical space based on the first spatial program data identifying one or more parameters of the first physical space and the first geometric model. Based on the first geometric model, the computing platform can score the first plurality of spatial models generated for the first physical space, which can generate a score for each of the first plurality of spatial models. The computing platform can sort the first plurality of spatial models generated for the first physical space based on the score of each of the first plurality of spatial models, which can generate a sorted list of spatial models. The computing platform can generate user interface data including the sorted list of spatial models. The computing platform can send the user interface data including the sorted list of spatial models to the first user computing device via the communication interface, which can cause the first user computing device to display a user interface including at least a portion of the sorted list of spatial models.

[0007] In some embodiments, the computing platform may receive first spatial program data identifying one or more parameters of the first physical space by receiving information identifying architectural details, organizational details, work style details, and budget details of the first physical space. In some embodiments, the computing platform may load a first geometric model from a database storing one or more geometric models by selecting a first geometric model from multiple geometric models generated by the computing platform using a machine learning engine trained on one or more best-in-class spatial designs.

[0008] In some embodiments, the computing platform can load a first geometric model from a database storing one or more geometric models by selecting a first geometric model based on first spatial program data identifying one or more parameters of the first physical space. In some embodiments, the computing platform can generate a first plurality of spatial models for a first physical space based on the first spatial program data identifying one or more parameters of the first physical space and the first geometric model in the following ways: 1) generating a plurality of block models for the first physical space; 2) scoring the plurality of block models generated for the first physical space based on the first geometric model, which can generate a score for each of the plurality of block models; 3) selecting a subset of the plurality of block models based on the score of each of the plurality of block models; 4) generating a plurality of setting models for the first physical space, each setting model potentially corresponding to a specific block model within the subset of the plurality of block models; 5) scoring the plurality of setting models generated for the first physical space based on the first geometric model. The model is scored, which can generate a score for each of the multiple setup models; 6) a subset of the multiple setup models is selected based on the scores of each of the multiple setup models; 7) multiple furniture models are generated for the first physical space, wherein each of the multiple furniture models corresponds to a specific setup model in the subset of the multiple setup models; 8) the multiple furniture models generated for the first physical space are scored based on the first geometric model, which can generate a score for each of the multiple furniture models; and 9) a subset of the multiple furniture models is selected based on the scores of each of the multiple furniture models, wherein the subset of the multiple furniture models corresponds to a first plurality of space models generated for the first physical space.

[0009] In some embodiments, each of the plurality of block models may indicate the potential location of different neighborhoods in the first physical space, each of the plurality of setting models may indicate the potential location of different work settings in the first physical space, and each of the plurality of furniture models may indicate the potential location of different furniture items in the first physical space. In some embodiments, the score of each of the first plurality of spatial models may indicate the level of conformity with one or more metrics defined by the first geometric model.

[0010] In some embodiments, sending user interface data including a sorted list of spatial models to a first user computing device may cause the first user computing device to display one or more of the scores determined for each of a first plurality of spatial models. In some embodiments, a computing platform may receive data from the first user computing device via a communication interface instructing the selection of a first spatial model from the sorted list of spatial models. In response to receiving the data instructing the selection of a first spatial model from the sorted list of spatial models, the computing platform may generate a visual rendering of the first spatial model. The computing platform may send the visual rendering of the first spatial model to the first user computing device via the communication interface, which may cause the first user computing device to display a user interface including at least a portion of the visual rendering of the first spatial model.

[0011] In some embodiments, the computing platform may receive data indicating user modifications to the first spatial model from a first user computing device via a communication interface. Based on the received data indicating user modifications to the first spatial model, the computing platform may update the machine learning engine running on the computing platform.

[0012] In some embodiments, the computing platform may receive data from a first user computing device via a communication interface, indicating a request to export a first spatial model to a design tool. In response to receiving the data indicating the request to export the first spatial model to the design tool, the computing platform may generate one or more drawing files based on the first spatial model. The computing platform may then send the one or more drawing files generated based on the first spatial model to the first user computing device via the communication interface.

[0013] In some embodiments, the computing platform may receive second spatial program data identifying one or more parameters of a second physical space from a second user computing device via a communication interface. The computing platform may load a second geometric model from a database storing one or more geometric models, the second geometric model including information defining a second plurality of design rules. The computing platform may generate a second plurality of spatial models for the second physical space based on the second spatial program data identifying one or more parameters of the second physical space and the second geometric model. Based on the second geometric model, the computing platform may score the second plurality of spatial models generated for the second physical space, which may generate a score for each of the second plurality of spatial models. The computing platform may sort the second plurality of spatial models generated for the second physical space based on the score of each of the second plurality of spatial models, which may generate a second sorted list of spatial models. The computing platform may generate second user interface data including the second sorted list of spatial models. The computing platform may send the second user interface data including the second sorted list of spatial models to the second user computing device via a communication interface, which may cause the second user computing device to display a user interface including at least a portion of the second sorted list of spatial models.

[0014] According to one or more additional embodiments, a computing platform having at least one processor, a communication interface, and a memory can receive multiple drawing models corresponding to different spatial designs from a data server via the communication interface. The computing platform can identify multiple design parameters associated with each of the multiple drawing models corresponding to different spatial designs. The computing platform can train a machine learning engine based on the multiple drawing models corresponding to different spatial designs and the multiple design parameters associated with each of the multiple drawing models corresponding to different spatial designs, which can generate at least one geometric model corresponding to the multiple drawing models. The computing platform can store at least one geometric model corresponding to the multiple drawing models in a database storing one or more additional geometric models.

[0015] In some embodiments, when receiving multiple drawing models corresponding to different spatial designs, the computing platform may receive at least one two-dimensional computer-aided design (CAD) model or PDF drawing. In some embodiments, the computing platform may identify multiple design parameters associated with each of the multiple drawing models corresponding to different spatial designs by identifying multiple design features, which may include one or more of the following: total square feet, total number of offices, total number of meeting spaces, total number of community spaces, number of seats per office, number of seats per meeting space, number of seats per community space, percentage of total square feet allocated to offices, percentage of total square feet allocated to meeting spaces, percentage of total square feet allocated to community spaces, average office size, or average meeting space size.

[0016] In some embodiments, the computing platform may identify multiple design parameters associated with each of the multiple drawing models corresponding to different spatial designs by applying cognitive machine learning to select multiple design features based on the organization corresponding to each of the multiple drawing models before identifying multiple design parameters. In some embodiments, the computing platform may select multiple design features based on one or more of the organization's industry, geographic data, size, or personality.

[0017] In some embodiments, the computing platform may select multiple design features by selecting multiple design features based on user input, and the multiple design features may be consistent for each of the multiple drawing models. In some embodiments, the computing platform may generate at least one geometric model by identifying one or more design rules that can be applied to score the conformity of at least one spatial model with the multiple drawing models, wherein the one or more design rules include one or more data ranges or numerical constraints.

[0018] In some embodiments, the computing platform may receive spatial program data identifying one or more parameters of a physical space from a user computing device via a communication interface. The computing platform may load at least one geometric model from a database storing one or more additional geometric models. The computing platform may generate multiple spatial models for the physical space based on the spatial program data identifying one or more parameters of the physical space and the at least one geometric model. The computing platform may score the multiple spatial models generated for the physical space based on the at least one geometric model, which may generate a score for each of the multiple spatial models. The computing platform may then sort the multiple spatial models generated for the physical space based on the scores of each of the multiple spatial models, which may generate a sorted list of spatial models. The computing platform may generate user interface data including the sorted list of spatial models. The computing platform may then send the user interface data including the sorted list of spatial models to the user computing device via the communication interface, which may enable the user computing device to display a user interface including at least a portion of the sorted list of spatial models.

[0019] In some embodiments, the computing platform can generate multiple spatial models for a physical space based on spatial program data identifying one or more parameters of the physical space and at least one geometric model in the following ways: 1) generating multiple block models for the physical space; 2) scoring the multiple block models generated for the physical space based on at least one geometric model, which can generate a score for each of the multiple block models; 3) selecting a subset of the multiple block models based on the scores of each of the multiple block models; 4) generating multiple setup models for the physical space, wherein each setup model corresponds to a specific block model in the subset of the multiple block models; 5) scoring the multiple block models generated for the physical space based on at least one geometric model. 6) Set up models for scoring, which can generate a score for each of the multiple set models; 7) Select a subset of multiple set models based on the scores of each of the multiple set models; 8) Generate multiple furniture models for the physical space, wherein each of the multiple furniture models corresponds to a specific set model in the subset of the multiple set models; 9) Set up multiple furniture models generated for the physical space based on at least one geometric model, which can generate a score for each of the multiple furniture models; and 10) Select a subset of multiple furniture models based on the scores of each of the multiple furniture models, wherein the subset of multiple furniture models corresponds to multiple space models generated for the physical space.

[0020] In some embodiments, each of the multiple block models can indicate the potential location of different neighborhoods in the physical space, each of the multiple setting models can indicate the potential location of different work settings in the physical space, and each of the multiple furniture models can indicate the potential location of different furniture items in the physical space.

[0021] According to one or more additional embodiments, a computing platform having at least one processor, a communication interface, and a memory can receive data from a first computing device via the communication interface, instructing a request to export a spatial model to a first design tool, and the spatial model can be defined in multiple data formats. In response to receiving the data instructing the export of the spatial model to the first design tool, the computing platform can generate one or more first drawing files based on the spatial model by: 1) selecting a first data format from multiple data formats based on the first design tool; 2) extracting first format-specific data from the spatial model, wherein the first format-specific data is defined in the first data format; and 3) generating one or more first drawing files using the first format-specific data extracted from the spatial model, wherein the one or more first drawing files are generated according to the first data format. The computing platform can send the one or more first drawing files generated based on the spatial model to the first computing device via the communication interface.

[0022] In some embodiments, the computing platform may receive user input defining spatial information corresponding to one or more elements from a first computing device, or the computing platform may automatically generate spatial information corresponding to one or more elements using cognitive machine learning based on best-in-class floor plans. In some embodiments, the computing platform may generate a spatial model based on the spatial information corresponding to one or more elements before receiving data instructing the export of the spatial model to a first design tool.

[0023] In some embodiments, the one or more elements can be one or more of the following: blocks, settings, or furniture items, wherein a block can be an office department, a setting can be a room type, and a furniture item can be a single piece of furniture. In some embodiments, the computing platform can send one or more commands to instruct a client computing device to display a graphical user interface including optional furniture purchase elements, which can cause the client computing device to display a graphical user interface including optional furniture purchase elements. Subsequently, the computing platform can receive furniture selection information indicating an order for one or more furniture items. The computing platform can then process the order for one or more furniture items.

[0024] In some cases, multiple data formats may include one or more of the following: Computer-Aided Design (CAD), CET, Revit, or SketchUp. In some cases, the computing platform may receive data from a second computing device via a communication interface instructing the export of a spatial model to a second design tool. In response to receiving the data instructing the export of a spatial model to a second design tool, the computing platform may generate one or more second drawing files based on the spatial model by: 1) selecting a second data format from a plurality of data formats based on the second design tool; 2) extracting second format-specific data from the spatial model, wherein the second format-specific data is defined by the second data format; and 3) generating one or more second drawing files using the second format-specific data from the spatial model, wherein the one or more second drawing files are generated according to the second data format. The computing platform may send one or more second drawing files generated based on the spatial model to the second computing device via a communication interface.

[0025] In some embodiments, the computing platform may generate a spatial model by: 1) receiving spatial program data identifying one or more parameters of a physical space from a first computing device via a communication interface; 2) loading a geometric model from a database storing one or more geometric models, wherein the geometric model contains information defining a plurality of design rules; 3) generating a plurality of block models for the physical space; 4) scoring the plurality of block models generated for the physical space based on the geometric models, which may generate a score for each of the plurality of block models; 5) selecting a subset of the plurality of block models based on the scores of each of the plurality of block models; and 6) generating a plurality of setup models for the physical space, wherein each of the plurality of setup models corresponds to a subset of the plurality of block models. 7) Scoring multiple setup models generated for the physical space based on the geometric model, which can generate a score for each of the multiple setup models; 8) Selecting a subset of the multiple setup models based on the score of each of the multiple setup models; 9) Generating multiple furniture models for the physical space, wherein each of the multiple furniture models corresponds to a specific setup model in the subset of the multiple setup models; 10) Scoring multiple furniture models generated for the physical space based on the geometric model, which can generate a score for each of the multiple furniture models; and 11) Selecting a subset of the multiple furniture models based on the score of each of the multiple furniture models, wherein the subset of the multiple furniture models includes a space model.

[0026] In some embodiments, each of the multiple block models can indicate the potential location of different neighborhoods in the physical space, each of the multiple setting models can indicate the potential location of different work settings in the physical space, and each of the multiple furniture models can indicate the potential location of different furniture items in the physical space. In some embodiments, generating a spatial model can include generating the spatial model in each of a multiple data formats. Attached Figure Description

[0027] This disclosure is illustrated by way of example and is not limited to the accompanying drawings, in which similar reference numerals indicate similar elements, and wherein:

[0028] Figure 1A and Figure 1B An illustrative operating environment for generating spatial and geometric models using a machine learning system with a multi-platform interface is described according to one or more example embodiments;

[0029] Figure 1C , Figure 1D , Figure 1E , Figure 1F and Figure 1G Illustrative data structures for various models that can be generated, stored and / or otherwise used according to one or more example embodiments are described;

[0030] Figures 2A to 2H An illustrative sequence of events is described, according to one or more example embodiments, for generating spatial and geometric models using a machine learning system with a multi-platform interface;

[0031] Figures 3 to 6 An illustrative user interface for generating spatial and geometric models using a machine learning system with a multi-platform interface, according to one or more example embodiments, is described.

[0032] Figure 7 An illustrative method for generating spatial and geometric models using a machine learning system with a multi-platform interface is described according to one or more example embodiments;

[0033] Figure 8 Additional illustrative user interfaces for generating spatial and geometric models using a machine learning system with a multi-platform interface, according to one or more example embodiments, are described.

[0034] Figures 9A to 9B An illustrative sequence of events for providing a workplace configuration interface is described according to one or more example embodiments; and

[0035] Figure 10An illustrative workplace configuration interface is depicted according to one or more example embodiments. Detailed Implementation

[0036] In the following description of various illustrative embodiments, reference is made to the accompanying drawings, which form part of this document, and illustrate various embodiments in which aspects of this disclosure can be practiced. It should be understood that other embodiments can be utilized, and structural and functional modifications can be made, without departing from the scope of this disclosure. Various connections between elements are discussed in the following description. It should be noted that these connections are general and, unless otherwise specified, can be direct or indirect, wired or wireless, and this specification is not intended to limit this aspect.

[0037] Some aspects of this disclosure relate to using a machine learning system with multi-platform interfaces to generate spatial models (which may also be referred to as spatial planning, test fitting, and / or floor planning) and geometric models (which may also be referred to as circulation networks or circulation paths). For example, the computing platform may receive spatial program data, which in some cases may identify one or more parameters of a physical space. The computing platform may load geometric models from a database storing one or more geometric models. In some cases, the geometric model may define multiple design rules. Additionally or alternatively, the geometric model may define rules for dividing floor slabs, placing circulation paths, and / or placing furniture arrangements. The computing platform may generate multiple spatial models (e.g., floor planning, test fitting, or other models for recording and / or otherwise specifying how the space and / or its contents are configured) for a physical space based on the spatial program data identifying one or more parameters of the physical space and the geometric models. Based on the geometric models, the computing platform may score the multiple spatial models generated for the physical space, which may generate a score for each of the multiple spatial models. The computing platform may sort the multiple spatial models generated for the physical space based on the scores of each of the multiple spatial models, which may produce a sorted list of spatial models. The computing platform can generate user interface data including a sorted list of spatial models and can send the user interface data including the sorted list of spatial models to a user computing device via a communication interface, which can enable the user computing device to display a user interface including at least a portion of the sorted list of spatial models.

[0038] In doing so, the computing platform can automatically generate a target series of spatial models with very little (if any) user input. Furthermore, by implementing generative design algorithms that use a hierarchical approach to generate spatial models (which in some cases may include generating block models at different design stages (e.g., which can be used to locate departments, rooms, spaces, and / or other areas within a floor space) and scoring and setting models (e.g., which can be used to create configurations of rooms or spaces)), the computing platform can reduce processing time and computational bandwidth. For example, by solving only the settings of a given physical space when the blocks have already been solved, determined, and / or otherwise defined relative to the physical space, and by solving only the furniture of the physical space when the settings have already been solved, determined, and / or otherwise defined relative to the physical space, the computing platform can generate a relatively small number of spatial models optimized and / or suited to desired and / or undesired preferred parameters, compared to the processing required to simultaneously generate models for blocks, settings, and furniture. Therefore, in at least some cases, and as an example, furniture can be solved for only a subset of the setup, and this subset can itself be selected from a subset of blocks, rather than solving for furniture for all blocks. This hierarchical generative design algorithm offers several technical advantages, including reduced processing load and reduced consumption of network bandwidth and other computing resources. Furthermore, and in some arrangements described in more detail below, a computing platform implementing some aspects of this disclosure can generate spatial models in multiple data formats. When implementing this multi-format approach, the computing platform can generate output elements only once at the start of the modeling process, rather than at the end of the process in response to a request for an associated data file of the spatial model or an alternative format. When implemented, this multi-format approach for spatial model generation can provide additional technical advantages, including reduced processing load and increased processing efficiency, as well as enhanced interoperability.

[0039] Figure 1A and Figure 1B An illustrative operating environment is described, according to one or more example embodiments, for generating spatial and geometric models using a machine learning system with a multi-platform interface. References Figure 1AThe computing environment 100 may include various computer systems, computing devices, networks, and / or other operational infrastructure. For example, the computing environment 100 may include a generative design computing platform 110, an internal data server 120, an external data server 130, a first designer user computing device 140, a second designer user computing device 150, a client user computing device 160, and a network 170. It should be noted that the computing environment 100 is exemplary, and in some cases, the generative design computing environment may include more or fewer computer systems, computing devices, networks, and / or other operational interfaces, or the computing functions may be combined or distributed across fewer or more devices, while still operating according to the methods and principles disclosed herein.

[0040] Network 170 may include one or more wired networks and / or one or more wireless networks interconnecting generative design computing platform 110, internal data server 120, external data server 130, first designer user computing device 140, second designer user computing device 150, client user computing device 160, and / or other computer systems and / or devices. Furthermore, each of the generative design computing platform 110, internal data server 120, external data server 130, first designer user computing device 140, second designer user computing device 150, and client user computing device 160 may be a dedicated computing device configured to perform a specific function, as illustrated in more detail below, and may include specific computing components such as processors, memory, and communication interfaces.

[0041] One or more internal data servers, such as internal data server 120, may be configured to host and / or otherwise provide internal block models, setup models, furniture models, and / or other data. For example, internal data server 120 may be maintained or otherwise controlled by an organization (e.g., a furniture company, architecture firm, design firm) that maintains or otherwise controls the generative design calculation platform 110. Furthermore, internal data server 120 may be configured to maintain product information, best-in-class floor plans, geometric models, design rules (e.g., design principles), and / or other design data developed, used, and / or otherwise associated with the organization.

[0042] One or more external data servers, such as external data server 130, may be configured to host and / or otherwise provide external block models, setup models, furniture models, and / or other data. For example, external data server 130 may be maintained or otherwise controlled by a third-party organization (e.g., an alternative furniture company, an alternative architecture company, an alternative design company) that is different from the enterprise organization maintaining or otherwise controlling the generative design calculation platform 110. Furthermore, external data server 130 may be configured to maintain product information, best-in-class floor plans, geometric models, design rules, and / or other design data developed, used, and / or otherwise associated with the third-party organization.

[0043] The first designer user computing device 140 can be configured for use by a first user (which may be, for example, an enterprise user associated with an organization operating the generative design computing platform 110, such as a designer, architect, etc.). In some cases, the first designer user computing device 140 can be configured to present one or more user interfaces generated and / or otherwise associated with a first design tool (e.g., tools associated with computer-aided design (CAD), CET, Revit, SketchUp, etc.), a local browser, and / or one or more other software applications.

[0044] The second designer user computing device 150 can be configured for use by a second user (which may be, for example, an enterprise user associated with an organization operating the generative design computing platform 110, such as a designer, architect, etc., and may be different from the first user of the first designer user computing device 140). In some cases, the second designer user computing device 150 can be configured to present one or more user interfaces generated and / or otherwise associated with a second design tool different from the first design tool (e.g., tools associated with computer-aided design (CAD), CET, Revit, SketchUp, etc.), a local browser, and / or one or more other software applications.

[0045] The client-user computing device 160 can be configured for use by a third user (who may be, for example, a customer or client of an enterprise organization operating the generative design computing platform 110, and who may be different from the first user of the first designer-user computing device 140 and the second user of the second designer-user computing device 150). In some cases, the client-user computing device 160 can be configured to present one or more user interfaces associated with a local browser, which can receive information from, send information to, and / or otherwise exchange information with the generative design computing platform 110 during a browser session. For example, the client-user computing device 160 can be configured to present one or more furniture purchasing interfaces, floor plan viewing interfaces, design viewing interfaces, and / or other user interfaces associated with one or more space models generated by the generative design computing platform 110 and / or other information received from the generative design computing platform 110.

[0046] refer to Figure 1B The generative design computing platform 110 may include one or more processors 111, one or more memories 112, and one or more communication interfaces 113. In some cases, the generative design computing platform 110 may consist of multiple different computing devices distributed across a single data center or multiple different data centers. In these cases, the one or more processors 111, one or more memories 112, and one or more communication interfaces 113 included in the generative design computing platform 110 may be part of and / or otherwise associated with the different computing devices forming the generative design computing platform 110.

[0047] In one or more arrangements, processor 111 can control the operation of generative design computing platform 110. Memory 112 can store instructions that, when executed by processor 111, cause generative design computing platform 110 to perform one or more of the functions described herein. Communication interface 113 may include one or more wired and / or wireless network interfaces, and communication interface 113 can connect generative design computing platform 110 to one or more networks (e.g., network 170) and / or enable generative design computing platform 110 to exchange information and / or otherwise communicate with one or more devices connected to such networks.

[0048] In one or more arrangements, memory 112 may store and / or otherwise provide multiple modules (which may include, for example, instructions that can be executed by processor 111 to cause generative design computing platform 110 to perform various functions), a database (which may, for example, store data used by generative design computing platform 110 in performing various functions), and / or other elements (which may include, for example, processing engines, services, and / or other elements). For example, memory 112 may store and / or otherwise provide generative design module 112a, generative design database 112b, geometric model engine 112c, and machine learning engine 112d. In some cases, generative design module 112a may store instructions that cause generative design computing platform 110 to generate spatial models and / or perform one or more other functions described herein. Additionally, generative design database 112b may store data used by generative design computing platform 110 in generating spatial models and / or performing one or more other functions described herein. The geometric model engine 112c can be used to generate and / or store geometric models that can be used by the generative design module 112a and / or the generative design computing platform 110 during spatial model generation and sorting. The machine learning engine 112d may have instructions that guide and / or enable the generative design computing platform 110 to set, define, and / or iteratively refine optimization rules and / or other parameters used by the generative design computing platform 110 and / or other systems in the computing environment 100.

[0049] Figure 1C , Figure 1D , Figure 1E , Figure 1F and Figure 1G Illustrative data structures for various models that can be generated, stored, and / or otherwise used according to one or more example embodiments are described. References Figure 1CThe example block model 180 is depicted. Block model 180 may include, for example, batch size data 180a, exterior wall data 180b, exterior feature data 180c, interior wall data 180d, interior feature data 180e, neighborhood data 180f (which may include, for example, department information, team information, group information, and / or other information), corridor data 180g (which may include, for example, circulation data about corridors, passageways, walkways, staircases, elevators, and / or other areas used for accessing spaces within the building), and other block data 180h. For example, batch size data 180a may include information defining one or more dimensions and / or features of a batch of land or other plots on which one or more buildings and / or other structures may be located. Exterior wall data 180b may include information defining the location and / or features of one or more exterior walls of such buildings and / or other structures, and exterior feature data 180c may include information defining other exterior features (e.g., windows, views, exterior columns, decorations, etc.) of such buildings and / or other structures. Interior wall data 180d may include information defining the location and / or other features of one or more interior walls within such buildings and / or structures, and interior feature data 180e may include information defining other interior features (e.g., windows, HVAC systems and components, interior columns, lounges, vertical ventilation, mechanical / electrical rooms, storage rooms, etc.). In some cases, these interior and / or exterior walls may be two-dimensional or three-dimensional walls, and they may be dragged, dropped, and / or otherwise modified (e.g., materials may be changed, and / or other modifications may be performed). Neighborhood data 180f may include information defining the location of various organizational departments, office neighborhoods, and / or other groupings within a physical space. Corridor data 180g may include information defining the location of various corridors, walkways, and / or other boundaries within a physical space, and other block data 180h may include information defining other features of specific areas within a physical space. In some cases, and as illustrated in more detail below, some aspects of the block model can be defined based on inputs received by the generative design computing platform 110, such as the size and / or external features of a batch of land or buildings located on such a batch of land, while other aspects of the block model can be determined by the generative design computing platform 110 using one or more processes described herein, such as the location and layout of various neighborhoods, corridors and / or other block model features.

[0050] refer to Figure 1DThe illustration depicts an example setup model 182. Setup model 182 may include, for example, block model 182a, room data 182b, public (shared) space data 182c, and other setup data 182d. Block model 182a may include block models that have been generated and / or stored for a specific physical space (e.g., the same space to which setup model 182 is applied). For example, block model 182 may include block model 180 and / or any of its content data. Room data 182b may include information defining the location and / or other characteristics of various rooms (e.g., private offices, meeting rooms, etc.) in the physical space. Public (shared) space data 182c may include information defining the location and / or other characteristics of various public (shared) spaces (e.g., cafes, reception areas, libraries, outdoor patios, indoor gardens, etc.) in the physical space. Other setup data 182d may include information defining other characteristics of specific setups within the physical space. In some cases, and as illustrated in more detail below, certain aspects of the setup model can be determined by the generative design computing platform 110 using one or more processes described herein, such as the location and layout of various rooms, public (shared) spaces and / or other setup model features.

[0051] refer to Figure 1EThe illustration depicts an example furniture model 184. Furniture model 184 may include, for example, block model 184a, setup model 184b, furniture identification data 184c, furniture location data 184d, and other furniture data 184e. Block model 184a may include block models that have been generated and / or stored for a specific physical space (e.g., the same space to which furniture model 184 is applied). For example, block model 184a may include block model 180 and / or any of its content data. Setup model 184b may include setup models that have been generated and / or stored for a specific physical space (e.g., the same space to which furniture model 184 is applied). For example, setup model 184b may include setup model 182 and / or any of its content data. Furniture identification data 184c may include information defining one or more specific pieces of furniture (e.g., tables, chairs, etc.) for the physical space, such as one or more stock units (SKUs) corresponding to such furniture, names and / or other identifiers corresponding to such furniture, color details and / or other specifications of such furniture, and / or other identification information. Furniture location data 184d may include information defining the location of one or more specific pieces of furniture within the physical space, such as identifiers indicating the location of tables, chairs, and / or other furniture components at a specific work point, coordinates indicating the location of each piece of furniture within the physical space, and / or other location information. Other furniture data 184e may include information defining other characteristics of the furniture within the physical space. In some cases, and as illustrated in more detail below, aspects of the furniture model may be determined by the generative design computation platform 110 using one or more of the processes described herein, such as including and positioning specific furniture at a specific work point within the physical space.

[0052] refer to Figure 1FThe diagram depicts an example spatial model 186. Spatial model 186 may include, for example, a block model 186a, a setup model 186b, and a furniture model 186c. Block model 186a may include block models that have been generated and / or stored for a specific physical space (e.g., the same space to which spatial model 186 is applied). For example, block model 186a may include block model 180 and / or any content data thereof. Setup model 186b may include setup models that have been generated and / or stored for a specific physical space (e.g., the same space to which spatial model 186 is applied). For example, setup model 186b may include setup model 182 and / or any content data thereof. Furniture model 186c may include furniture models that have been generated and / or stored for a specific physical space (e.g., the same space to which spatial model 186 is applied). For example, furniture model 186c may include furniture model 184 and / or any content data thereof. In some cases, and as illustrated in more detail below, certain aspects of the spatial model can be determined by the generative design computing platform 110 using one or more processes described herein, such as by iteratively generating and optimizing block models, setup models, and / or furniture models for a specific physical space.

[0053] refer to Figure 1G An example geometric model 188 is depicted. Geometric model 188 may include, for example, one or more design rule sets, such as design rule set 188a and design rule set 188n. Each design rule set may include, for example, one or more block rules, setting rules, and / or furniture rules. Such block rules, setting rules, and / or furniture rules may be used, for example, by the generative design computation platform 110 to generate and / or optimize one or more block models, setting models, and / or furniture models, respectively. For example, design rule set 188a may include one or more block rules 188a-1, one or more setting rules 188a-2, and one or more furniture rules 188a-3. Block rule 188a-1 may include information defining one or more rules that define block layout, block adjacency, and / or other block characteristics. Setting rule 188a-2 may include information defining one or more rules that define setting layout, setting adjacency, and / or other setting characteristics. Furniture rule 188a-3 may include information defining one or more rules that define furniture layout, furniture grouping, and / or other furniture characteristics.

[0054] Figures 2A to 2H An illustrative operating environment is described, according to one or more example embodiments, for generating spatial and geometric models using a machine learning system with a multi-platform interface. References Figure 2AAt step 201, the generative design computing platform 110 may receive one or more drawing models from internal data server 120 and / or external data server 130. These drawing models may correspond to different spatial designs (e.g., floor plans, furniture location information, best-in-class designs, etc.). For example, when receiving one or more drawing models, the generative design computing platform 110 may receive one or more two-dimensional computer-aided design (CAD) models, which can be used to train one or more machine learning models to identify design parameters and / or distinguish different design parameters. In some cases, when receiving one or more drawing models, the generative design computing platform 110 may receive a sufficient number of drawing models that is satisfactory and / or sufficient to train one or more machine learning models to distinguish different room types (e.g., meeting rooms, offices, public spaces, etc.) and / or other design features. This training may, for example, be configured and / or enable the generative design computing platform 110 to determine insights and / or relationships related to square feet, adjacency (which may, for example, define and / or indicate the proximity and / or location of various departments, settings, rooms, and / or other spatial features) and / or other typical and / or preferred features and / or characteristics of the physical space. In addition to receiving one or more drawing models at step 201, or as an alternative, the generative design computing platform 110 may also receive, request, or otherwise access photographs, videos, and / or other media corresponding to physical spaces, and may use photographs, videos, and / or other media to generate one or more drawing models.

[0055] At step 202, the generative design computing platform 110 can identify multiple design parameters associated with each of the multiple drawing models corresponding to different spatial designs. In some cases, the generative design computing platform 110 can identify multiple design parameters based on user input (which may be received, for example, at a first designer user computing device 140, a second designer user computing device 150, and / or another computing device, and then sent to the generative design computing platform 110). For example, the user can manually identify design parameters derived from each drawing model and / or otherwise associated with each drawing model. In these cases, when identifying multiple design parameters, the generative design computing platform 110 can identify a common set of design parameters for each of the multiple drawing models. Additionally or alternatively, the generative design computing platform 110 can apply cognitive machine learning to the multiple drawing models to identify multiple design parameters. In these cases, the generative design computation platform 110 can identify multiple design parameters based on graphical features derived from and / or linked to the drawing models, such as metadata information indicating the organization's industry, geographic location, size, personality, and / or other characteristics linked to each of the multiple drawing models. Furthermore, in these cases, the generative design computation platform 110 can identify different design parameters for each of the multiple drawing models.

[0056] In some cases, when identifying multiple design parameters associated with each of the multiple drawing models received at step 201, the generative design computation platform 110 may identify multiple design features prior to identifying the multiple design parameters at step 202. In some cases, these design features may be relatively common to organizations of the same business type (e.g., architecture and design firms may typically occupy spaces with a first set of common features, and these common features may be reflected in the drawing model of a space occupied by such a design firm, while law firms may typically occupy spaces with a second set of common features, and these common features may be reflected in the drawing model of a space occupied by such a law firm). To identify, group, and / or otherwise select these common features from the various drawing models associated with different types of organizations, the generative design computation platform 110 may execute and / or otherwise use one or more cognitive machine learning algorithms. For example, the generative design computation platform 110 may identify, group, and / or otherwise select multiple design features associated with a specific drawing model among the multiple drawing models by applying cognitive machine learning based on the organization and / or occupant corresponding to a specific drawing model among the multiple drawing models. For example, the generative design computing platform 110 can identify features that are most suitable for drawing models associated with a particular organization, which in turn enables the generative design computing platform 110 to infer features that may be applicable when creating spatial and / or geometric models for other similar organizations.

[0057] For example, in some cases, the generative design computing platform 110 can select design features based on the industry, geographic location, size, personality, and / or other characteristics of an organization corresponding to each of the multiple drawing models. For example, for each organization and / or for each drawing model, the generative design computing platform 110 can identify total square feet, total number of offices, total number of meeting spaces, total number of community spaces, number of seats per office, number of seats per meeting space, number of seats per community space, percentage of total square feet allocated to offices, percentage of total square feet allocated to meeting spaces, percentage of total square feet allocated to community spaces, average office size, average meeting space size, and / or other spatial metrics.

[0058] At step 203, the generative design computation platform 110 can train a machine learning engine (e.g., machine learning engine 112d) based on multiple drawing models corresponding to different spatial designs and multiple design parameters associated with each of the multiple drawing models corresponding to different spatial designs. When training the machine learning engine, the generative design computation platform 110 can generate at least one geometric model corresponding to the multiple drawing models. Specifically, when generating the at least one geometric model, the generative design computation platform 110 can determine and / or otherwise generate a set of ranges, numerical constraints, and / or other quantifiable features and / or rules that can be applied by the generative design computation platform 110 to generate a spatial model of physical space based on spatial program data (e.g., as illustrated in more detail below). Additionally or alternatively, when generating the at least one geometric model, the generative design computing platform 110 may generate a hierarchical model, which may have sub-step specific rules for performing different sub-steps of the generative design process (e.g., block rules for performing steps associated with generating a block model, setup rules for performing steps associated with generating a setup model, furniture rules for performing steps associated with generating a furniture model, and / or other layer-specific rules).

[0059] At step 204, the generative design computation platform 110 may store the at least one geometric model. In some cases, the generative design computation platform 110 may store the at least one geometric model locally (e.g., in memory 112 and / or specifically in the generative design database 112b). Additionally or alternatively, the generative design computation platform 110 may store the at least one geometric model at a remote source such as an internal data server 120.

[0060] refer to Figure 2BAt step 205, the generative design computing platform 110 may receive spatial program data from the first designer user computing device 140. For example, the generative design computing platform 110 may receive first spatial program data identifying one or more parameters of the first physical space. In some cases, upon receiving the first spatial program data, the generative design computing platform 110 may receive information identifying architectural details of the first physical space, such as line drawings identifying the shell of the building corresponding to the first physical space, window locations, ceiling heights, preferred views, planarable areas, elevator locations, column locations, entrances, exits, doors, and / or other spatial features (which may be included, for example, in computer-aided design documents). Additionally or alternatively, upon receiving the first spatial program data, the generative design computing platform 110 may receive organizational details of the organization occupying or to occupy the first physical space, such as information indicating the total number of employees of the organization, projected growth rate, and organizational breakdowns (e.g., departments, teams, team composition, relationships between teams and / or departments). Additionally or alternatively, when receiving the first space procedural data, the generative design computation platform 110 may receive working style details of the first physical space, such as information indicating preferences related to open or closed floor plans, privacy concerns, ambiance, and / or other style preferences. Additionally or alternatively, when receiving the first space procedural data, the generative design computation platform 110 may receive budget details of the first physical space, such as information indicating targets and / or maximum price per square foot, and / or measurement details of the first physical space, such as information indicating that some scoring factors are more important than others in the overall selection process.

[0061] In some cases, the first spatial program data (which may be received by the generative design computing platform 110, for example, at step 205) may be set by or for the occupant of the first physical space, and may be received as user input received via a spreadsheet or survey. For example, the occupant of the first physical space may be prompted (e.g., via the generative design computing platform 110, via one or more graphical user interfaces presented on one or more user computing devices) to select images, word clouds (e.g., graphical representations of words and / or phrases displayed in a cloud format and associated with different themes and / or styles that may indicate different design preferences), and / or match possible design elements to their conception of the first physical space. As illustrated below, the generative design computing platform 110 may use any and / or all of the user input when generating multiple spatial models for the first physical space. Additionally or alternatively, upon receiving the first spatial information, the generative design computation platform 110 may receive user input defining specific preferences for one or more design elements of the first physical space. These specific preferences may be such as specific preferences for blocks (e.g., office departments and / or other different areas of the physical space), settings (e.g., room types and / or other sub-block characteristics), and / or furniture items (e.g., individual furniture and / or other sub-setting characteristics). Additionally or alternatively, upon receiving the first spatial information, the generative design computation platform 110 may receive information indicating trends in third-party data, industry standards, best-in-class floor plans, and / or other external data. Also as illustrated below, the generative design computation platform 110 may use any one and / or all of this information when generating multiple spatial models for the first physical space.

[0062] At step 206, the generative design computing platform 110 may load a first geometric model from a database storing one or more geometric models (e.g., stored in memory 112 or on an internal data server 120). When loading the first geometric model, the generative design computing platform 110 may load information defining a first plurality of design rules, which may be part of and / or otherwise associated with the first geometric model, such as design rules for controlling and / or influencing the number, location, size, and / or other characteristics of various spatial model design elements, such as blocks, settings, furniture, and / or other elements (e.g., the number of blocks, settings, furniture, and / or other features; the type of blocks, settings, furniture, and / or other features; the location of blocks, settings, furniture, and / or other features; the location of corridors; and / or other features). In some cases, when loading the first geometric model, the generative design computing platform 110 may use a machine learning engine trained on one or more best-in-class designs to select the first geometric model from multiple geometric models. For example, when loading the first geometric model, the generative design computation platform 110 can use the machine learning engine 112d to select the geometric model generated and / or produced at step 203. Additionally or alternatively, when loading the first geometric model, the generative design computation platform 110 can select the first geometric model based on first spatial program data (e.g., the first spatial program data received at step 205). For example, architectural details of the physical space, organizational details of the physical space, stylistic details of the physical space, and / or budget details of the physical space can influence the selection of the first geometric model and can therefore be used by the generative design computation platform 110 as selection parameters when selecting the first geometric model.

[0063] At step 207, the generative design calculation platform 110 can generate a first plurality of spatial models for a first physical space based on the first spatial program data and the first geometric model. For example, the generative design calculation platform 110 can generate a plurality of spatial models (which may, for example, be a floor plan including block models, setup models, and furniture models, as illustrated in more detail below) based on elements corresponding to the spatial program data received at step 205 and the geometric model loaded at step 206. In some cases, when generating the first plurality of spatial models, the generative design calculation platform 110 can generate each spatial model in the first plurality of spatial models in multiple different data formats. For example, the generative design calculation platform 110 can generate each spatial model in the first plurality of spatial models in CAD format, CET format, Revit format, SketchUp format, and / or one or more other formats. As illustrated in more detail below, by generating each spatial model in a different format at step 207, the generative design computation platform 110 can define the spatial model and / or the elements included in the spatial model only once at the start of the design process. This eliminates the need for downstream adjustments during formatting and thus provides improved efficiency when generating, editing, and / or exporting models.

[0064] In some cases, when generating the first plurality of spatial models, the generative design computing platform 110 can generate multiple block models for the first physical space (which may, for example, indicate how departments and / or furniture settings are arranged on the floor slab). For example, the generative design computing platform 110 can generate multiple block models based on the first spatial program data and the first geometric model to determine where different departments of the organization can be located within the physical space, where corridors and / or walls can be located (e.g., between departments), and / or where and / or how other block-level features can be implemented in the physical space. In some cases, the geometric model may also include adjacency rules (e.g., rules indicating that certain departments should be adjacent to other departments; for example, there may be rules indicating that the legal department should be adjacent to the accounting department), and the generative design computing platform 110 may use and / or consider these adjacency rules when generating the block models (and / or when generating the setup models and / or furniture models, as discussed in more detail below). Furthermore, the various block models that can be generated by the generative design computing platform 110 may correspond to different variations (e.g., variations in department locations and / or other implementation details and / or other block-level features). For example, generative design computing platform 110 may determine that the legal department requires 6,000 square feet and the marketing department requires 20,000 square feet, and the generative design computing platform 110 may place these departments in different locations within the first physical space and / or have different variations across different block models when placed within the first physical space (which may be further refined into detailed floor plans, for example, when generating setup models and furniture models, as described below). In some cases, when generating multiple block models for the first physical space, the generative design computing platform 110 may consider existing offices, rooms, and / or other elements in fixed locations within the first physical space (e.g., immovable elements, preferred and / or undesirable elements (e.g., due to cost issues, workload issues, and / or other issues), and / or elements with characteristics that make them fixed and / or otherwise immovable). In these cases, the generative design computing platform 110 may incorporate the predetermined existing locations of these fixed elements into the multiple block models being generated.

[0065] In some cases, when generating multiple block models for a first physical space, the generative design computing platform 110 may perform one or more preprocessing steps. For example, when performing one or more preprocessing steps, the generative design computing platform 110 may perform flooding to create initial assumed locations for one or more specific departments (e.g., initially placing the legal department in the first part of the block model, the human resources department in the second part, etc.). In some cases, the generative design computing platform 110 may perform flooding based on the known size of each department (which may be expressed, for example, in area (such as square feet) or occupancy (such as number of people or seats)). In some cases, when performing one or more preprocessing steps, the generative design computing platform 110 may iteratively generate various flooding solutions and score these solutions accordingly (e.g., using one or more scoring methods described below regarding the scoring of block models).

[0066] After generating multiple block models, the generative design computation platform 110 can score the multiple block models based on a first geometric model, which can generate a score for each block model. For example, the first geometric model may include multiple design rules, constraints, and / or metrics that define the ideal locations and / or other attributes of block-level features. When scoring each block model, the generative design computation platform 110 can calculate how closely a particular block model fits the design rules, constraints, and / or metrics defined by the geometric model (e.g., by calculating the distance between the “actual” values ​​of the block model and the “ideal” values ​​of the geometric model, and then subtracting these distances from a perfect score of 1 or 100). Based on the scores of the block models, the generative design computation platform 110 can select a subset of multiple block models. For example, the generative design computation platform 110 can sort the multiple block models based on their corresponding scores and then select a subset of the highest-scoring block models (e.g., the generative design computation platform 110 can select block models with the five highest scores). In this way, the generative design computing platform 110 can use higher-scoring block models that more closely fit the “ideal” values ​​defined in the geometric model when generating the setup model, as illustrated in more detail below, while other lower-scoring block models can be discarded (this can, for example, bring technical advantages such as increased computational efficiency, reduced processing load and / or reduced network resource usage).

[0067] For each subset of multiple block models, the generative design computation platform 110 can generate multiple setup models, and each setup model can indicate a specific location of a different office or environmental setup within a different block, such as an office, meeting room, public (shared) space, and / or other setups within a different block, as well as other characteristics of these various setups, such as their size, shape, quantity, intended purpose, and / or other features. Furthermore, each setup model (which can be generated, for example, by the generative design computation platform 110) can correspond to a specific block model within a subset of the multiple block models. After generating the multiple setup models, the generative design computation platform 110 can score the multiple setup models based on a first geometric model (e.g., using one or more evaluation measures), which can generate a score for each setup model. Similar to how the first geometric model can include multiple design rules, constraints, and / or measures defining the ideal location and / or other properties of block-level features, as discussed above, the first geometric model can also include multiple design rules, constraints, and / or measures defining the ideal location and / or other properties of setup-level features. Therefore, similar to scoring block models, when scoring each setup model, the generative design computation platform 110 can calculate how closely a particular setup model fits the design rules, constraints, and / or metrics defined by the geometric model (e.g., by calculating the distance between the “actual” values ​​of the setup model and the “ideal” values ​​of the geometric model, and then subtracting these distances from a perfect score of 1 or 100). Based on the setup model scores, the generative design computation platform 110 can select a subset of multiple setup models. For example, the generative design computation platform 110 can sort multiple setup models based on their corresponding scores and then select a subset of the highest-scoring setup models (e.g., the generative design computation platform 110 can select setup models with the five highest scores). In this way, the generative design computation platform 110 can use higher-scoring setup models that fit more closely to the “ideal” values ​​defined in the geometric model when generating furniture models, as illustrated in more detail below, while other lower-scoring setup models can be discarded (this can, for example, bring technical advantages such as increased computational efficiency, reduced processing load, and / or reduced network resource usage).

[0068] For each subset of multiple setup models, the generative design computation platform 110 can generate multiple furniture models that indicate which specific furniture should be located in which office or environmental setting. Furthermore, each furniture model (which may be generated, for example, by the generative design computation platform 110) may correspond to a specific setup model within the subset of multiple setup models.

[0069] After generating multiple furniture models, the generative design computation platform 110 can score the multiple furniture models based on a first geometric model, which can generate a score for each furniture model. Similar to how the first geometric model can include multiple design rules, constraints, and / or measures defining the ideal locations and / or other properties of block-level and setting-level features, as discussed above, the first geometric model can also include multiple design rules, constraints, and / or measures defining the ideal locations and / or other properties of furniture-level features. Therefore, similar to scoring block and setting models, when scoring each furniture model, the generative design computation platform 110 can calculate how closely a particular furniture model fits the design rules, constraints, and / or measures defined by the geometric model (e.g., by calculating the distance between the "actual" values ​​of the furniture model and the "ideal" values ​​of the geometric model, and then subtracting these distances from a perfect score of 1 or 100). Based on the scores of the furniture models, the generative design computation platform 110 can select a subset of multiple furniture models. For example, the generative design computation platform 110 can sort multiple furniture models based on their corresponding ratings and then select a subset of the highest-rated furniture models (e.g., the generative design computation platform 110 can select a setting model with five highest ratings). Furthermore, the generative design computation platform 110 can output the selected subset of furniture models as a first plurality of spatial models. In this way, the generative design computation platform 110 can use the highest-rated furniture models that more closely fit the “ideal” values ​​defined in the geometric model when determining and / or outputting the spatial model (which may, for example, include full details of block-level features, setting-level features, and furniture-level features). Moreover, this phased and rating-based approach (which can be implemented by the generative design computation platform 110, for example, when determining and / or outputting the spatial model) can provide various technical advantages, such as increased computational efficiency, reduced processing load, and / or reduced network resource usage.

[0070] Furthermore, by generating a plurality of spatial models using the iterative generative design algorithm illustrated above (e.g., by iteratively generating, scoring, and improving block models, setting models, and furniture models), the generative design computation platform 110 can efficiently generate and output optimal spatial models and / or a set of optimal spatial models. Moreover, by moving through the stage gates illustrated above (e.g., generating only the setting model when the block model has been solved, and generating only the furniture model when the setting model has been solved), the generative design computation platform 110 can reduce computational bandwidth consumption and achieve faster computational performance. These benefits can be achieved while taking into account occupant and designer preferences (e.g., as indicated in the spatial program data) and a more agnostic set of design rules (e.g., as defined in the geometric model).

[0071] In some cases, when generating the first plurality of spatial models, the generative design computing platform 110 can generate one or more multi-story stacking plans (which may be, for example, floor plans spanning multiple levels across a building, group of buildings, campus, or other space). In these cases, using a method similar to that described above regarding block models to determine the most suitable location for each department in a specific section of a particular floor, the generative design computing platform 110 can identify a particular floor on which one or more specific departments should be located, and where each of these departments should fit within that particular floor. In this way, the generative design computing platform 110 can place different departments on different floors of a given space, thereby generating multi-story stacking plans.

[0072] In some cases, using methods similar to those described above regarding multi-story stacked planning, the generative design computing platform 110 can generate block models spanning multiple buildings and / or other spaces across a campus. This enables the generative design computing platform 110 to perform campus and / or other large-scale planning. For example, the generative design computing platform 110 can place different departments on different floors of different buildings within a given campus, thereby generating a campus plan that may include one or more multi-story stacked plans (which, for example, may also include block models of each floor). In some cases, the generative design computing platform 110 can generate spatial models involving different building floor slab types. For example, a particular space may have multiple planarable areas on the same floor of a building (e.g., in two related and / or connected towers of a building), and the generative design computing platform 110 can generate models for these different planarable areas using techniques similar to those discussed above and / or below (e.g., by placing blocks, settings, and / or furniture in different planarable areas, while considering other elements already placed in such areas).

[0073] At step 208, the generative design computation platform 110 may score the first plurality of spatial models based on the first geometric model. For example, the generative design computation platform 110 may calculate and / or otherwise generate a score for each spatial model based on design rules, constraints, and / or metrics included in the first geometric model. When scoring the first plurality of spatial models, the generative design computation platform may identify the level and / or degree of conformity of the first plurality of spatial models to one or more metrics defined by the first geometric model. For example, the first geometric model may include one or more metrics, and the generative design computation platform 110 may calculate and / or otherwise evaluate the degree of conformity of the first plurality of spatial models to the first geometric model (e.g., by calculating one or more distances, as described above in the examples concerning block models, setup models, and furniture models that may provide the basis and / or constitute the spatial model; and then summing and / or averaging such distance values). In some cases, the geometric model may include measures such as external views and / or preferred views, daylight, suitability of the setting (e.g., an evaluation of each furniture setting and whether it is placed in a suitable area—e.g., is the work café located near a busy area / is the workstation located in a quiet area?), spatial syntax, aggregate compliance, adjacency (which may include, for example, one or more rules defining that one or more specific departments should preferably be located adjacent to or within a predetermined distance of one or more other specific departments, such as a rule specifying that the product management department should be located adjacent to the engineering department or a rule specifying that the legal department should be located adjacent to the accounting department), and / or noise / distraction (which may include, for example, one or more rules for balancing incidental social encounters between space occupants with potential distractions encountered or experienced by space occupants due to certain layout features). In these cases, when scoring the first plurality of spatial models, the generative design computation platform 110 may calculate and / or otherwise generate a score for each spatial model based on the degree to which each spatial model provides features aligned with these measures.

[0074] For example, when scoring the first plurality of spatial models, the generative design computing platform 110 can quantify and / or otherwise assess the external view, preferred view, and / or access to daylighting metrics by identifying a line from a chair at the work point to a window in the physical space (e.g., as indicated in a given spatial model), calculating the distance of that line, and identifying whether there are any objects between the work point (e.g., chair, sofa, seat, etc.) and the window (e.g., wall, partition, etc.), or whether that distance exceeds a predetermined threshold (e.g., whether the distance is too far for a person at the work point to enjoy the view). In performing this assessment, the generative design computing platform 110 may also consider what the view from a given work point includes (e.g., a view from a courtyard might be more desirable than a view from a parking lot or the walls of an adjacent building). The generative design computing platform 110 may also consider the building's position relative to the sun. Any and / or all of these considerations can be quantified by metrics and can be used by the generative design computing platform 110 to score each of the first plurality of spatial models.

[0075] As another example, when scoring the first plurality of space models, the generative design computing platform 110 can quantify and / or otherwise evaluate the suitability of each work point and / or each setting in the space model by identifying the surrounding environment of a given work point or setting and determining whether and / or to what extent the work point or setting conforms to the rules of the setting model. For example, in performing such an evaluation, in some cases, the generative design computing platform 110 can determine the occupancy rate of a given space, predict the decibel level in the space based on the predicted occupancy rate, and based on that decibel level, identify how far away offices or other work points should be from the space in order to maintain a quiet office or work point.

[0076] As another example, when scoring the first plurality of spatial models, the generative design computing platform 110 can quantify, evaluate, and / or otherwise score the spatial syntax of a given spatial model by identifying predicted traffic patterns in the physical space based on the layout of the spatial model (e.g., the number of turns required to move from one location to another in the space, the unobstructedness of corridors in the space, the proximity of relevant teams, the degree to which the space provides the possibility of chance encounters, and / or other spatial syntax factors). For example, when scoring a given spatial model, the generative design computing platform 110 can balance maintaining short distances between frequently visited parts of the physical space for individuals with allowing individuals in that space to experience chance encounters (e.g., it might be expected that everything in the space is easily accessible to people belonging to different teams, while still allowing people belonging to different teams to occasionally encounter someone from another team). After quantifying and / or otherwise evaluating one or more of the above features, the generative design computing platform 110 can calculate and / or otherwise determine a score (e.g., 1 to 10, etc.) for each metric of each spatial model in the first plurality of spatial models. Then, the generative design computation platform 110 can calculate the total score of each of the first plurality of spatial models, for example, by calculating the average of the metric scores determined for a particular spatial model.

[0077] refer to Figure 2C At step 209, the generative design computation platform 110 may sort the first plurality of spatial models based on the score of each spatial model (e.g., the score generated at step 208). In doing so, the generative design computation platform 110 may generate a first sorted list of spatial models for a first physical space. In some cases, the generative design computation platform 110 may sort the first plurality of spatial models based on the metric score of each spatial model and / or the aggregate score of each spatial model.

[0078] At step 210, the generative design computation platform 110 can generate first user interface data, which includes a first sorted list of the spatial model generated at step 209. The first user interface data generated by the generative design computation platform 110 can define a graphical user interface (such as the following combination) Figure 3(A more detailed description of the user interface) or one or more portions thereof. At step 211, the generative design computing platform 110 may send first user interface data to the first designer user computing device 140 via communication interface 113. In some cases, by sending the first user interface data to the first designer user computing device 140, the generative design computing platform 110 may cause the first designer user computing device 140 to display a user interface that includes at least a portion of a first sorted list of spatial models. In some cases, sending the user interface data to the first designer user computing device 140 may cause the first designer user computing device 140 to display one or more of the scores calculated for each spatial model at step 208 (e.g., metric scores, aggregate scores, and / or other scores discussed in the above examples).

[0079] At step 212, the first designer user computing device 140 may display a user interface including at least a portion of a first sorted list of spatial models. For example, the first designer user computing device 140 may display based on first user interface data received from the generative design computing platform 110. Figure 3 The graphical user interface shown is similar to the graphical user interface 300. For example... Figure 3 As shown, the graphical user interface 300 may include information identifying one or more different spatial models generated by the generative design computing platform 110, sorting information indicating the sorting and / or scoring of one or more spatial models, and / or visual information indicating graphical views of one or more spatial models and / or portions thereof generated by the generative design computing platform 110. In some cases, the first designer user computing device 140 may display such a user interface based on or in response to user interface data received from the generative design computing platform 110. Additionally or alternatively, when displaying a user interface including at least a portion of a first sorted list of spatial models, the first designer user computing device 140 may display the metric score and / or total score calculated at step 208 (and / or other scores discussed in the above examples).

[0080] In some cases, when displaying a user interface that includes at least a portion of a first sorted list of spatial models, the first designer user computing device 140 may display each of the spatial models in a grid, along with measurements corresponding to each spatial model (e.g., based on user interface data received from the generative design computing platform 110). In these cases, in response to receiving user input selecting a portion of the displayed spatial model, the first designer user computing device 140 may display renderings of one or more workstations (e.g., seats) in the spatial model and / or other graphics associated with those workstations, as well as calculations of the external view from each of the one or more workstations and / or other measurements associated with each workstation.

[0081] refer to Figure 2D At step 213, the generative design computing platform 110 may receive data instructing the selection of a first spatial model from a first list of sorted spatial models. For example, the generative design computing platform 110 may receive data instructing the selection of a first spatial model from a first designer user computing device 140 via communication interface 113.

[0082] At step 214, the generative design computing platform 110 can generate a visual rendering of the first spatial model. For example, the generative design computing platform 110 can generate a visual rendering of the first spatial model in response to or based on data indicating that the first spatial model is selected from a first sorted list of spatial models. In some cases, when generating a visual rendering of the first spatial model, the generative design computing platform 110 can generate a two-dimensional or three-dimensional rendering of the first spatial model. In some cases, when generating such a rendering, the generative design computing platform 110 can use rendering software built into the drawing tools to convert blocks, settings, furniture, and / or other elements indicated in the spatial model into two-dimensional and / or three-dimensional objects that can be viewed by the user and / or reflect a view of the space if the spatial model is to be implemented.

[0083] At step 215, the generative design computing platform 110 may send a visual rendering of the first spatial model to the first designer user computing device 140 (e.g., via communication interface 113). In some cases, sending the visual rendering of the first spatial model to the first designer user computing device 140 may cause the first designer user computing device 140 to display a user interface that includes at least a portion of the visual rendering of the first spatial model. For example, by sending the visual rendering of the first spatial model to the first designer user computing device 140, the generative design computing platform 110 may cause the first designer user computing device 140 to display a user interface similar to a graphical user interface 400, which in... Figure 4It is shown in the figure and described below.

[0084] In some cases, users of the First Designer User Computing Device 140 can modify the parameters of the spatial model for various reasons, such as to further refine the spatial model, optimize parameters beyond the calculations performed by the generative design computing platform 110, and / or refine the spatial model to take into account the need for maintaining social distancing. For example, the First Designer User Computing Device 140 can display... Figure 4 The graphical user interface 400 shown is similar to a graphical user interface. In doing so, the first designer user computing device 140 can allow the user to modify variables that can control, change, and / or otherwise affect the layout of the space model and / or other parameters of the space model. For example, the first designer user computing device 140 can display and / or otherwise present one or more user-selectable controls that allow the user to modify variables such as circulation percentage (which can, for example, affect corridor width and / or other parameters that affect the circulation of individuals within the space), group space percentage (which can, for example, affect the relative amount of space allocated to public spaces), office and workstation sizes (which can, for example, affect the size of various workstations to optimize for social distancing needs), sharing ratios (which can, for example, affect whether and to what extent work areas are configured as shared hotspots rather than reserved desktops) and / or other variables.

[0085] In these cases, when one or more of the variables are modified, the first designer user computing device 140 can demonstrate the impact of the modification (e.g., by displaying updated information indicating how many people can be placed in the office and / or other effects of the variable modification). This updated data can be determined, for example, by the first designer user computing device 140, or the first designer user computing device 140 can send the modification to the generative design computing platform 110 (which can, for example, calculate and / or otherwise determine the impact of the variable modification and return data indicating the impact of the variable modification to the first designer user computing device 140). In some cases, the first designer user computing device 140 can receive user input corresponding to new building blocks, such as barriers to be deployed between workers and / or other spatial materials designed to have antiviral properties. The first designer user computing device 140 can then send this user input and / or other information associated with the new building blocks to the generative design computing platform 110, which can incorporate them into the spatial model (e.g., by regenerating the spatial model and / or one or more other spatial models, for example, by re-performing one or more of the steps described above). Additionally or alternatively, the first designer user computing device 140 may receive user input identifying one or more pieces of furniture already owned by the space occupant, and may send this user input and / or other information associated with the one or more pieces of furniture already owned by the space occupant to the generative design computing platform 110. The generative design computing platform 110 may then incorporate such furniture into the space model (e.g., by regenerating the space model and / or one or more other space models, for example, by re-performing one or more of the steps described above). In this way, the generative design computing platform 110 may generate one or more space models that indicate the potential reconfiguration of already owned furniture (e.g., to facilitate compliance with new social distancing requirements in existing spaces, such as existing office spaces), rather than proposing new space models involving the purchase and / or deployment of entirely new furniture sets.

[0086] At step 216, the generative design computing platform 110 may receive data indicating user modifications to the first spatial model (e.g., received from the first designer user computing device 140 via communication interface 113). In some cases, the data indicating user modifications to the first spatial model may correspond to user modifications received by the first designer user computing device 140 via the graphical user interface displayed at step 215. For example, at step 216, the generative design computing platform 110 may receive data indicating user modifications such as refinement of the spatial model and / or manual optimization of one or more parameters of the lower layers of the spatial model, as in the examples discussed above.

[0087] refer to Figure 2E At step 217, based on or in response to receiving data indicating user modifications to the first spatial model, the generative design computing platform 110 may update the machine learning engine executed on the generative design computing platform 110. For example, the generative design computing platform 110 may update and / or retrain the machine learning engine based on the data indicating user modifications to the first spatial model. For example, in terms of the degree to which the user manually refines the layout of the spatial model and / or manually optimizes one or more parameters of the lower layers of the spatial model, such refinement and / or optimization may be captured by the generative design computing platform 110 and used to retrain the machine learning engine, so that when generating future spatial models, the generative design computing platform 110 can automatically implement such refinement and / or optimization. Additionally or alternatively, when updating the machine learning engine, the generative design computing platform 110 may cause the machine learning engine 112d to automatically update the first geometric model and / or the metrics corresponding to the first geometric model. For example, in terms of the degree of manual refinement and / or optimization of elements relating to the geometric model and / or its associated metrics, the generative design computation platform 110 can update the geometric model and / or the metrics corresponding to the geometric model, so that when a future spatial model is generated based on the same geometric model, the generative design computation platform 110 can automatically perform refinement and / or optimization.

[0088] Subsequently, the generative design computing platform 110 can continue to process spatial program data and / or generate spatial models for other physical spaces, similar to how the generative design computing platform 110 can process spatial program data and generate spatial models in the examples discussed above. For example, at step 218, the generative design computing platform 110 can receive second spatial program data (e.g., received from the second designer user computing device 150 via communication interface 113). For example, the generative design computing platform 110 can receive information identifying one or more parameters of a second physical space that is different from the first physical space. In some cases, the actions performed at step 218 can be similar to those described above at step 205 regarding the receipt of the first spatial program data.

[0089] At step 219, the generative design computation platform 110 can load a second geometric model from a database storing one or more geometric models. For example, the generative design computation platform 110 can load information defining a second plurality of design rules. In some cases, the actions performed at step 219 may be similar to those described above in step 206 regarding loading the first geometric model. At step 220, the generative design computation platform 110 can generate a second plurality of spatial models for a second physical space based on the second spatial program data and the second geometric model. In some cases, the actions performed at step 220 may be similar to those described above in step 207 regarding generating the first plurality of spatial models.

[0090] refer to Figure 2F At step 221, based on the second geometric model, the generative design computation platform 110 can score the second plurality of spatial models. In some cases, when scoring the second plurality of spatial models, the generative design computation platform 110 can generate a score for each spatial model in the second plurality of spatial models. In some cases, the actions performed at step 221 can be similar to those described above in step 208 regarding the scoring of the first plurality of spatial models. At step 222, the generative design computation platform 110 can sort the second plurality of spatial models based on the score of each spatial model in the second plurality of spatial models. In some cases, by sorting the second plurality of spatial models, the generative design computation platform 110 can generate a second sorted list of spatial models. In some cases, the actions performed at step 222 can be similar to those described above in step 209 regarding the sorting of the first plurality of spatial models.

[0091] At step 223, the generative design computing platform 110 may generate second user interface data including a second sorted list of spatial models. In some cases, the actions performed at step 222 may be similar to those described above regarding the generation of the first user interface data at step 210. At step 224, the generative design computing platform 110 may send the second user interface data to a second designer user computing device 150 (e.g., via communication interface 113). In some cases, when sending the second user interface data to the second designer user computing device 150, the generative design computing platform 110 may cause the second user computing device to display a user interface including at least a portion of the second sorted list of spatial models. In some cases, the actions performed at step 224 may be similar to those described above regarding the sending of the first user interface data at step 211.

[0092] refer to Figure 2GAt step 225, based on the second user interface data, the second designer user computing device 150 can display a user interface including at least a portion of a second sorted list of spatial models. In some cases, the second designer user computing device 150 can display... Figure 5 The graphical user interface 500 shown is similar to a graphical user interface. In some cases, the actions performed at step 225 may be similar to those described above regarding the display of the user interface at step 212. For example, as... Figure 5 As shown, the graphical user interface 500 may include information identifying one or more different spatial models generated by the generative design computing platform 110 for the second physical space, sorting information indicating the sorting and / or scoring of one or more spatial models, and / or visual information indicating graphical views of one or more spatial models and / or portions thereof generated by the generative design computing platform 110.

[0093] Subsequently, the generative design computation platform 110 can receive and process requests to export one or more spatial models. As illustrated in more detail below, when processing such requests, the generative design computation platform 110 can output data in various different formats using one or more of the multi-platform interoperability features described herein. Specifically, and as described above (e.g., with respect to step 207), the generative design computation platform 110 can generate each of the multiple spatial models in multiple different data formats (e.g., in CAD format, CET format, Revit format, SketchUp format, and / or one or more other formats), and this multi-format generation can accelerate the process of exporting data in different formats.

[0094] For example, at step 226, the generative design computing platform 110 may receive data instructing a request to export a spatial model (e.g., a first spatial model) to a first design tool. In some cases, the generative design computing platform 110 may receive the data instructing a request to export a spatial model to a first design tool from a first designer user computing device 140 and via communication interface 113. In some cases, upon receiving the data instructing a request to export a spatial model to a first design tool, the generative design computing platform 110 may receive the data instructing the export of a spatial model defined in multiple different data formats (e.g., in CAD format, CET format, Revit format, SketchUp format, and / or one or more other formats) in a specific format compatible with the first design tool and / or in a format that can be processed by the first design tool.

[0095] At step 227, in response to receiving data instructing the export of the spatial model to a first design tool, the generative design computation platform 110 may generate one or more first drawing files based on the first spatial model. In some cases, when generating such drawing files, the generative design computation platform 110 may select a first data format from a plurality of data formats based on the first design tool (e.g., based on the compatibility of the first design tool with different drawing file formats). In these cases, once the first data format is selected, the generative design computation platform 110 may extract first format-specific data (which may be defined, for example, in the first data format) from the first spatial model. Specifically, and as discussed above, the first spatial model may initially be generated in a plurality of different data formats (e.g., in CAD format, CET format, Revit format, SketchUp format, and / or one or more other formats). Therefore, in order to generate drawing files from the first spatial model in any specific format, the generative design computing platform 110 may only need to extract format-specific data from the first spatial model (this can, for example, provide many technical advantages, such as increased efficiency, reduced processing load, and / or reduced network resource consumption). Once the first format-specific data has been extracted, the generative design computing platform 110 can create one or more first drawing files by writing the first format-specific data extracted from the first spatial model into one or more new drawing files defined according to the first data format.

[0096] At step 228, the generative design computing platform 110 may send one or more first drawing files to the first designer user computing device 140 (e.g., via communication interface 113). In some cases, by sending one or more first drawing files to the first designer user computing device 140, the generative design computing platform 110 may cause the first designer user computing device 140 to display one or more first drawing files. (See reference...) Figure 2H At step 229, the first designer user computing device 140 can receive and display one or more first drawing files.

[0097] Subsequently, the generative design computing platform 110 can generate and / or provide one or more user interfaces that enable a customer (e.g., an occupant of a physical space) to purchase one or more furniture elements associated with the space model and / or otherwise view and / or implement the space model. For example, at step 230, the generative design computing platform 110 can generate and send one or more commands to instruct the customer user computing device 160 to display a graphical user interface including user-selectable furniture purchase elements. In some cases, when generating and sending one or more commands to instruct the customer user computing device 160 to display a graphical user interface including user-selectable furniture purchase elements, the generative design computing platform 110 can cause the customer user computing device 160 to display a graphical user interface including user-selectable furniture purchase elements. For example, the customer user computing device 160 can display... Figure 6 The graphical user interface shown is similar to the graphical user interface 600. For example... Figure 6 As shown, the graphical user interface 600 may include information about the spatial model (e.g., measurements, ratings, details associated with blocks, settings, and / or furniture, and / or other information), one or more renderings of the spatial model, and / or one or more user-selectable options enabling the adoption of the spatial model and / or the purchase of one or more furniture items associated with the spatial model. In some cases, the graphical user interface 600 may include final pricing information for the spatial model (e.g., based on the included blocks, settings, furniture, and / or other information), which may be based on pricing information pulled from internal and / or external data sources. For example, when generating the user interface and / or displaying such a user interface on the customer user computing device 160, the generative design computing platform 110 may calculate and / or otherwise determine cost estimates and / or price estimates that indicate the predicted costs of constructing and / or otherwise implementing the spatial model. For example, the generative design computation platform 110 can calculate and / or otherwise determine the estimated cost of constructing a specified setting in the spatial model (e.g., based on data maintained and / or stored by the generative design computation platform 110 that indicates the standard and / or average cost of similar settings in similar spaces). Additionally or alternatively, the generative design computation platform 110 can calculate and / or otherwise determine the estimated cost of purchasing one or more specified pieces of furniture in the spatial model (e.g., based on unit-level pricing data and / or other details, which may be retrieved by the generative design computation platform 110 from another system or database, such as a harbor database).

[0098] In some cases, when generating one or more user interfaces associated with a spatial model, the generative design computing platform 110 may determine that there is additional space in the plan (e.g., positive flex) or insufficient space in the plan (e.g., negative flex), and may generate such user interfaces to indicate and / or otherwise enable interaction with that positive and / or negative flex. Therefore, when displaying one or more user interfaces associated with a spatial model, the client-user computing device 160 may display a spatial model with positive and / or negative flex. For example, when displaying a spatial model with positive flex, the client-user computing device 160 may display a floor plan with rooms for additional furniture. In these cases, the user of the client-user computing device 160 (who may be, for example, a designer) can select additional furniture to fill the space, and these selections can be communicated by the client-user computing device 160 to the generative design computing platform 110, which can update one or more data records to indicate the selection and / or other changes to the spatial model. When displaying a space model with negative flexibility, the client-user computing device 160 can display furniture exceeding the available space in the floor plan (e.g., sofas and / or other furniture exceeding the dimensions of a specific space). In these cases, the user of the client-user computing device 160 can expand the corresponding blocks within the space model to accommodate any additional required space, and this expansion and / or other associated changes can be communicated by the client-user computing device 160 to the generative design computing platform 110, which can update one or more data records to indicate the expansion and / or other changes to the space model. In some cases, instead of presenting the user with the option to reconfigure furniture, the computing device (such as the client-user computing device 160) can mimic the designer's flexibility and automatically modify the floor plan accordingly based on the available space. In some cases, when displaying positive and / or negative flexibility, the client-user computing device 160 can display furniture exceeding the available space in the floor plan. Figure 8 The graphical user interface shown is similar to the graphical user interface 800.

[0099] At step 231, the generative design computing platform 110 may receive furniture selection information indicating an order for one or more furniture items. For example, the furniture selection information may be based on user input received at step 230 via a graphical user interface displayed on the client user computing device 160, and may be sent from the client user computing device 160 to the generative design computing platform 110. At step 232, the generative design computing platform 110 may process the order for one or more furniture items specified in the furniture selection information received at step 231. For example, the generative design computing platform 110 may cause one or more furniture items to be purchased and sent to an address specified by the user of the client user computing device 160.

[0100] Subsequently, the generative design computing platform 110 can repeat one or more steps of the example sequence discussed above when generating other geometric models, generating other spatial models, and / or outputting other drawing files associated with various spatial models. Furthermore, the generative design computing platform 110 can continuously update its machine learning engine 112d based on user input and / or other data received by the generative design computing platform 110 to continuously and automatically optimize the generation of geometric and spatial models.

[0101] In some cases, user applications integrated with the features described in steps 201 to 231 can be designed and implemented, allowing for further customization and functionality beyond those described above. For example, one or more of the features described above can be hosted on and / or provided by a cloud-based Software as a Service (SaaS) platform, on which various designers and / or developers can build custom applications for their own or others' use. These custom applications can, for example, be hosted on the generative design computing platform 110 or on different and / or external computing platforms. In some cases, these custom applications can be integrated with, use, and / or replace the functions and / or features of the tools described above. For example, any and / or all aspects of the custom application can be presented as additional or alternative options in a settings selector tool that can be executed on and / or integrated with the generative design computing platform 110.

[0102] In some cases, any and / or all of the data generated and / or used by the generative design computing platform 110 may be stored and / or otherwise maintained in a single centralized project asset and designer database. Such a database may also include entries from other sources, such as sales data and / or reconnaissance data. In some arrangements, such a centralized database may consist of multiple tabular and / or subsidiary databases, such as a project asset database (which may, for example, store data related to a specific project, such as a spatial model and / or other entries for that project), a designer database (which may, for example, store designer preferences), and a settings database (which may, for example, store data about specific furniture items and / or may be connected to one or more external databases, such as Herman Miller's port database).

[0103] Figure 7 An illustrative method for generating spatial and geometric models using a machine learning system with a multi-platform interface is described, according to one or more example embodiments. References Figure 7At step 705, a computing platform having at least one processor, communication interface, and memory can receive one or more drawing models. At step 710, the computing platform can identify design parameters based on one or more drawing models. At step 715, based on the design parameters, the computing platform can generate one or more geometric models. At step 720, the computing platform can store the geometric models. At step 725, the computing platform can receive spatial program data. At step 730, the computing platform can load one or more geometric models based on the spatial program data. At step 735, the computing platform can generate one or more spatial models based on one or more geometric models and spatial program data. At step 740, the computing platform can store one or more spatial models. At step 745, the computing platform can score one or more spatial models and sort one or more spatial models based on the scores. At step 750, the computing platform can send user interface data to a designer user computing device, which can enable the designer user computing device to display a graphical user interface including a sorted list of one or more spatial models. At step 755, the computing platform can receive model selection data indicating the selection of a first spatial model. At step 760, the computing platform can generate a visual rendering of the first spatial model and send the visual rendering to the designer's user computing device. At step 765, the computing platform can determine whether it has received data indicating a modification to the first spatial model. If not, the computing platform can proceed to step 775. If it has received data indicating a modification to the first spatial model, the computing platform can proceed to step 770.

[0104] At step 770, the computing platform may update the machine learning engine used to generate the geometric model and / or spatial model. At step 775, the computing platform may determine whether a request to export data for the first spatial model has been received. If not, the method may terminate. If a request to export data for the first spatial model has been received, the computing platform may proceed to step 780.

[0105] At step 780, the computing platform may send one or more drawing files based on the first spatial model in response to the export request. At step 785, the computing platform may send one or more commands to instruct the client user's computing device to display a user interface that prompts the user to select furniture (e.g., from the first spatial model) for purchase. At step 790, the computing platform may indicate whether furniture selection data has been received. If not, the method may terminate. If furniture selection data has been received, the computing platform may proceed to step 795. At step 795, the computing platform may process the order corresponding to the furniture selection data.

[0106] Figures 9A to 9BAn illustrative sequence of events for providing a workplace configuration interface is described according to one or more example embodiments. Figures 9A to 9B The actions described herein can be performed alternatively or as an alternative to the aforementioned spatial procedural approach for providing insights into a specific project. (See reference) Figure 9A At step 901, the first designer user computing device 140 may display a graphical user interface that allows input of department information, and may receive department information through this graphical user interface. For example, the first designer user computing device 140 may receive input indicating which departments (e.g., administration, legal, finance, sales, marketing, communications, design, and / or other departments) should be included in the final workplace configuration.

[0107] At step 902, the first designer user computing device 140 may display a graphical user interface that allows input of job information, and may receive job information through this graphical user interface. For example, the first designer user computing device 140 may receive input indicating which jobs correspond to various departments (e.g., executives, supervisors, vice presidents, directors, managers, staff, and / or other positions).

[0108] At step 903, the first designer user computing device 140 can display a graphical user interface that allows input of total employee information, and can receive the total employee information through this graphical user interface. For example, the first designer user computing device 140 can receive input indicating the number of employees in each position (identified at step 902) of each department (identified at step 901). As a specific example, the first designer user computing device 140 can receive input indicating that the legal department has two supervisors.

[0109] At step 904, the first designer user computing device 140 can communicate with the generative design computing platform 110 to share the information received in steps 901 to 903 (e.g., department, position, and total number of employees information). At step 905, the generative design computing platform 110 can receive the information sent in step 904.

[0110] refer to Figure 9BGenerative design computing platform 110 can use the information received at step 905 to generate a workpoint configuration interface (or information that can be used to generate a workpoint configuration interface). For example, generative design computing platform 110 can identify the total number of employees for each position, the corresponding default workspace for each position (e.g., offices and workstations and their corresponding sizes), the square feet of a single corresponding workspace (e.g., a single office, workstation, etc. for a specific position), and the square feet of the workspace for all employees for each position (e.g., the total square feet occupied by seven chief operating officers is 1925 square feet, and each chief operating officer's office is 275 square feet). After identifying these metrics, generative design computing platform 110 can identify the total square feet occupied by all expected employees for all positions / departments (e.g., by adding all identified total square feet for each role), and can then add group space sizes (e.g., meeting rooms, open collaboration rooms, and / or other group spaces), support space sizes (e.g., printer / copying areas, storage rooms, LAN rooms, and / or other support spaces) and / or any other square feet to reach a total available area in square feet.

[0111] In some cases, when generating the workpoint configuration interface, the generative design computing platform 110 may include controls that allow modification of breakpoints (e.g., where in the hierarchy employees are assigned offices and workstations), workpoint sizes (e.g., office or workstation size), and / or other parameters. For example, this could allow a user to modify the workpoint configuration so that only managers or higher-level employees can have offices, rather than supervisors or higher-level employees. In doing so, the user could reduce the square footage occupied by supervisor-level employees by moving them from offices to workstations. Additionally or alternatively, the workpoint configuration could be modified to reduce the office size for employees with specific roles, thereby reducing the square footage occupied by those employees. Conversely, if additional available space exists, the user could modify the workpoint configuration to increase the number of employees occupying offices and / or increase the size of individual offices.

[0112] In some cases, the generative design computation platform 110 can generate one or more workpoint configuration options that include the information described above, and can include each option on the workpoint configuration interface. In this way, users can identify which option is most desired and modify that option as needed.

[0113] By automatically generating these working point configuration options, the generative design computation platform 110 can save significant time that would otherwise be consumed by designing a complete test fit for a specific space and then refining the fit as needed based on whether it exceeds the total available area of ​​that space, the total available area of ​​the unused space, and / or other factors.

[0114] In some cases, the generative design computing platform 110 can also be configured to generate cost estimates for each workplace (e.g., based on the price per square foot from the corresponding lease and the identified total square feet for each workplace configuration). In doing so, the generative design computing platform 110 can enable users to identify cost savings associated with each workplace configuration (e.g., cost savings associated with a smaller space compared to a larger space).

[0115] In step 907, the generative design computing platform 110 can send a workpoint configuration interface to the first designer user computing device 140 for display. In step 908, the first designer user computing device 140 can receive the workpoint configuration interface.

[0116] At step 909, the first designer user computing device 140 can display a workpoint configuration interface. For example, the first designer user computing device 140 can display... Figure 10 The graphical user interface 1000 illustrated is similar to a graphical user interface. For example, a first designer user computing device 140 can display a graphical user interface that allows the user to adjust the workplace configuration and observe the corresponding cost savings as described in step 906 above.

[0117] Although the systems, methods, and event sequences described above primarily illustrate use cases involving commercial office design, they can be similarly applied to other use cases, such as residential design, outdoor design, manufacturing facilities, etc., without departing from the scope of this disclosure. For example, the generative design computation platform 110 can perform one or more steps similar to those described above when generating spatial models for residential spaces, outdoor spaces, manufacturing facility spaces, and / or other types of spaces.

[0118] One or more aspects of this disclosure may be embodied in computer-usable data or computer-executable instructions, such as in one or more program modules, executed by one or more computers or other devices to perform the operations described herein. Program modules may include routines, programs, objects, components, data structures, etc., which, when executed by one or more processors in a computer or other data processing device, perform a specific task or implement a specific abstract data type. Computer-executable instructions may be stored as computer-readable instructions on a computer-readable medium such as a hard disk, optical disk, removable storage medium, solid-state memory, RAM, etc. In various embodiments, the functionality of program modules may be combined or distributed as desired. Furthermore, functionality may be embodied wholly or partially in firmware or hardware equivalents, such as integrated circuits, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), etc. Specific data structures may be used to more effectively implement one or more aspects of this disclosure, and such data structures are envisioned to be within the scope of the computer-executable instructions and computer-usable data described herein.

[0119] One or more aspects described herein may be embodied as a method, apparatus, or one or more computer-readable media storing computer-executable instructions. Therefore, these aspects may take the form of a completely hardware embodiment, a completely software embodiment, a completely firmware embodiment, or an embodiment combining software, hardware, and firmware aspects in any combination. Furthermore, various signals representing the data or events described herein may be transmitted between a source and a destination in the form of light waves or electromagnetic waves that travel through signal transmission media such as metal wires, optical fibers, and / or wireless transmission media (e.g., air and / or space). One or more computer-readable media may be and / or include one or more non-transitory computer-readable media.

[0120] As described herein, various methods and actions can operate across one or more computing servers and one or more networks. Functionality can be distributed in any manner or may reside within a single computing device (e.g., a server, client computer, etc.). For example, in an alternative embodiment, one or more of the computing platforms discussed above can be combined into a single computing platform, and the various functions of each computing platform can be performed by the single computing platform. In such an arrangement, any and / or all of the communications discussed above between computing platforms can correspond to data accessed, moved, modified, updated, and / or otherwise used by the single computing platform. Additionally or alternatively, one or more of the computing platforms discussed above can be implemented in one or more virtual machines provided by one or more physical computing devices. In such an arrangement, the various functions of each computing platform can be performed by one or more virtual machines, and any and / or all of the communications discussed above between computing platforms can correspond to data accessed, moved, modified, updated, and / or otherwise used by one or more virtual machines.

[0121] Various aspects of this disclosure have been described in accordance with their illustrative embodiments. Numerous other embodiments, modifications, and variations within the scope and spirit of the appended claims will arise to those skilled in the art from a review of this disclosure. For example, one or more of the steps depicted in the illustrative drawings may be performed in an order other than that listed, and one or more of the depicted steps may be optional according to various aspects of this disclosure.

Claims

1. A computing platform, comprising: At least one processor; Communication interface; as well as The memory stores computer-readable instructions that, when executed by the at least one processor, cause the computing platform to: Receive first space program data, identifying one or more parameters of the first physical space, from the first user computing device via the communication interface; Load a first geometric model from a database that stores one or more geometric models, wherein the first geometric model includes information defining a first plurality of design rules; Based on the first spatial program data identifying the first physical space and the first geometric model, a first plurality of spatial models are generated for the first physical space; The first plurality of spatial models generated for the first physical space are scored based on the first geometric model, wherein scoring the first plurality of spatial models generated for the first physical space generates a score for each of the first plurality of spatial models. The first plurality of spatial models generated for the first physical space are sorted based on the score of each of the first plurality of spatial models, wherein sorting the first plurality of spatial models generated for the first physical space produces a sorted list of spatial models; Generate user interface data that includes a sorted list of the spatial model; as well as A visual rendering of a first spatial model selected from a sorted list of spatial models is sent to the first user computing device via the communication interface, wherein sending the visual rendering of the first spatial model to the first user computing device causes the first user computing device to display a user interface including at least a portion of the visual rendering of the first spatial model.

2. The computing platform of claim 1, wherein receiving the first space program data identifying the one or more parameters of the first physical space includes receiving information identifying architectural details of the first physical space, organizational details of the first physical space, working style details of the first physical space, and budget details of the first physical space.

3. The computing platform of claim 1, wherein loading the first geometric model from the database storing the one or more geometric models comprises selecting the first geometric model from a plurality of geometric models generated by the computing platform using a machine learning engine trained on one or more best-in-class spatial designs.

4. The computing platform of claim 1, wherein loading the first geometric model from the database storing the one or more geometric models comprises selecting the first geometric model based on the first spatial program data identifying the first physical space.

5. The computing platform according to claim 1, wherein generating the first plurality of spatial models for the first physical space based on the first spatial program data identifying the first physical space and the first geometric model comprises: Generate multiple block models for the first physical space; The plurality of block models generated for the first physical space are scored based on the first geometric model, wherein scoring the plurality of block models generated for the first physical space generates a score for each of the plurality of block models. A subset of the multiple block models is selected based on the score of each of the multiple block models; Generate multiple setup models for the first physical space, wherein each of the multiple setup models corresponds to a specific block model in the subset of the multiple block models; The plurality of setup models generated for the first physical space are scored based on the first geometric model, wherein scoring the plurality of setup models generated for the first physical space generates a score for each of the plurality of setup models. A subset of the plurality of setting models is selected based on the rating of each of the plurality of setting models; Generate multiple furniture models for the first physical space, wherein each of the multiple furniture models corresponds to a specific setting model in the subset of the multiple setting models; The plurality of furniture models generated for the first physical space are scored based on the first geometric model, wherein scoring the plurality of furniture models generated for the first physical space generates a score for each of the plurality of furniture models. as well as A subset of the plurality of furniture models is selected based on the rating of each of the plurality of furniture models, wherein the subset of the plurality of furniture models corresponds to the first plurality of space models generated for the first physical space.

6. The computing platform of claim 5, wherein each of the plurality of block models indicates a potential location in a different neighborhood in the first physical space, each of the plurality of setting models indicates a potential location in a different work setting in the first physical space, and each of the plurality of furniture models indicates a potential location in a different furniture item in the first physical space.

7. The computing platform of claim 1, wherein the scoring indication of each of the first plurality of spatial models is consistent with the level of conformity of one or more metrics defined by the first geometric model.

8. The computing platform of claim 1, wherein sending the user interface data, including the sorted list of the spatial models, to the first user computing device further causes the first user computing device to display one or more of the ratings determined by each of the first plurality of spatial models.

9. The computing platform of claim 1, wherein the memory stores additional computer-readable instructions, which, when executed by the at least one processor, cause the computing platform to: Receive data from the first user computing device via the communication interface instructing the selection of the first spatial model from the sorted list of spatial models; and In response to receiving an instruction to select the data of the first spatial model from the sorted list of the spatial models, the visual rendering of the first spatial model is generated.

10. The computing platform of claim 9, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to: Receive data indicating user modifications to the first spatial model from the first user computing device via the communication interface; and Based on the data received indicating user modifications to the first spatial model, the machine learning engine executing on the computing platform is updated.

11. The computing platform of claim 9, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to: Receive data from the first user computing device via the communication interface, indicating a request to export the first spatial model to the design tool; In response to receiving the data indicating a request to export the first spatial model to the design tool, one or more drawing files are generated based on the first spatial model; as well as The one or more drawing files generated based on the first spatial model are sent to the first user computing device via the communication interface.

12. The computing platform of claim 1, wherein the memory stores additional computer-readable instructions, which, when executed by the at least one processor, cause the computing platform to: Receive second space program data, identifying one or more parameters of the second physical space, from the second user computing device via the communication interface; A second geometric model is loaded from the database storing the one or more geometric models, wherein the second geometric model includes information defining a second plurality of design rules; Based on the second spatial program data and the second geometric model that identify the second physical space, a second plurality of spatial models are generated for the second physical space; The second plurality of spatial models generated for the second physical space are scored based on the second geometric model, wherein scoring the second plurality of spatial models generated for the second physical space generates a score for each of the second plurality of spatial models. The second plurality of spatial models generated for the second physical space are sorted based on the score of each of the second plurality of spatial models, wherein sorting the second plurality of spatial models generated for the second physical space produces a second sorted list of spatial models; Generate second user interface data that includes a second sorted list of the spatial model; as well as Sending second user interface data, including a second sorted list of the spatial model, to the second user computing device via the communication interface, wherein sending the second user interface data, including a second sorted list of the spatial model, to the second user computing device causes the second user computing device to display a user interface including at least a portion of the second sorted list of the spatial model.

13. The computing platform of claim 1, wherein the memory stores additional computer-readable instructions, which, when executed by the at least one processor, cause the computing platform to: The user interface data, comprising a sorted list of the spatial model, is sent to the first user computing device via the communication interface, wherein sending the user interface data, comprising a sorted list of the spatial model, to the first user computing device causes the first user computing device to display a user interface comprising at least a portion of the sorted list of the spatial model.

14. A data processing method, comprising: At a computing platform that includes at least one processor, communication interface, and memory: The at least one processor receives first space program data identifying one or more parameters of the first physical space from the first user computing device via the communication interface; The first geometric model is loaded from a database storing one or more geometric models by the at least one processor, wherein the first geometric model includes information defining a first plurality of design rules; The at least one processor generates a first plurality of spatial models for the first physical space based on the first spatial program data, which identifies the first physical space, and the first geometric model; The at least one processor scores the first plurality of spatial models generated for the first physical space based on the first geometric model, wherein scoring the first plurality of spatial models generated for the first physical space generates a score for each of the first plurality of spatial models. The first plurality of spatial models generated for the first physical space are sorted by the at least one processor based on the score of each of the first plurality of spatial models, wherein sorting the first plurality of spatial models generated for the first physical space produces a sorted list of spatial models; User interface data, including a sorted list of the spatial model, is generated by the at least one processor; as well as The at least one processor sends a visual rendering of a first spatial model selected from a sorted list of spatial models to the first user computing device via the communication interface, wherein sending the visual rendering of the first spatial model to the first user computing device causes the first user computing device to display a user interface including at least a portion of the visual rendering of the first spatial model.

15. The data processing method of claim 14, wherein receiving the first space program data identifying the one or more parameters of the first physical space includes receiving information identifying architectural details of the first physical space, organizational details of the first physical space, working style details of the first physical space, and budget details of the first physical space.

16. The data processing method of claim 14, wherein loading the first geometric model from the database storing the one or more geometric models comprises selecting the first geometric model from a plurality of geometric models generated by the computing platform using a machine learning engine trained on one or more best-in-class spatial designs.

17. The data processing method of claim 14, wherein loading the first geometric model from the database storing the one or more geometric models comprises selecting the first geometric model based on first spatial program data identifying the first physical space.

18. The data processing method of claim 14, wherein generating the first plurality of spatial models for the first physical space based on the first spatial program data identifying the first physical space and the first geometric model comprises: Generate multiple block models for the first physical space; The plurality of block models generated for the first physical space are scored based on the first geometric model, wherein scoring the plurality of block models generated for the first physical space generates a score for each of the plurality of block models. A subset of the multiple block models is selected based on the score of each of the multiple block models; Generate multiple setup models for the first physical space, wherein each of the multiple setup models corresponds to a specific block model in the subset of the multiple block models; The plurality of setup models generated for the first physical space are scored based on the first geometric model, wherein scoring the plurality of setup models generated for the first physical space generates a score for each of the plurality of setup models. A subset of the plurality of setting models is selected based on the rating of each of the plurality of setting models; Generate multiple furniture models for the first physical space, wherein each of the multiple furniture models corresponds to a specific setting model in the subset of the multiple setting models; The plurality of furniture models generated for the first physical space are scored based on the first geometric model, wherein scoring the plurality of furniture models generated for the first physical space generates a score for each of the plurality of furniture models. as well as A subset of the plurality of furniture models is selected based on the rating of each of the plurality of furniture models, wherein the subset of the plurality of furniture models corresponds to the first plurality of space models generated for the first physical space.

19. The data processing method of claim 18, wherein each of the plurality of block models indicates the potential location of different neighborhoods in the first physical space, each of the plurality of setting models indicates the potential location of different work settings in the first physical space, and each of the plurality of furniture models indicates the potential location of different furniture items in the first physical space.

20. A non-transitory computer-readable medium storing one or more instructions, said instructions causing the computing platform, when executed by a computing platform including at least one processor, a communication interface, and a memory, to: Receive first space program data, identifying one or more parameters of the first physical space, from the first user computing device via the communication interface; Load a first geometric model from a database that stores one or more geometric models, wherein the first geometric model includes information defining a first plurality of design rules; Based on the first spatial program data identifying the first physical space and the first geometric model, a first plurality of spatial models are generated for the first physical space; The first plurality of spatial models generated for the first physical space are scored based on the first geometric model, wherein scoring the first plurality of spatial models generated for the first physical space generates a score for each of the first plurality of spatial models. The first plurality of spatial models generated for the first physical space are sorted based on the score of each of the first plurality of spatial models, wherein sorting the first plurality of spatial models generated for the first physical space produces a sorted list of spatial models; Generate user interface data that includes a sorted list of the spatial model; as well as A visual rendering of a first spatial model selected from a sorted list of spatial models is sent to the first user computing device via the communication interface, wherein sending the visual rendering of the first spatial model to the first user computing device causes the first user computing device to display a user interface including at least a portion of the visual rendering of the first spatial model.

Images