Abstract: In various embodiments, a computer-implemented method for determining compositions of buildings includes receiving a plurality of building images associated with a building, generating a conditional image based on the plurality of building images, generating a plurality of tokens characterizing the building, providing the conditional image and the plurality of tokens to a neural network to cause the neural network to generate an output, generating, via a generative artificial intelligence (AI) model, a structural floorplan of the building based on the output and the plurality of tokens, and determining a composition of the building based on the structural floorplan.

Abstract: One embodiment sets forth a technique for completing computerized representations of physical structures using knowledge graphs. According to some embodiments, the technique includes the steps of generating a knowledge graph that characterizes relationships between various features of computerized representations of physical structures; training one or more machine learning models based on the knowledge graph; receiving a request for adding a selected feature to a computerized representation of a physical structure; generating predicted feature data using the one or more trained machine learning models and the request; and causing the predicted feature data to be rendered at a graphical user interface (GUI) to suggest a predicted feature to be included with the selected feature. Another embodiment sets forth a technique for training machine learning models using knowledge graphs associated with computerized representations of physical structures.

Abstract: One embodiment sets forth a technique for completing computerized representations of physical structures using knowledge graphs. According to some embodiments, the technique includes the steps of generating a knowledge graph that characterizes relationships between various features of computerized representations of physical structures; training one or more machine learning models based on the knowledge graph; receiving a request for adding a selected feature to a computerized representation of a physical structure; generating predicted feature data using the one or more trained machine learning models and the request; and causing the predicted feature data to be rendered at a graphical user interface (GUI) to suggest a predicted feature to be included with the selected feature. Another embodiment sets forth a technique for training machine learning models using knowledge graphs associated with computerized representations of physical structures.

Abstract: An urban design pipeline automatically generates design options for an urban design project based on a financial model. The urban design pipeline includes a geometry engine that generates candidate designs to meet various design criteria and design objectives associated with the urban design project. The urban design pipeline also includes an evaluation engine that evaluates the candidate designs based on the financial model. The financial model quantifies the financial performance of any given candidate design. The urban design pipeline operates iteratively to improve the candidate designs until one or more convergence criteria are met, thereby generating a set of design options with acceptable financial performance.

Abstract: Techniques are disclosed for designing manufacturing facilities. A design application imports a computer-aided design (CAD) model and metadata associated with a manufacturing facility to generate a data set that specifies geometric and manufacturing constraints of the manufacturing facility. The design application performs optimization operations based on the data set to identify one or more high-performing designs that assign components to docks in the manufacturing facility. The optimization operations can include genetic optimization operations that generate multiple generations of designs, each of which is evaluated based on a travel distance, congestion, and number of turns associated with paths traversed by components for the design.

Abstract: Techniques are disclosed for designing manufacturing facilities. A design application imports a computer-aided design (CAD) model and metadata associated with a manufacturing facility to generate a data set that specifies geometric and manufacturing constraints of the manufacturing facility. The design application performs optimization operations based on the data set to identify one or more high-performing designs that assign components to docks in the manufacturing facility. The optimization operations can include genetic optimization operations that generate multiple generations of designs, each of which is evaluated based on a travel distance, congestion, and number of turns associated with paths traversed by components for the design.

Abstract: A design generation and simulation workflow for generating design layouts and executing simulations of the design layouts to generate energy and comfort performance metrics that indicate the energy efficiency and occupant-comfort levels associated with the design layouts. As described herein, the design workflow includes five phases including 1) configuring a design generation application, 2) generating design layouts via the design generation application, 3) generating energy models for the design layouts, 4) configuring simulations on the energy models, and 5) executing the configured simulations on the energy models and generating comparative results based on the simulation results of multiple simulations.

Abstract: A design generation and simulation workflow for generating design layouts and executing simulations of the design layouts to generate energy and comfort performance metrics that indicate the energy efficiency and occupant-comfort levels associated with the design layouts. As described herein, the design workflow includes five phases including 1) configuring a design generation application, 2) generating design layouts via the design generation application, 3) generating energy models for the design layouts, 4) configuring simulations on the energy models, and 5) executing the configured simulations on the energy models and generating comparative results based on the simulation results of multiple simulations.

Abstract: Techniques for generative design based on large language models include receiving a plurality of design examples; evaluating each of the design examples using performance metrics to generate corresponding design attributes for each of the design examples; storing the design examples in a design grid as initial candidate design layouts at a location in the design grid based on the corresponding design attributes; selecting one or more candidate design layouts from the design grid as parent candidate design layouts; generating a new candidate design layout from the parent candidate design layouts; evaluating the new candidate design layout using the performance metrics to generate new design attributes; storing the new candidate design layout in the design grid based on the new design attributes; and generating training data for a large language model based on the candidate design layouts in the design grid and the corresponding design attributes for the candidate design layouts.

Abstract: Techniques for generative design based on large language models include receiving one or more design prompts; generating, based on the one or more design prompts, a plurality of design tokens using a large language model; generating a conceptual design layout from the plurality of design tokens; and optimizing the conceptual design layout to generate a compliant design layout.

Abstract: In various embodiments, a generative design application generates and evaluates automotive designs. In operation, the generative design application computes a first set of metric values based on a set of metrics associated with design goal(s) and a first set of parameter values for a parameterized automobile model. The generative design application then performs optimization operation(s) on the first set of parameter values based on the first set of metric values to generate a second set of parameter values. Subsequently, the generative design application generates at least one design based on the second set of parameter values that is more convergent with respect to at least one of the design goals than a previously generated design. Advantageously, less time and effort are required to generate and evaluate multiple designs and then optimize those designs relative to more manual prior art approaches.

Abstract: An urban design pipeline automatically generates design options for an urban design project. The urban design pipeline includes a geometry engine and an evaluation engine. The geometry engine analyzes design criteria and design objectives associated with the urban design project and then generates numerous candidate designs that meet the design criteria and optimize the design objectives to varying degrees. The evaluation engine evaluates each candidate design to generate a set of metrics. The geometry engine modifies the candidate designs based on corresponding metrics to generate candidate designs that better meet the design criteria and more effectively achieve the design objectives.

Abstract: An urban design pipeline is configured to automatically generate design options that meet competing design objectives. A geometry engine within the urban design pipeline generates candidate designs for an urban design project. An evaluation engine within the urban design pipeline evaluates the degree to which each candidate design addresses the competing design objectives to produce a set of design metrics. A tradeoff engine within the urban design pipeline generates a design tradeoff space based on the candidate designs and corresponding design metrics. The tradeoff engine traverses the design tradeoff space based on performance modifications to adjust the degree to which the competing design objectives are addressed. The performance modifications can be obtained from any number of stakeholders in the urban design project.

Abstract: An urban design pipeline automatically generates design options for an urban design project. The urban design pipeline includes a geometry engine and an evaluation engine. The geometry engine analyzes design criteria, design objectives, and design heuristics associated with the urban design project and then generates numerous candidate designs. The design criteria specify a property boundary associated with a region of land to be developed. The design objectives indicate a specific type of topology that is derived from an existing urban layout. The design heuristics include different sets of construction rules for generating designs with specific types of topologies. The geometry engine generates candidate designs that conform to the property boundary and have topological characteristics in common with the existing urban layout.

Abstract: One embodiment of the present invention sets forth a technique for generating a modular design for a construction project. The technique includes determining a first set of candidate designs for the construction project. The technique also includes for each candidate design included in the first set of candidate designs, generating a set of design options based on one or more portions of the candidate design and determining a set of performance metrics and a set of attributes associated with the set of design options. The technique further includes generating a second set of candidate designs for the construction project based on the sets of design options associated with the first set of candidate designs, the sets of performance metrics associated with the sets of design options, and the sets of attributes associated with the sets of design options.

Abstract: An urban design pipeline automatically generates design options for an urban design project. The urban design pipeline includes a geometry engine and an evaluation engine. The geometry engine analyzes design criteria, design objectives, and design heuristics associated with the urban design project and then generates numerous candidate designs. The design criteria specify a property boundary associated with a region of land to be developed. The design objectives indicate a specific type of topology that is derived from an existing urban layout. The design heuristics include different sets of construction rules for generating designs with specific types of topologies. The geometry engine generates candidate designs that conform to the property boundary and have topological characteristics in common with the existing urban layout.

Abstract: A computer-implemented method for generating a building design for a building site having a complex shape comprises: generating a first group of processing areas from a plurality of processing areas associated with the building design and a second group of processing areas from the plurality of processing areas; generating a first building footprint leg that is sized to include the first group of processing areas and a second building footprint leg that is sized to include the second group of processing areas; at a pivot location, joining the first building footprint leg and the second building footprint leg at an aperture angle; performing a numerical simulation to position the first group of processing areas in the first building footprint leg and the second group of processing areas in the second building footprint leg to generate a candidate building design for the building site; and rendering the candidate building design.

Abstract: A computer-implemented method for generating a building design for a building site having a complex shape comprises: generating a first group of processing areas from a plurality of processing areas associated with the building design and a second group of processing areas from the plurality of processing areas; generating a first building footprint leg that is sized to include the first group of processing areas and a second building footprint leg that is sized to include the second group of processing areas; at a pivot location, joining the first building footprint leg and the second building footprint leg at an aperture angle; performing a numerical simulation to position the first group of processing areas in the first building footprint leg and the second group of processing areas in the second building footprint leg to generate a candidate building design for the building site; and rendering the candidate building design.

Abstract: An urban design pipeline automatically generates design options for an urban design project. The urban design pipeline includes a geometry engine and an evaluation engine. The geometry engine analyzes design criteria, design objectives, and design heuristics associated with the urban design project and then generates numerous candidate designs. The design criteria specify a property boundary associated with a region of land to be developed. The design objectives indicate a specific type of topology that is derived from an existing urban layout. The design heuristics include different sets of construction rules for generating designs with specific types of topologies. The geometry engine generates candidate designs that conform to the property boundary and have topological characteristics in common with the existing urban layout.

Abstract: An urban design pipeline is configured to automatically generate design options that meet competing design objectives. A geometry engine within the urban design pipeline generates candidate designs for an urban design project. An evaluation engine within the urban design pipeline evaluates the degree to which each candidate design addresses the competing design objectives to produce a set of design metrics. A tradeoff engine within the urban design pipeline generates a design tradeoff space based on the candidate designs and corresponding design metrics. The tradeoff engine traverses the design tradeoff space based on performance modifications to adjust the degree to which the competing design objectives are addressed. The performance modifications can be obtained from any number of stakeholders in the urban design project.