AI-Based Manufacturing Part Design Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current engineering design of parts is slow, manual, and suboptimal in terms of cost, manufacturability, and quality, requiring iterative and labor-intensive processes for design engineers to achieve desired part designs.
Innovation Solution
An artificial intelligence-based method and system that uses a neural network to optimize part designs by encoding user inputs, comparing them to a space of realized and imagined part designs, and generating an optimal design based on specified objectives and weightings, allowing for automated design feedback and improvement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual iterative design process is used, then design flexibility and control are maintained, but design time and labor cost increase significantly
Solution Approach 1:
An AI-based intermediary system is introduced between the designer and the design space. This system encodes design requirements, searches through encoded part designs using machine learning algorithms, and decodes optimal solutions, thereby automating the iterative search process while maintaining design control through user-defined objectives and constraints.
Solution Approach 2:
The manual mechanical iterative design process is replaced with an automated computational system using machine learning algorithms. The system encodes design parameters, performs automated similarity searches and optimizations in the encoded design space, and generates optimal designs without requiring manual iteration by designers.
2Manufacturing precision
If parametric rules are manually entered into design programs, then design precision can be improved, but labor intensity and skill requirements increase
Solution Approach 1:
The system performs self-service by automatically encoding design requirements, searching the design space, and generating optimized parametric rules without requiring manual input of complex parameters. The machine learning model autonomously processes design objectives and generates precise design solutions, eliminating the need for designers to manually enter complex parametric rules.
Solution Approach 2:
The system transforms design parameters from manual input form to encoded numerical representations that can be processed by machine learning algorithms. By changing the parameter representation and using automated optimization, the system achieves high design precision without requiring manual expertise in parametric modeling.
3Reliability
If conventional design methods are used, then existing design knowledge is utilized, but optimization for cost and manufacturability is insufficient
Solution Approach 1:
The system incorporates feedback loops where design objectives including cost and manufacturability constraints are continuously evaluated during the optimization process. The AI model learns from feedback about manufacturing requirements and adjusts design solutions to optimize both quality and manufacturability, generating designs that are evaluated against multiple criteria including production cost and manufacturing feasibility.
Data Source
AI summary
Systems, methods, and apparatus for artificial intelligence-based manufacturing part design are disclosed. A system for designing a part comprises at least one processor configured: to encode the desired part design to generate an encoded desired part design; to identify a group of part designs within a space that is similar to the desired part design by comparing the encoded desired part design to encoded realized part designs, encoded imagined part designs, real metadata, and imagined metadata within the space; to generate an encoded optimal part design by analyzing the group of part designs according to objectives and weightings provided by a user; and to decode the encoded optimal part design to generate an optimal part design. Further, the system comprises a display configured to display, to the user, the optimal part design, which the user may use as a guide to modify the desired part design accordingly.