Design Assessment Ecosystem for Automated Requirement Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current engineering design tools face challenges in automatically assessing designs against engineering requirements, particularly in verifying complex system requirements and integrating disparate models across distributed development environments, which hinders productivity and accuracy in design evaluation and collaboration.
Innovation Solution
The development of an ecosystem for design assessment and verification that integrates with CAD systems, utilizing machine learning and digital assistants to provide automatic requirement verification, enabling efficient user interaction, nonlinear analysis of civil structures, and facilitating collaboration across distributed teams through a plug-and-play approach for modeling engineering structures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automatic requirement verification is implemented using machine learning, then productivity of design instructors and engineering departments is improved, but device complexity increases
Solution Approach 1:
A digital assistant acts as an intermediary between instructors and the complex machine learning-based verification system. The digital assistant handles user interactions, simplifies complex operations, and manages the interface between human users and the automated requirement verification engine, thereby improving productivity without exposing users to system complexity.
Solution Approach 2:
The system implements self-service capabilities where the automatic requirement verification engine autonomously assesses student designs against ABET learning outcomes without requiring manual configuration or intervention. The system automatically parses deliverables, identifies contributions, and generates assessment reports, reducing the need for instructor involvement in routine verification tasks.
2Measurement precision
If machine learning is used to assess student deliverables, then measurement precision of individual student contributions is improved, but difficulty of detecting and measuring increases
Solution Approach 1:
The patent replaces manual mechanical assessment processes with machine learning-based automated analysis. The system uses ML algorithms to parse and evaluate student deliverables, automatically measuring individual contributions to team projects. This substitution enables precise measurement of student work without the difficulty of manual detection and evaluation.
Solution Approach 2:
The system creates digital copies of student deliverables and assessments, enabling automated analysis and measurement. By converting physical or semi-structured student work into structured digital data, the system can apply machine learning algorithms to accurately measure and compare individual contributions across multiple students and projects.
3Ease of operation
If automatic requirement verification is integrated with CAD systems, then ease of operation is improved, but device complexity increases
Solution Approach 1:
The patent merges the automatic requirement verification functionality directly into existing CAD systems through plugin architecture. By integrating verification tools within the familiar CAD environment, the system improves ease of operation as users can perform verification without switching between separate tools, while the underlying complexity is managed through modular integration.
Solution Approach 2:
The verification system is designed with universal functionality that can operate across multiple CAD platforms and design types. The plugin architecture provides multi-functional capabilities that work with various CAD systems, reducing the need for separate specialized tools and simplifying the user experience while managing complexity through standardized interfaces.
4Ease of manufacture
If plug-and-play approach is used for modeling engineering structures, then ease of manufacture is improved, but device complexity increases
Solution Approach 1:
The patent implements segmentation by dividing complex engineering models into modular, reusable components that can be independently created and assembled. The plug-and-play approach allows individual model elements to be developed separately and then integrated into larger systems, simplifying the manufacturing and deployment process while managing complexity through modular architecture.
Solution Approach 2:
The system enables ease of manufacture through parameter-based modeling where engineering structures are defined by adjustable parameters rather than fixed geometries. This allows models to be easily modified and adapted by changing parameters, facilitating rapid customization and assembly of engineering systems without requiring complex manual reconfiguration.
Data Source
AI summary
1. The invention presents a mechanism for assessing student designs against learning outcomes from the Accreditation Board for Engineering and Technology. Changes in formulation of the learning outcomes can be simply accommodated by changing association between the learning outcomes and performance indicators, but keeping assessment rubrics the same.2. We present a plugin for automatically verifying engineering requirements formulated in the SysML system modeling language. The plugin employs a generic verification method consisting of check points systematically positioned along the function chain.3. Smart Modeling and Simulation is an automated system for creating simulation models, performing rapid simulations and interactively interpreting the results. It can account for complex multi-physics interactions not traditionally supported by commercial analysis software.4. We also present a system for efficient non-linear analysis of civil engineering structures, one accounting for complex interactions, and yet offering an intuitive interface based on a minimal set of assumptions.