Data Acquisition for Enhancing Human-Informed Topology Optimization

Abstract

This thesis presents a survey application designed for the future development of HumanInformed Topology Optimization (HiTop) towards the deeper integration of optimization and real-world feasibility. Topology optimization produces high-performance designs by optimally distributing material, but its application in professional environments remains limited due to fabrication constraints and inflexible design workflows. To address this, the Carstensen Group developed HiTop, which integrates optimization algorithms with human experience, allowing engineers to modify the computer design based on their professional judgment. Thus, the future development of HiTop requires real-world data on human preferences. This project introduces a web-based survey app integrated with Qualtrics. It presents users with various design scenarios and computer-optimized designs, and records their modifications and reasoning. A preliminary survey collected responses from 13 professionals and engineering students. Preliminary findings suggest that engineers consistently focus on similar regions of interest, even when motivated by different reasons. However, the sample size is too small to make any statistically significant conclusions. While the platform mostly performed as intended, a bug related to data storage was discovered during analysis. The issue has since been resolved, and the tool is now fully functional and ready for broader deployment.