EngiBench: A Framework for Data-Driven Engineering Design Research

TL;DR

EngiBench is an open-source framework with datasets and tools for data-driven engineering design optimization, enabling fair comparisons of algorithms across diverse engineering problems.

Contribution

It introduces EngiBench and EngiOpt, providing a unified, modular platform with datasets and algorithms for reproducible, data-driven engineering design research.

Findings

Demonstrates versatility through experiments across multiple engineering domains.

Reveals challenges for standard machine learning methods on complex design manifolds.

Facilitates fair comparison and reproducibility in engineering optimization research.

Abstract

Engineering design optimization seeks to automatically determine the shapes, topologies, or parameters of components that maximize performance under given conditions. This process often depends on physics-based simulations, which are difficult to install, computationally expensive, and require domain-specific expertise. To mitigate these challenges, we introduce EngiBench, the first open-source library and datasets spanning diverse domains for data-driven engineering design. EngiBench provides a unified API and a curated set of benchmarks -- covering aeronautics, heat conduction, photonics, and more -- that enable fair, reproducible comparisons of optimization and machine learning algorithms, such as generative or surrogate models. We also release EngiOpt, a companion library offering a collection of such algorithms compatible with the EngiBench interface. Both libraries are modular, letting users plug in novel algorithms or problems, automate end-to-end experiment workflows, and leverage built-in utilities for visualization, dataset generation, feasibility checks, and performance analysis. We demonstrate their versatility through experiments comparing state-of-the-art techniques across multiple engineering design problems, an undertaking that was previously prohibitively time-consuming to perform. Finally, we show that these problems pose significant challenges for standard machine learning methods due to highly sensitive and constrained design manifolds.