MECHBench: A Set of Black-Box Optimization Benchmarks originated from Structural Mechanics

2025-11-13

AI Summary

Existing black-box optimization benchmarks predominantly rely on abstract or synthetic problems, failing to capture the nonsmoothness, strong nonlinearity, and high-fidelity constraints characteristic of real-world engineering applications—thus limiting their practical relevance for algorithm evaluation. Method: We introduce the first engineering-grade black-box optimization benchmark suite tailored to structural mechanics, systematically formalizing real physical problems—such as vehicle crashworthiness—as standardized black-box optimization tasks, integrating high-fidelity simulation with gradient-free optimization paradigms. Contribution/Results: This work pioneers the rigorous, reproducible, and comparable transformation of complex nonsmooth structural mechanics problems into standardized black-box benchmarks. It provides precise problem definitions, a unified API, comprehensive documentation, and tools for analyzing algorithmic search behavior and engineering applicability. The suite significantly enhances the evaluability and practical utility of black-box optimization methods in realistic, computationally demanding engineering scenarios.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Constraint Satisfaction and Optimization: Constraint Optimization
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Benchmarking is essential for developing and evaluating black-box optimization algorithms, providing a structured means to analyze their search behavior. Its effectiveness relies on carefully selected problem sets used for evaluation. To date, most established benchmark suites for black-box optimization consist of abstract or synthetic problems that only partially capture the complexities of real-world engineering applications, thereby severely limiting the insights that can be gained for application-oriented optimization scenarios and reducing their practical impact. To close this gap, we propose a new benchmarking suite that addresses it by presenting a curated set of optimization benchmarks rooted in structural mechanics. The current implemented benchmarks are derived from vehicle crashworthiness scenarios, which inherently require the use of gradient-free algorithms due to the non-smooth, highly non-linear nature of the underlying models. Within this paper, the reader will find descriptions of the physical context of each case, the corresponding optimization problem formulations, and clear guidelines on how to employ the suite.

Problem

• Addresses the gap in real-world engineering benchmarks for black-box optimization
• Proposes structural mechanics-based benchmarks derived from vehicle crashworthiness scenarios
• Provides non-smooth, non-linear problems requiring gradient-free optimization algorithms

Innovation

• Structural mechanics benchmarks for black-box optimization
• Gradient-free algorithms for non-smooth vehicle crashworthiness
• Curated problem sets from real-world engineering applications