🤖 AI Summary
This study addresses key challenges in engineering problem-solving—ambiguous requirements, difficulty in cross-domain mapping, weak multimodal data integration, and inefficient exploration of solution spaces—by proposing “Generative Optimization,” a novel paradigm. Methodologically, it synergistically integrates generative AI (e.g., large language models, diffusion models, multimodal generators) with classical optimization techniques (e.g., gradient-based and evolutionary algorithms): GenAI drives requirement inference, cross-domain semantic bridging, multimodal input understanding, and large-scale solution generation, while optimization modules ensure constraint satisfaction and solution precision. We establish, for the first time, a systematic conceptual framework, elucidate complementary mechanisms between generative and optimization components, and propose hybrid algorithm design principles and critical research directions. Empirical validation across multiple engineering domains demonstrates the paradigm’s scalability, interpretability, and robustness, offering a principled pathway toward AI-augmented engineering decision-making.
📝 Abstract
The field of engineering is shaped by the tools and methods used to solve problems. Optimization is one such class of powerful, robust, and effective engineering tools proven over decades of use. Within just a few years, generative artificial intelligence (GenAI) has risen as another promising tool for general-purpose problem-solving. While optimization shines at finding high-quality and precise solutions that satisfy constraints, GenAI excels at inferring problem requirements, bridging solution domains, handling mixed data modalities, and rapidly generating copious numbers of solutions. These differing attributes also make the two frameworks complementary. Hybrid generative optimization algorithms present a new paradigm for engineering problem-solving and have shown promise across a few engineering applications. We expect significant developments in the near future around generative optimization, leading to changes in how engineers solve problems using computational tools. We offer our perspective on existing methods, areas of promise, and key research questions.