Demystifying Axiomatic Design: A Fresh Look at Problem Formulation

Axiomatic design stands as a cornerstone for effective engineering, yet its initial step, translating customer needs into first-level functional requirements (FRs), is often misunderstood. This misstep can derail entire projects. Why, then, is such a critical component so frequently brushed aside?

Unpacking First-Level FRs

First-level FRs should be unambiguous and consistent across designers, provided the needs and constraints are the same. The confusion over what constitutes an appropriate FR isn't a minor oversight. It's a systemic issue that reflects a broader lack of clarity in the design framework itself. Nam P. Suh's extensive work, particularly his books 'The Principles of Design,' 'Axiomatic Design Advances and Applications,' and 'Complexity Theory,' delves into these foundational principles.

Crucially, Suh emphasizes that first-level FRs aren't subjective interpretations. Instead, they're objective targets essential for a coherent design strategy. So, why do designers often get it wrong? The paper suggests that intrinsic difficulties and recurring pitfalls in this early phase lead to substantial design setbacks. The ablation study reveals these pitfalls frequently stem from insufficient problem formulation guidance during the design process.

The Role of Large Language Models

In today's rapidly evolving tech landscape, large language models (LLMs) like GPT-3 have transformed the way we approach problem-solving. But can they truly enhance the axiomatic design process? This paper cautiously explores the possibilities and limitations of LLMs in assisting with first-level FR formulation.

While these models can offer valuable insights and generate potential FRs, they lack the nuanced understanding that human designers possess. The key contribution of the paper lies in its argument that while LLMs are beneficial, they can't replace the intuition and experience required for precise FR formulation.

Practical Implications

Designers and engineers must prioritize a deep understanding of first-level FRs if they wish to avoid the pitfalls that lead to failure. With the right guidance, as elucidated by Suh's principles, teams can achieve more reproducible and successful design outcomes. However, the integration of AI tools like LLMs should be seen as complementary, rather than a panacea.

, what's missing isn’t technology but the application of sound design principles. The question remains: Will designers heed this call for clarity, or will the allure of technology overshadow foundational truths?