Struggling with ill-defined objectives in engineering design? This paper introduces an Interactive Evolutionary Design System (IEDS) to support designers in the early stages of the design process. The IEDS facilitates design search and exploration, contributing to a better understanding of the problem domain. It emphasizes multiobjective satisfaction in the face of uncertain objectives and relative importance. The system allows for offline processing of information, leading to a redefinition of the design space. This redefinition includes the ability to add or remove objectives, change their relative importance, and reduce variable ranges. The paper describes elements within the IEDS that enable multiobjective information gathering and subsequent design space redefinition. Instead of focusing on multiobjective optimization over a set number of generations, the IEDS promotes a continuous interactive evolutionary search. This approach aims to optimize both the variable and objective spaces relating to the design problem, leading to more effective and adaptable design outcomes. The insights from this research can significantly enhance conceptual design environments.
This paper, published in Evolutionary Computation, fits squarely within the journal's scope by exploring the application of evolutionary algorithms to design problems. The research extends the field by focusing on interactive search and exploration, which contributes to the journal's ongoing discussion of innovative computational methods for optimization and design.