Optimizing Design Circularity: Managing Complexity in Design for Circular Economy Through Single and Multi-Objective Optimisation

Abstract

This paper advances the application of computational optimization to design for circular economy (CE) by comparing results of scalarized single-objective optimization (SOO) and multi-objective optimization (MOO) to a furniture design case study. A framework integrating both methods is put forward based on results of the case study. Existing design frameworks for CE emphasize optimization through an iterative process of manual assessment and redesign (Ellen MacArthur Foundation, 2015). Identifying good design solutions for CE, however, is a complex and time-consuming process. Most prominent CE design frameworks list at least nine objectives, several of which may conflict (Reike et al., 2018). Computational optimization responds to these challenges by automating search for best solutions and assisting the designer to identify and manage conflicting objectives. Given the many objectives outlined in circular design frameworks, computational optimisation would appear a priori to be an appropriate method. While results presented in this paper show that scalarized SOO is ultimately more time-efficient for evaluating CE design problems, we suggest that given the presence of conflicting circular design objectives, pareto-set visualization via MOO can initially better support designers to identify preferences.