Automating morphological chart exploration: a multi-objective genetic algorithm to address compatibility and uncertainty

Abstract

A novel approach using a Genetic Algorithm (GA) is presented for extracting globally satisficing (Pareto optimal) solutions from a morphological chart, where the evaluation and combination of 'means to subfunctions' is modelled as a combinatorial multi-objective optimisation problem. A fast and robust GA is developed to solve the resulting optimisation problem. Customised crossover and mutation operators that are specifically tailored to solve this combinatorial optimisation problem are discussed. Proof-of-concept simulation results on industrial design problems, viz. work-cell fixture design and automobile headlamp design, are presented, analysed, and compared with the solutions obtained by the design team. The described genetic algorithm incorporates features to prevent the redundant evaluation of identical solutions and a method for handling the compatibility matrix (feasible/infeasible combinations) and addressing desirable/undesirable combinations. The proposed approach is limited by its reliance on the user-specified metrics for evaluating the objectives and existence of a mathematical representation of the combined solutions. A method motivated from robust optimisation techniques is presented to reduce the effect of these limitations by addressing the sensitivity to variation in the input parameters. The optimisation architecture is designed to be a scalable and flexible procedure which can be easily modified to accommodate a wide variety of design methods that are based on the morphological chart.