Higher and Lower-level Innovizations for Knowledge Discovery in Practical Problems

Innovization (innovation through optimization) is a relatively new concept in the field of multi-objective engineering design optimization. It involves the use of the Pareto-optimal solutions of a problem, to unveil hidden mathematical relationships between variables, objectives and constraint functions. Since the relationships themselves come from the Pareto-optimal dataset, they can be thought of as essential properties which make a feasible solution Pareto-optimal. This paper proposes and discusses two major extensions to the basic concept of innovization, namely higher-level innovization and lower-level innovization. While the former deals with the discovery of common features among solutions from different fronts, the latter concerns commonalities among the trade-off solutions that belong to a specified (or preferred) part of the Pareto-optimal front. The knowledge of such lower-level information is extremely beneficial to a decision maker, since it focuses on a set of preferred designs. On the other hand, higher-level innovization reveals interesting knowledge about the general problem structure. Neither of these crucial aspects concerning multi-objective designs have been addressed before to the authors’ knowledge. In this study, generic methodologies for handling both levels of innovization are developed by extending authors’ earlier automated innovization algorithm. These methodologies are clearly illustrated on three well-known engineering design problems. Results demonstrate that the proposed methodologies are generic and are ready to be applied to other engineering design problems.