Accounting for Changing Customer NeedsWith s-Pareto Frontiers

Abstract

The development of new products often results from natural changes in customer needs. Products which anticipate, account for, and allow for these natural changes through the addition of modules, although challenging for manufacturers and design engineers to develop, capitalize on product commonality to reduce production costs. In this paper, the Pareto-based multiobjective optimization design method previously developed by the authors is extended to satisfy changing needs by using s-Pareto frontiers ‐ sets of non-dominated designs from disparate design models. The novel intent of the presented method is to design a product that moves from one location on the s-Pareto frontier to another, thus satisfying changing customer needs, through the addition of a module. The expanded six-step method is described as follows: (A) Characterize the multiobjective design space. (B) Identify the anticipated regions of interest within the search space based on predicted future needs. (C) Identify the platform design variables that minimize the performance loses due to commonality across the anticipated regions of interest. (D) Assemble the s-Pareto frontier within each region of interest. (E) Determine the values of all design variables for the optimal product design in each region of interest by multiobjective optimization. (F) Identify the module design variables, and identify the platform and module designs by constrained module design. The design of a manual irrigation pump is used to demonstrate the method, and results in the development of a platform and two module designs that adapt the pump to provide specified changes in pump water output verses pump cost. The authors conclude that the developed method provides a new and general framework for selecting platform and module designs in the presence of multiple design models, and is capable of providing a set of designs based on predicted changes in needs.