In this paper, we present a method incorporating axiomatic design, TRIZ, and mixed integer programming (MIP) to solve engineering design problems. Axiomatic design decomposes the problem into several mutually independent sub-problems, TRIZ generates all feasible design concepts, and MIP optimizes cost and the numerical configuration among available design options. The method is illustrated on a locomotive ballast arrangement case study. Ballast arrangement is a key process for a locomotive assembly, which determines the carrying capacity. Due to the unsophisticated technology requirements, the ballast arrangement process has received little attention. The trend of mass customization, however, demands locomotive manufacturers to provide diverse products with affordable cost and reduced time. Thus, a flexible and easy to implement ballast arrangement process design is sought. The proposed method determines what material combinations, in what quantity, and where in the limited cavities should the ballast be allocated to minimize cost. Using the case study, we demonstrate the advantages in cost reduction and time savings. The synergy of these improvements not only can enhance productivity and agility but also competitive advantage.
Read full abstract