Multi-objective optimisation of engineering design by using Excel

Abstract

The different objectives of engineering design often conflict with one another such as achieving high product quality while reducing its cost, for example. The design of engineering systems usually involves, in addition to the technical and economic aspects, other aspects that also have to be considered such as the safety, environmental, and social factors. The task of finding acceptable trade-offs between the different objectives is made even more difficulty by the numerous available options and the increasing complexity of the systems themselves. Therefore, the engineers of today need proper tools to help them take the different design factors and options into consideration. Although such tools have been made available by the recent development of evolutionary multi-objective optimisation (MOO), unfortunately, the evolutionary MOO software are expensive to develop and, therefore, they are unavailable or unaffordable to most engineers and engineering students particularly in developing countries. This paper presents a technique for conducting MOO analyses by integrating Excel’s single-objective optimisation (SOO) solver with the TOPSIS decision-making method. The Solver-TOPSIS technique (STT) is illustrated by considering the case of a tip-loaded concrete beam provided by Deb [1] who also provided optimised results obtained by using the widely-used NSGA-II algorithm [2]. The results of the technique are also compared to those obtained by using the more recent MOO solver MIDACO [3]. Finally, the paper demonstrates the capability of the technique to deal with MOO analyses of engineering design with predefined preference of the objectives involved.