A bio-inspired simulation-based optimization framework for multi-objective optimization

Abstract

Owing to the complexity of real-world problems, optimization of these problems normally involves multiple objectives to be optimized at the same time. These objectives are frequently non-commensurable and competing. Very often, a number of trade-off solutions rather than a single solution are found in multi-objective optimization problems. In the absence of any further preference information such as rankings of the objectives, all these solutions are optimal in a wider sense that no other solutions in the search space are better than them when all objectives have been considered and none of these solutions can be said to be better than the others. For example, material handling is a vital element of industrial processes, which requires a wide range of manual, semi-automated and automated equipment, as well as implicates various activities such as the movement, protection, storage and control of materials, products and wastes throughout the processes of manufacturing, warehousing, distribution, consumption, and disposal while having to satisfy multiple objectives. Having efficient and effective material handling systems are of great importance to various industries, such as manufacturing and logistics industries, for maintaining and facilitating a continuous flow of materials through the workplace and guaranteeing that required materials are available when needed. For this reason, the capabilities of analyzing and evaluating these systems that involve the use of material handling systems and the presence of uncertainty and randomness, is essential for manufacturing and distribution practitioners. In this respect, computer-based discrete-event modeling and simulation methodologies together with optimization have emerged as a very useful tool to facilitate the effective analysis of these complex operations and systems. In this study, a multi-objective simulation-based optimization framework is applied. The framework consists of a newly developed hybrid immune-inspired optimization algorithm named Suppression-controlled Multi-objective Immune Algorithm (SCMIA) and a simulation model for solving real-life multi-objective simulation-based optimization problems. The proposed multi-objective optimization algorithm is largely developed based on biological immune system concepts and it hybridizes the biological immune system concepts, namely, clonal selection principle and immune network theory with an idea derived from the biological evolution. The reason behind such hybridization is to further improve the diversity of the clone population and the convergence of the algorithm. In this research, a set of numerical experiments are carried out for evaluating the performance of the proposed algorithm by the use of several benchmark problems. The proposed algorithm is also applied together with simulation modeling technique to two multi-objective simulation-based problems on material handling systems in order to demonstrate the applicability of the proposed algorithm in real-life cases. The results indicate that the proposed algorithm outperforms the other benchmark algorithms particularly in population diversity for most cases. The results also reveal that the simulation-based optimization framework is capable of solving large scale real-life problems with a large number of input parameters involved.