Fault-tolerance Optimization of Reconfigurable Manufacturing Systems: Multi-objective Artificial Bee Colony Algorithm for Process Planning

Abstract

A reconfigurable manufacturing system (RMS) is a modern manufacturing system class that aims to provide functionality and capacity as and when needed. Reconfigurable machine tools (RMTs) are the main component of RMS, which have a modular structure, such that they can be reconfigured into different configurations. In RMSs, process planning aims to assign to each operation the appropriate configuration. During the fulfillment of operations along the process plan, RMTs may defect. Monitoring the state of RMS during process planning allows for diagnosing the defect of broken-down components. After detecting the defect, the repairing team of the company is responsible for repairing this defect. Furthermore, periodic maintenance for the system components is required for facilitating the monitoring of RMS. The accomplishment of all steps of each process, i.e., monitoring, maintenance, and repairing, requires time and a specific cost. This paper deals with the problem of maximizing fault-tolerance optimization in RMSs. A multi-objective optimization model is proposed considering the minimum total time and minimum total cost of process plans incorporating repair time of configurations and maintenance cost of configurations, during the process planning. This problem represents an NP-hard problem, for solving it, the multi-objective artificial bee colony (MOABC) algorithm is adapted to generate a set of non-dominated solutions (i.e., process plans). With the help of a numerical example, the results obtained by MOABC are compared with a well-known meta-heuristic algorithm, called non-dominated Sorting Genetic Algorithm-II (NSGA-II), and demonstrated the superiority of MOABC over the NSGA-II in terms of solution quality, convergence, and solutions' diversity.