Reliability Modeling of Automated Guided Vehicles in Cellular Manufacturing Systems: A Non-Dominated Sorting Cuckoo Search (NSCS) algorithm

Reliability of the cellular manufacturing systems and automated guided vehicles (AGVs) in the production systems has been recently become one of the most challenging issues. This paper focuses on proposing a multi-objective mathematical model in order to optimize three objectives including the minimization of production costs, minimization of the intracellular and intercellular part transportations and finally maximization of the reliability of AGVs. The failure rates of intracellular AGVs at any of the cells follow either an exponential or a Weibull distribution whereas, this rate is considered to be constant for the intercellular AGVs. As reliability calculation in the Weibull case is very hard (if not impossible), a simulation approach is applied to estimate the reliability of system in this case. It is assumed that all the intracellular and intercellular transportations are done through the AGVs. This means that the production system halts when all AGVs are failed. In order to validate the proposed model, some numerical examples are generated and solved by implementation of a non-dominated sorting cuckoo search (NSCS) and a multi-objective invasive weeds optimization (MOIWO) algorithm. Finally, a hybrid analytic hierarchy process and TOPSIS method (AHP-TOPSIS) is utilized to select the better algorithm in terms of some multi-objective metrics, simultaneously.