Search

A parallel Simulated Annealing algorithm with multi-threaded architecture is proposed to solve a real bi-objective maintenance scheduling problem with conflicting objectives: the minimisation of the total equipment downtime caused by maintenance jobs and the minimisation of the multi-skilled workforce requirements over the given horizon. The maintenance jobs have different priorities with some precedence relations between different skills. The total weighted flow time is used as a scheduling criterion to measure the equipment availability. The multi-threaded architecture is used to speed up a multi-objective Simulated Annealing algorithm to solve the considered problem. Multi-threading is a form of parallelism based on shared memory architecture where multiple logical processing units, so-called threads, run concurrently and communicate via shared memory. The performance of the parallel method compared to the exact method is verified using a number of test problems. The obtained results imply the high efficiency and robustness of the proposed heuristic for both solution quality and computational effort. [ABSTRACT FROM AUTHOR]

In this paper, a multi-objective integer programming approach is developed to investigate the impact of the use-based preventive maintenance (UPM) policy on the performance of the cellular manufacturing system (CMS). Under the UPM policy a maintenance schedule is established which provides for the performance of preventive maintenance (PM) only after a predetermined number of operating hours of machine use. This research indicates how PM and failure repair (FR) actions affect the effective availability of the machines and accordingly the machine and inter/intra-cell material handling costs under the UPM policy. The objective is to minimise the machine cost, inter- and intra-cell material handling and PM/FR costs. The proposed model is solved by an interactive fuzzy programming (IFP) approach to determine the best compromise solution from the decision maker point of view. IFP assumes that each objective function has a fuzzy goal and focuses on minimising the worst upper bound to obtain an efficient solution which is close to the best lower bound of each objective function. Compromise solutions are prioritised by two efficiency criteria, i.e. grouping efficiency and system availability. The performance of the proposed model is verified by a comprehensive numerical example. [ABSTRACT FROM AUTHOR]