Meta heuristics based optimal maintenance and repair policies for cement industry

Abstract

In the present competitive market environment, it is imperative for organizations seeking to achieve performance excellence, to continuously enhance their capability to improve the cost-effectiveness of their operations. In this study, a maintenance decision model to determine the optimal maintenance and repair policy which would minimize the total discounted maintenance cost which includes both the scheduled and unscheduled maintenance and repair costs is proposed. Several maintenance alternatives, such as simple maintenance, preventive replacement, different levels of improvement in Mean Time Between Failure (MTBF) and Mean Time To Repair (MTTR) of the critical subsystems have been considered. The application of the proposed maintenance framework is demonstrated with a real life Raw-Mill System (RMS) of a cement industry. Since, the number of maintenance alternatives considered in the study is very large, this maintenance optimization problem is a complex combinatorial optimization problem, and therefore the use of meta-heuristics is inevitable. In this study an attempt is made to explore the application of meta-heuristics such as the Genetic Algorithm (GA), Simulated Annealing (SA), and Memetic Algorithm (MA) to solve this maintenance optimization problem. The Monte Carlo simulation model is used to compute the present value of the total maintenance cost during the planning horizon. The results obtained are compared based on the objective function value and computational efficiency. Sensitivity analysis is carried out to study the effect of variations in the model parameters on the proposed maintenance model. newline newline newline