A multi-objective production smoothing model with compressible operating times

• Published in: Applied mathematical modelling 2011-07, Vol.35 (7), p.3596-3610
• Main Authors: Karimi-Nasab, M., Aryanezhad, M.B.
• Format: Article
• Language: English
• Subjects: Applied sciences; Available time; Compressible operating time; Cost engineering; Decision theory. Utility theory; Economics; Exact sciences and technology; Genetic algorithms; Inventory control, production control. Distribution; Marketing; Mathematical analysis; Mathematical models; Multi-objective; Multi-period; Multi-product; Operational research and scientific management; Operational research. Management science; Production planning; Production smoothing; Smoothing
• ISSN: 0307-904X
• DOI: 10.1016/j.apm.2011.01.038

Real life multi-product multi-period production planning often deals with several conflicting objectives while considering a set of technological constraints. The solutions of these problems can provide deeper insights to the decision makers/managers than those of single-objective problems. Some managers want to use from a production plan that is corresponding to minimum change in production policy along with minimum total cost simultaneously as possible. On the other hand, these two objectives have intrinsic conflicts such that producing in a fixed rate will cause huge costs than producing economically or according to JIT. So this paper presents a novel multi-objective model for the production smoothing problem on a single stage facility that some of the operating times could be determined in a time interval for. The model is to: ( a) smooth the variations of production volume, and ( b) minimize total cost of the corresponding production plan, while satisfying a set of technological constraints such as limited available time. The proposed model is developed in a real case study and is solved by a new genetic algorithm. The proposed genetic algorithm uses a novel achievement function for exploring the solution space, based on LP-metric concepts. Computational experiences reveal the sufficiency and efficiency of the proposed approach in contrast to previous researches.