An efficient multi-unit production planning strategy based on continuous variables

Abstract

Production planning can significantly enhance the competitiveness and hence plays a crucial role in manufacturing industries. In the production planning problem discussed in this article, a set of processes are available which can be operated at various capacity levels with the production and investment costs being proportional to the production of the processes. The determination of the optimal production plan, in terms of the selection of product to be manufactured, the selection and operating capacity of the manufacturing processes in the presence of multiple resource constraints so as to obtain maximum profit is a combinatorial optimization problem. In this article, we state the limitations of the formulation/strategies employed in literature and propose a multi-unit strategy which utilizes only continuous variables to overcome them. The proposed strategy is generic and demonstrated in the context of production planning in a petrochemical industry that has been previously used to potentially guide the petrochemical industries. The proposed strategy is demonstrated with multiple computational intelligence algorithms viz., artificial bee colony, dynamic neighbourhood learning particle swarm optimizer, multi-population ensemble differential evolution, and sanitized–teaching–learning–based optimization. For the cases discussed in literature, the proposed strategy shows an improvement of up to 12.1% in the profit.