A monolithic batch-centric MILP approach for a real-world integrated production and pipeline distribution scheduling problem

Abstract

This paper addresses a real-world, comprehensive production and distribution operations scheduling problem within the context of an integrated phosphate supply chain. The supply chain considered belongs to the Khouribga mine site of the OCP Group. A monolithic batch-centric, continuous-time Mixed-Integer Linear Programming (MILP) model was developed to solve this industrial problem, encompassing a production scheduling problem, a multiproduct pipeline network transportation problem, and a multiproduct storage-sensitive management problem. The objective is to create a scheduling plan for all detailed operations in the integrated supply chain while maximizing the use of pipelines for product transportation and complying with production, transport, and storage constraints, as well as meeting customer demand requirements. The proposed MILP model was evaluated using real dataset instances from OCP’s supply chain; the experimental results demonstrate its efficiency in providing optimal and near-optimal solutions within a reasonable CPU time.