Stochastic production routing problem for perishable products: Modeling and a solution algorithm

• Published in: Computers & operations research Vol. 142; p. 105725
• Main Authors: Mousavi, Razieh, Bashiri, Mahdi, Nikzad, Erfaneh
• Format: Journal Article
• Language: English
• Published: New York Elsevier Ltd 01.06.2022; Pergamon Press Inc
• Subjects: Algorithms; Freshness; Mathematical analysis; Mathematical models; Matheuristic algorithm; Numerical analysis; Operations research; Pandemic; Perishable product; Production routing problem; Shelf life; Supply chains; Transportation management; Production routing problem; Pandemic; Freshness; Matheuristic algorithm; Perishable product
• ISSN: 0305-0548; 1873-765X; 1873-765X; 0305-0548
• DOI: 10.1016/j.cor.2022.105725

The freshness of perishable products, in addition to other economic aspects such as production, inventory, and transportation management, is one of the main challenges of food supply chains. Therefore, integration of production, inventory, and routing decisions is essential. In this study, a new production routing model for perishable products with uncertain demand is presented. The aim is to minimize the costs of production, inventory, routing, wasted products, and penalties for non-fresh products. The model is more applicable for perishable products with limited, discrete shelf life with a high freshness value. A five-phase matheuristic algorithm is proposed to solve the stochastic mathematical model. Computational experiments show that the proposed mathematical model can result in a significant reduction of wasted products, particularly when consumer buying patterns change due to various occurrences, such as a pandemic. Also, numerical analysis for small, medium, and large instances confirms the validity and efficiency of the proposed matheuristic algorithm when compared with an exact solver. •Proposing a two-stage stochastic programming model of production routing problem for perishable products.•Considering the wastage and freshness of perishable products in model.•Development of a five phases matheuristic algorithm to solve the considered problem.•Performing computational experiments on pandemic and normal situation instances.