Loading...

A robust optimization approach for the production-inventory-routing problem with simultaneous pickup and delivery

Hemmati Golsefidi, A ; Sharif University of Technology | 2020

457 Viewed
  1. Type of Document: Article
  2. DOI: 10.1016/j.cie.2020.106388
  3. Publisher: Elsevier Ltd , 2020
  4. Abstract:
  5. The production routing problem (PRP) merges the lot-sizing problem and the vehicle routing problem, two classical problems that have been the focus of comprehensive studies for over half a century. Solving the PRP is an effort to optimize decisions about the production, inventory, distribution, and routing in an integrated manner. In the literature of the recent decade, due to economic changes and regulatory issues, reverse logistics has become a focal point. Subsequently, the vehicle routing problem with simultaneous pickup and delivery (VRPSPD) has drawn more and more attention for its considerable effect on the reverse logistics problem. In addition, one of the major arguments in supply chain management is uncertainty of demand, which it is highly crucial to make accurate decisions. Most of the previous studies have disregarded the concept of uncertainty and reverse flow in the PRP. In this study, we present a mixed integer linear programming model (MILP) for the production-inventory-routing problem with simultaneous pickup and delivery, while considering a reverse flow of products. To reduce the total cost of the system, the proposed model plans the decisions with respect to production–reproduction setups, quantities of the production–reproduction, visiting/not-visiting the retailers, supplier inventory management, retail inventory management under the vendor-managed inventory (VMI) policy, the quantity of the defect-free products delivered to retailers and that of the defective products collected from them at the same time, and vehicle routing in each period. We also propose a new robust MILP formulation for the problem, while considering multiple uncertainty conditions, including the quantity of each retailer's demand, the quantity of the defective products collected from each retailer in each period, and the reproduction cost. Presenting the mentioned problem with a robust optimization approach which applies to equality constraints is the main contribution of this research. The performance of the model under uncertainty was depicted based on comparison between deterministic and robust model considering various random examples. Given the NP-hardness of the problem and to evaluate the performance of various algorithms in solving the model, we develop two meta-heuristic algorithms, viz. the simulated annealing algorithm and genetic algorithm. The performances of the proposed algorithms are analyzed and compared based on the numerical examples in three small, medium, and large sizes which for each size three states, including standard, high transportation cost, and high inventory cost are considered. © 2020
  6. Keywords:
  7. Genetic algorithm ; Production inventory routing problem ; Robust optimization ; Simulated annealing algorithm ; Simultaneous pickup and delivery ; Cell proliferation ; Decision making ; Defects ; Genetic algorithms ; Heuristic algorithms ; Inventory control ; Pickups ; Production control ; Sales ; Simulated annealing ; Supply chain management ; Vehicle routing ; Vehicles ; Production inventories ; Reverse flow ; Simulated annealing algorithms ; Simultaneous pickup and deliveries ; Integer programming
  8. Source: Computers and Industrial Engineering ; Volume 143 , May , 2020
  9. URL: https://www.sciencedirect.com/science/article/abs/pii/S0360835220301224