Examining additive manufacturing in supply chain context through an optimization model

Abstract

This study explores the problem of integrated jobs and vehicles scheduling in a two-stage supply chain, where parts are processed on additive manufacturing (AM) machines and then distributed to customers. In this study, we develop an optimization model for the problem with the objective of makespan minimization. Besides, several structural properties and lower bound formulation are provided to explain the main characteristics of the problem. In this regard, this study also contributes to the existing academic literature by investigating the two-stage supply chain scheduling problem with the additive manufacturing technology. Because the addressed problem belongs to non-deterministic polynomial-time hardness (NP-hard) problem class, a best-fit heuristic-based approach and several selection rules are developed to solve the problem. Each selection rule is designed by considering a distinct structural property of the problem and, thus, each of which is considered to be a different algorithm. A comprehensive experimental study is conducted through random instances, which are generated for small- and large-sized problems by considering real-production data. According to the computational results, the best-fit capacity utilization based selection (BFCUBS) algorithm is superior to others and yields a substantial improvement in the makespan. The reason behind this fact is that the high utilization of capacity enables a large number of jobs to be completed in a short time. Besides, as the number of jobs decreases and the capacity utilization rates increases all algorithms provide better results.