Content area
In today's ever-evolving production landscape, characterized by a growing demand for personalized products to enhance consumer satisfaction, the strategy of pursuing high-mix, low-volume manufacturing has gained importance. More than ever, manufacturers are adopting the Make-To-Order (MTO) approach, aiming to balance efficient cost management by meeting strict customer deadlines. This research explores the complex state of job shop scheduling, a significant challenge faced by manufacturing entities trying to optimize production time and costs while making the best use of their machinery and resources. The primary concern is the dynamic relationship between order acceptance and scheduling within a job shop environment, which includes a diverse range of discrete and batch processing machines.
Order acceptance involves evaluating incoming orders based on available capacity and profitability. Once an order is accepted, scheduling allocates the necessary machines for each operation and organizes the process in the most efficient sequence. This research focuses on managing customer orders in a job shop, where jobs are processed through multiple operations on different machines. The job environment adheres to specific rules, including operation orders, fixed processing times, deadlines, job sizes, and prices. Each operation is carried out on an available machine, and some jobs may need to revisit machines for additional steps. The objective is to maximize profits while completing all accepted orders on time. The job shop consists of discrete machines and a batch processing machine that can accommodate multiple jobs within its capacity.
The research aims to formulate the problem and develop efficient solution approaches for large problem instances. The formulation prescribes which orders to accept and provides a detailed schedule to follow such that the overall profit is maximized. Leveraging the power of a Mixed Integer Linear Program (MILP), the problem can be structured to assist operations managers in balancing profits while making wise order choices.
This research proposes a Dantzig-Wolfe decomposition method and Simulated Annealing to create an efficient solution approach. The experimental studies show that the proposed decomposition method was able to achieve better profits compared to the full model. This work provides practical insights into order acceptance and scheduling, offering a reliable solution for manufacturers aiming to streamline their production processes effectively.