| Summary: | 碩士 === 國立東華大學 === 運籌管理研究所 === 102 === In the make-to-order environment, manufacturing companies often receive a number of order requests in a short time. Due to the limited production capacities, companies need to decide which orders should be accepted and scheduled in suitable machines in order to maximize the profit. It depends on the orders’ profits, due dates and the resources consumed.
A mathematical model for earliness-tardiness order acceptance and parallel machine scheduling problem is formulated to maximize the total profit. In order to solve the problem, we use particle swarm optimization combined with variable neighborhood local search to solve the sequencing problem. The optimal timing algorithm is used to obtain the optimal production time in a fixed production sequence. Different sizes of problems are tested to validate the effect of the proposed algorithms. We find out that PSO4.2 is better if earliness penalty weight is equal to tardiness penalty weight. If earliness penalty weight is not equal to tardiness penalty weight, PSO4.1 is better. In the end, we tested different ranges of due dates in order to check if it will affect the computing time. We find that it takes more time to solve the problem in the wide range of due dates.
|
|---|
