A lagrange relaxation based algorithm for parallel injection machine scheduling problem Paralel enjeksiyon makine çizelgeleme problemi için lagrange gevşetme temelli bir algoritma


Creative Commons License

ARIK O. A.

Journal of the Faculty of Engineering and Architecture of Gazi University, vol.40, no.1, pp.277-286, 2024 (SCI-Expanded) identifier

  • Publication Type: Article / Article
  • Volume: 40 Issue: 1
  • Publication Date: 2024
  • Doi Number: 10.17341/gazimmfd.1425180
  • Journal Name: Journal of the Faculty of Engineering and Architecture of Gazi University
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Art Source, Compendex, TR DİZİN (ULAKBİM)
  • Page Numbers: pp.277-286
  • Keywords: batch processing, energy cost, job splitting, parallel machine, Plastic injection, scheduling
  • Erciyes University Affiliated: Yes

Abstract

Injection molding machines produce semi-finished and finished products necessary for many industries. Shops with these machines in parallel are referred to as parallel injection machine shops. In the production of orders, the connection of injection molds to the machines and the determination of the suitability of these molds for the machines are frequently addressed in the literature. This study is inspired by a parallel injection machine shop producing healthcare plastic products. The problem addressed in this study is significantly different from problems in the literature. It involves dividing orders among machines, processing a stack of orders from different customers, labor costs for production, penalty costs for products produced after the delivery time, injection machines with different production speeds, energy costs, and considering the compatibility of orders-molds-machines. A mathematical model is proposed for the problem incorporating these differences, ensuring linearity with all constraints and the objective function. Furthermore, a tool is developed for solving real-life problems using Lagrange relaxation technique. Test problems of various sizes are created to validate the proposed model and algorithm. It is observed that the proposed algorithm converges better to the optimum solution and performs better than the model.