Applying MILP/Heuristic Algorithms to Automated Job-Shop Scheduling Problems in Aircraft-Part Manufacturing

Abstract

This work presents efficient algorithms based on Mixed-Integer LinearProgramming (MILP) and heuristic strategies for complex job-shop scheduling problemsraised in Automated Manufacturing Systems. The aim of this work is to find alternative asolution approach of production and transportation operations in a multi-product multi-stageproduction system that can be used to solve industrial-scale problems with a reasonablecomputational effort. The MILP model developed must take into account; heterogeneousrecipes, single unit per stage, possible recycle flows, sequence-dependent free transferringtimes and load transfer movements in a single automated material-handling device. Inaddition, heuristic-based strategies are proposed to iteratively find and improve the solutionsgenerated over time. These approaches were tested in different real-world problems arising inthe surface-treatment process of metal components in the aircraft manufacturing industry