Decomposition Strategies for Solving Scheduling Problems in Industrial Applications

TL;DR

This paper explores decomposition strategies using logic programming to solve the Job-shop Scheduling Problem, aiming to improve scheduling efficiency in industrial applications through a phased research approach.

Contribution

It introduces a novel decomposition approach for JSP and evaluates it on benchmark instances before applying it to real-world factory scheduling.

Findings

Decomposition strategies show promise on benchmark instances.

Initial phase completed, second phase underway for real-world application.

Model aims to generate short-term factory schedules.

Abstract

This article presents an overview of a research study of a crucial optimization problem in the Computer Science/Operations research field: The Job-shop Scheduling Problem (JSP). The JSP is a challenging task in which a set of operations must be processed using a set of scarce machines to optimize a particular objective. The main purpose of the JSP is to determine the execution order of the processes assigned to each machine to optimize an objective. Our main interest in this study is to investigate developing decomposition strategies using logic programming to solve the JSP. We split our goal into two main phases. The first phase is to apply the decomposition approach and evaluate the proposed model by solving a set of known benchmark instances. The second phase is to apply the successful decomposition methods obtained from the first phase to solve a scheduling problem in the real-life application. In the current state, we finished the first phase and started the second one aiming to have a model that can provide a schedule of a factory for a short-time period.