EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

360. Context-aware processing rate guided production scheduling: a forecasting embedded branch-and-price heuristic method

Invited abstract in session MD-28: Advancements of OR-analytics in statistics, machine learning and data science 4, stream Advancements of OR-analytics in statistics, machine learning and data science.

Monday, 14:30-16:00
Room: 065 (building: 208)

Authors (first author is the speaker)

1. Yige Sun
2. Sai-Ho Chung
The Hong Kong Polytechnic University
3. Tsan-Ming Choi
University of Liverpool

Abstract

Production scheduling seeks to identify effective job execution plans on machines to enhance overall operational efficiency. Prior studies in this area usually assume job processing rates (JPRs) to be constant. However, based on our analysis of real-world production data, multiple factors such as machine use, operator skills, material supplies, and the processing of prior jobs may influence the actual processing of a job in manufacturing operations. To catch the effect of production variabilities and improve scheduling solutions, we introduce a measure named context-aware processing rate (CAPR) in production scheduling. Based on this measure, a novel deep learning-empowered CAPR-guided scheduling approach is developed. Specifically, the CAPR is predicted for each processing context produced along with the pricing process. The CAPR is used to find beneficial allocation and sequencing of jobs to operators and machines with an efficient labelling algorithm. Experiments based on real-world data from our studied printing company show that the proposed method enables a substantial decrease in overall job completion time.

Keywords

Status: accepted

Back to the list of papers