Publications / 2024 Proceedings of the 41st ISARC, Lille, France

Exploring the Potential of Reinforcement Learning in Pipe Spool Scheduling in Industrial Projects

Mohamed ElMenshawy, Lingzi Wu, Brian Gue, Simaan AbouRizk
Pages 965-971 (2024 Proceedings of the 41st ISARC, Lille, France, ISBN 978-0-6458322-1-1, ISSN 2413-5844)
Abstract:

Pipe spools are key components in industrial projects. Usually, they're built off-site in a fabrication shop, then shipped to the project location for installation. The fabrication shop deals with numerous spools, each tailored to specific requirements according to shop drawings. These specific requirements together with other factors, such as lead time of materials, different processing times, availability of resources, and the nature of pipe spools being engineer to order products, render the scheduling process within the shop challenging and time consuming. As such, the aim of this research is to automate the scheduling process by developing a reinforcement learning model that includes an agent that is capable of handling the scheduling process. The proposed model is applied to an illustrative example to investigate the concept of automating the scheduling process. The construction professionals highlighted the great protentional of the proposed model in the decision-making process, and its ability to minimize manual intervention.

Keywords: Pipe spools, Reinforcement learning, Industrial projects