Decentralized Scheduling Using The Multi-Agent System Approach For Smart Manufacturing Systems: Investigation And Design

Ebufegha, Akposeiyifa Joseph

Decentralized Scheduling Using The Multi-Agent System Approach For Smart Manufacturing Systems: Investigation And Design

dc.contributor.advisor	Li, Simon
dc.contributor.author	Ebufegha, Akposeiyifa Joseph
dc.contributor.committeemember	Brennan, Robert
dc.contributor.committeemember	Lee, Jihyun
dc.date.accessioned	2023-09-28T16:50:56Z
dc.date.available	2023-09-28T16:50:56Z
dc.date.issued	2023-09-20
dc.description.abstract	The advent of industry 4.0 has resulted in increased availability, velocity, and volume of data as well as increased data processing capabilities. There is a need to determine how best to incorporate these advancements to improve the performance of manufacturing systems. The purpose of this research is to present a solution for incorporating industry 4.0 into manufacturing systems. It will focus on how such a system would operate, how to select resources for the system, and how to configure the system. Our proposed solution is a smart manufacturing system that operates as a self-coordinating system. It utilizes a multi-agent system (MAS) approach, where individual entities within the system have autonomy to make dynamic scheduling decisions in real-time. This solution was shown to outperform alternative scheduling strategies (right shifting and dispatching priority rule) in manufacturing environments subject to uncertainty in our simulation experiments. The second phase of our research focused on system design. This phase involved developing models for two problems: (1) resource selection, and (2) layout configuration. Both models developed use simulation-based optimization. We first present a model for determining machine resources using a genetic algorithm (GA). This model yielded results comparable to an exhaustive search whilst significantly reducing the number of required experiments to find the solution. To address layout configuration, we developed a model that combines hierarchical clustering and GA. Our numerical experiments demonstrated that the hybrid layouts derived from the model result in shorter and less variable order completion times compared to alternative layout configurations. Overall, our research showed that MAS-based scheduling can outperform alternative dynamic scheduling approaches in manufacturing environments subject to uncertainty. We also show that this performance can further be improved through optimal resource selection and layout design.
dc.identifier.citation	Ebufegha, A. J. (2023). Decentralized scheduling using the multi-agent system approach for smart manufacturing systems: investigation and design (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca.
dc.identifier.uri	https://hdl.handle.net/1880/117165
dc.language.iso	en
dc.publisher.faculty	Graduate Studies
dc.publisher.institution	University of Calgary
dc.rights	University of Calgary graduate students retain copyright ownership and moral rights for their thesis. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission.
dc.subject	Multi-Agent System
dc.subject	Operations Research
dc.subject	Layout Design
dc.subject	Dynamic Job Shop Scheduling
dc.subject	Machine Selection Problem
dc.subject	Industry 4.0
dc.subject	Smart Manufacturing
dc.subject.classification	Engineering--Mechanical
dc.subject.classification	Engineering--Operations Research
dc.title	Decentralized Scheduling Using The Multi-Agent System Approach For Smart Manufacturing Systems: Investigation And Design
dc.type	doctoral thesis
thesis.degree.discipline	Engineering – Mechanical & Manufacturing
thesis.degree.grantor	University of Calgary
thesis.degree.name	Doctor of Philosophy (PhD)
ucalgary.thesis.accesssetbystudent	I do not require a thesis withhold – my thesis will have open access and can be viewed and downloaded publicly as soon as possible.