Agent-based holonic job allocation in manufacturing operations

Owliya, Mohammad (2012). Agent-based holonic job allocation in manufacturing operations. University of Birmingham. Ph.D.

Owliya12PhD.pdf
PDF - Accepted Version
Restricted to Repository staff only until 31 December 2040.
Download (2MB)

Abstract

Job shop scheduling, especially in machine intensive manufacturing is heavily reliant upon job assignment systems for the management of machines on the shop-floor. Considering the machines as the elements of a network in the shop-floor, similar approach could be extended to the elements of supply chain of the manufacturing enterprise. This approach, combined with relevant concepts of the agent-based and holonic systems, has caused motivation to this research, whose primary aim is proposing new job allocation models in multi-level manufacturing operations.
That aim has been achieved by bringing the concept of topology of distributed systems into the current agent-based and holonic manufacturing systems. After presenting a conceptual holonic model of multi-level manufacturing operations for fulfillment of an external customer order, the work has focused on shop-floor level. Interaction protocol in multi-agent systems and its relation with the architecture of the agents’ network were studied and new models for task allocation were proposed in the context of a machine-intensive shop-floor. Having examined the models in the shop-floor, the best performing models were further customized and extended to manufacturing supply network (top-floor level). At the both levels, the research uses information and data from case studies of real-world manufacturing.
The results obtained were very promising for a ring-like job allocation model, to improve the key performance indicators of operations at the both levels. Although the research seeks practical solutions in Manufacturing Systems, its developed concept on the platform of MAS topology might be utilized within other engineering fields that benefit from the agent technology.

Type of Work:

Thesis (Doctorates > Ph.D.)

Award Type:

Doctorates > Ph.D.

Supervisor(s):