A robust methodology for strategically designing environments subject to unpredictable and evolving conditions

The layout design process, a lean technique, has the potential to provide a manufacturer with significant cost reductions. The major challenge for layout designers is then ensuring this reduction can be maximized, but more so realized when implemented in practice. Guaranteeing this realization requires both the real-life behavior and characteristics of the environment as well as the market and business model conditions to adequately be captured. Unfortunately, current methods fail to accurately capture real-life considerations such as flow path feasibility, they neglect continuous detailed representations of evolving layouts subject to financial restrictions and uncertainty, and moreover they tend to provide insufficient insight into the problem. The objective of this dissertation was then to establish an improved methodology for exploring the design space of a detailed evolving environment, enabling more informed and collaborative design decisions to be made in the presence of evolving and uncertain conditions. In the process of achieving this goal, a three-step methodology (problem initialization, solution, analysis), titled LIVE, is formed. Along with its formation an extensive array of novel methods, revolutionary optimization techniques, and a detailed performance model are developed; all to facilitate effective solution to the uniquely complex and arduous layout problem formulation considered in this dissertation. It is then postulated, that if the problem of designing an environment subject to evolving and uncertain conditions was to be solved with this LIVE methodology, then designers would be capable of making more informed and collaborative design decisions. Substantiation of this is then pursued by systematically testing the methodology and the various models, methods, and solution approaches deployed by it. A series of compounding experiments are performed. During this testing, developed methods are proven to outperform existing approaches, consideration of flow path feasibility is proven imperative, and the novel bimodel multi-stage solution approach deployed by the LIVE methodology is well exercised whereby the best optimization parameter settings to ensure effective solution are identified. Finally, while applying the LIVE methodology to a real-world layout design problem, complete substantiation to the postulated hypothesis is achieved. It is shown that the methodology effectively facilitates improved insight and potential collaboration into the layout design process. The developed performance model proves significant in enabling new insights to be drawn and further for a richer understanding of the operations and layout design to be gained. Overall, the methodology demonstrates its ability to provide an improved layout design process that can effectively handle design problems subject to uncertain and evolving conditions; enabling strategic business decisions to be considered in parallel to the design of the layout.

Level thesis:degree_level

Doctoral

Department dc:contributor.department

Aerospace Engineering

Grantor dc:publisher

Georgia Institute of Technology

Year dc:date.issued

2019

Author dc:creator

• Minier, Ethan T.

Advisor dc:contributor.advisor

• Mavris, Dimitri N.

Committee members dc:contributor.committeemember

• Schrage, Daniel P.
• Pinon-Fischer, Olivia J.
• Briceno, Simon I.
• Rahmani, Morvarid

• Layout design
• Manufacturing
• Strategic decision making
• Optimization
• Uncertainty
• Robustness
• Methodology
• Model
• Facility layout problem
• Lean manufacturing
• Cellular manufacturing

Language dc:language.iso

en_US

Handle dc:identifier.uri

http://hdl.handle.net/1853/61208

OAI identifier oai:identifier

oai:repository.gatech.edu:1853/61208