Integrating Bayesian Learning and Discrete Event Modeling for Adaptive Facility Layout in Remanufacturing

The landscape of production has evolved drastically from its nascency. The emergence of diverse demand, globalization, environmental and alternative aspects of the global economy, constitute greater complexity in manufacturing. The need for companies to stay competitive warrant robust business models and systems capable of accommodating uncertainty in markets. Increased attention to sustainability in manufacturing is promoting re-manufacturing directives poised to extend product service life which could present uncertainty in supply. This paper proposes a framework and modeling approach to equip manufacturing systems to respond to uncertainty in market demand and supply, with motivation nested in remanufacturing techniques that mitigate compromise in stakeholder requirements whilst accommodating more sustainable practice. The proposed production model implements Bayesian inferential data driven capability to account for uncertainty, heuristics methods in the form of genetic algorithms for adaptability to system deliverables, and discrete modeling approaches to simulate shop floor behavior through the generation of sample paths.