Optimal data pooling for shared learning in maintenance operations
Presentation is building ... The process may take up to 3 minutes. If the page does not reload automatically, please reload it manually.
Voice is AI-generated
Connected to paperThis paper is a preprint and has not been certified by peer review
Optimal data pooling for shared learning in maintenance operations
Collin Drent, Melvin Drent, Geert-Jan van Houtum
AbstractThis paper addresses the benefits of pooling data for shared learning in maintenance operations. We consider a set of systems subject to Poisson degradation that are coupled through an a-priori unknown rate. Decision problems involving these systems are high-dimensional Markov decision processes (MDPs). We present a decomposition result that reduces such an MDP to two-dimensional MDPs, enabling structural analyses and computations. We leverage this decomposition to demonstrate that pooling data can lead to significant cost reductions compared to not pooling.