Multi-level condition-based maintenance planning for railway infrastructures – A scenario-based chance-constrained approach

摘要: Abstract This paper develops a multi-level decision making approach for the optimal planning of maintenance operations railway infrastructures, which are composed multiple components divided into basic units maintenance. Scenario-based chance-constrained Model Predictive Control (MPC) is used at high level to determine an long-term component-wise intervention plan infrastructure, and Time Instant Optimization (TIO) applied transform MPC optimization problem with both continuous integer variables nonlinear problem. The middle-level determines allocation time slots interventions suggested optimize trade-off between traffic disruption setup cost slots. Based on high-level plan, low-level clustering be treated by agent, subject limit imposed proposed treatment squats, real data from Eindhoven-Weert line in Dutch network.