Remanufacturing processes are characterised by uncertainty due to the unpredictable status of used parts, entailing variable processing times and the need for reworks. Robust scheduling approaches are required to plan remanufacturing activities while guaranteeing the efficient use of resources and respect for delivery times. This work proposes a novel scheduling framework to handle uncertain processing times and rework matching the requirements of remanufacturing processes for turbine blades. The approach integrates branch-and-bound and heuristic steps, leveraging a Markovian model and phase-type distributions for the execution of activities and pursuing robustness through the minimisation of the value-at-risk of the makespan. An application to an industrial case is also reported to demonstrate the viability of the approach. © 2023 CIRP