Real time failure recovery planning in flexible production systems is considered at the process level. Initially, the problem is structured and formulated within a systemic framework, and then, a generic roll-forward recovery approach that uses the process’s dynamic model, as well as general production knowledge, is proposed. The Colored Timed Petri Net formalism is used to represent, in a modular and compact way, the process discrete-event dynamics. In connection with this, an intelligent opportunistic recovery technique for finding, in real time, the recovery trajectory that conforms best within a specific operational strategy, is proposed. The application of the approach is demonstrated by considering cases taken from the chemicals batch-processing industry.