Plan maintenance turnarounds constitute a large fraction of all maintenance activities in the process industries. We consider turnaround planning problems over large networks of interconnected plants. The network interactions provide an opportunity to plan and coordinate the different turnaround activities to save on annual downtime and recover the associated revenue. We propose a stochastic optimization model to quantify the risk of loss in rescheduling maintenance turnarounds that have been previously planned and compare the proposed approach to alternative approaches incorporating different production planning strategies under uncertainty. The model also provides simultaneous hedging strategies using inventories for unplanned outages. Thus, our model offers additional flexibility to previous approaches that address long- and medium-term turnaround planning problems, and explicitly incorporates plant reliability in the planning process. Through extensive computational studies, we show that proactive planning strategies that take uncertainty over multiple time periods into account offer substantial benefits over a reactive strategy. (C) 2017 Elsevier Ltd. All rights reserved.