Technologies are generally described for systems and methods effective to schedule repair (e.g., allocate repair resources, determine a repair sequence, etc.) of a system effected by a large-scale failure caused by a natural disaster, malicious attack, faulty components, or the like. In an example, the system can generate a schedule that indicates amounts of repair resources allocated for repair of specific components of a disrupted system as well as a time or sequence in which the components are to be repaired. The schedule, in some instance, can operate to maximize an amount of restoration, at each stage of a recovery process, relative to the characteristic of the disrupted system. For example, with a communications network as the disrupted system, the schedule can maximize the amount of total traffic flow capacity recovered after respective steps of the recovery process.