Today, improving machine availability is vital for industries to compete in the global market. Spare parts play an essential role in the maintenance and repair of equipment, but planning an extensive network in strategic industries with various spare parts can be very challenging due to the existence of different decision factors. The spare parts supply chain deals with inventory management issues, which necessitates considering the related decisions such as determining the stock level and order quantity. Moreover, demand uncertainty and long supply time make decision-making more complex. This paper presents a repair and supply planning model for repairable spare parts while considering a modified formulation of demand uncertainty to minimize costs. The model determines the optimal stock level, lateral transshipment, assignment of spare part orders to suppliers, equipment to repair centers, and the number of intervals over the planning horizon used in demand estimation. This research contributes to the literature by integrating recent decisions, using demand approximation by piecewise linearization, and considering backorder in warehouses evaluated by queuing models. A hybrid approach, including heuristic and genetic algorithms, is used to optimize the model using data from an oil company. The results show that using piecewise linearization and integrated repair and supply planning decisions optimizes costs and improves performance. Also, the availability is affected by the demand estimation, which necessitates precision prediction.