Operational availability is a fundamental measure to assess the system performance after the installation. To achieve the desired availability goals, various strategies have been discussed, ranging from preventive maintenance, reliability-redundancy allocation (RRA), to spare parts logistics. RRA aims to extend the system uptime while spare parts logistics can reduce the downtime. These methods become difficult to choose if the fleet size changes over time. This situation often occurs in the new product introduction stage. This paper develops new cost model and analyzes the trade-off between redundancy allocation and spare parts stocking. Our model is built upon an integrated product-service mechanism where the firm manufactures the products and also provides after-sales support. We show that component redundancy is preferred over spare part inventory under long-term, performance-based contract. Examples from semiconductor equipment industry are used to demonstrate the application of the proposed method.