On the Importance of Service Parts When Taking Commonality and Reliability Decisions

Competitive Original Equipment Manufacturers (OEMs) do not only sell equipment, but also provide service contracts that ensure proper functioning and uptime of equipment after the sale. This makes OEMs responsible for a large part of an equipment’s life cycle. Therefore, OEMs aim to minimize the total life cycle costs of their equipment, in particular by commonality and reliability level decisions during the design phase. We consider these decisions for one component occurring in a family of systems. The commonality decision is about choosing a common component or dedicated components for the systems. The life cycle costs consist of design and production costs of all components, repair costs, inventory holding costs of service parts, and logistic downtime costs (i.e., downtime costs due to insufficient spare parts inventory). At many OEMs, the design department tends to exclude service parts considerations for the commonality and reliability level decisions. Excluding service parts leads to a simpler decision model, however, this may lead to non–optimal decisions. We compare two approaches for the commonality and reliability decisions: the anticipating approach, which includes inventory holding costs of service parts and logistic downtime costs, and the nonanticipating approach, which excludes these costs. Since the cost function under the anticipating approach is intractable, we first derive a cost function that approximates the actual costs via an asymptotic analysis. Next, we show that the anticipating approach selects the common component more frequently in comparison with the nonanticipating approach. For many problem instances, the nonanticipating approach leads to decisions that have a much higher overall cost than using the anticipating approach. The relative cost difference can become arbitrarily large when the family of systems consists of many systems. Funding: J. Driessen was funded by the Netherlands Organisation for Scientific Research [Grant 407-12-001].