A fleet of vehicles with the same type and age needs to retire at different time points. The problem is to choose a subset of vehicles that will be first retired based on the overall performance of each vehicle. The paper presents an approach to establish an overall performance model based on operational and maintenance data of vehicles. The first step of the approach is to identify the key performance indicators based on the engineering knowledge; the second step is to carry out a principal component analysis so as to reduce the dimension of the problem; and the final step is to establish the overall performance model using the closeness measure of TOPSIS. The approach is illustrated by a real-world example, and the results are compared with the actual retirement decision. It is shown that the model-based decision is fairly consistent with the actual decision.
