A DEA based approach for optimization of urban public transport system

The paper demonstrates a relatively new perspective application of data envelopment analysis (DEA) in the optimization of urban public transport (UPT) systems. For this goal a special two-phase approach is used. The first phase consists of applying a special adjusted DEA model-input oriented hybrid model with non-controllable and bounded inputs with categorical decision making units under constant returns to scale-for evaluating efficiency of UPT lines and for obtaining recommendations for improving their efficiency. Based on the recommendations from the first phase, in the second phase a possible assignment of dispatched vehicles of inefficient UPT lines is proposed. The results of the conducted analysis offer suggestions to improve efficiency of the UPT so that the balance between supply and demand for transport services is maintained. The proposed method is applied to the UPT system in a Slovak city, Banska Bystrica.