Mixed network DEA models with shared resources for measuring and decomposing performance of public transportation systems

Document Type

Journal Article

Publication Date


Subject Area

mode - subway/metro, operations - performance


Network DEA, Mixed network, Metro, Efficiency, Effectiveness


This study proposes a hybrid two-stage network model and a mixed network DEA with the shared-inputs model to jointly measure the efficiency and effectiveness of a metro transport system. Performance is determined by the hybrid two-stage network DEA with the shared-inputs model to account for the non-storable service feature. In contrast, the mixed network DEA model with the shared-inputs is used to describe the metro system's non-storable service, joint-product, and multi-subsystems features. Shared resources are analyzed by these two models to determine which common input resources are shared among multiple processes. A case study is conducted with the two models to evaluate and decompose the performance of a metro transit system. Results from the hybrid two-stage network and mixed network DEA models indicate that the performance measures or ranks differ significantly when using non-parametrical statistical tests. Transit system operators can gain more complete performance information on subsystems from the results of the mixed network DEA model than from the hybrid two-stage network model, enabling further improvements to their systems and sub-systems.


Permission to publish the abstract has been given by Elsevier, copyright remains with them.


Research in Transportation Business & Management Home Page: http://www.sciencedirect.com/science/journal/22105395