The literature on the measurement of technical efficiency has grown substantially since Farrell (1957) introduced an index based on the maximum radial reduction in inputs consistent with observed production. Based on Farrell's work, Charnes et al. (1978) introduced data envelopment analysis (DEA), a linear programming approach to performance evaluation when production is characterized by constant returns to scale. Theoretical contributions have been numerous, including allowing variable returns to scale (Banker et al., 1984). The purposes of this paper are twofold. First, introduction to the theory, method and process for the application of DEA are described. Secondly, three cases in the application of network DEA, multi-activity DEA and Malmquist productivity decomposition to the transportation sector will be illustrated.