Preface: data envelopment analysis

Abstract

Data envelopment analysis (DEA) is a methodology for data-oriented analytics. Over the years, we have seen new and novel developments and applications of DEA in a variety of areas. In this INFOR special issue, we present the second part of ‘DEA and its applications in operations’. This special issue aims to compile the state-of-the-art research papers spanning models, theory, empirical studies, applications and case studies on DEA. These contributions provide methodological advances, new and valuable insights, and implications to the practice of DEA for performance evaluation and benchmarking. Five articles are included in this issue. Super efficiency is a DEA approach that can be used to discriminate the performance of efficient units. A well-known problem of the super efficiency approach is that its models can be infeasible. Such infeasibility can be caused by various sources. In ‘Modified super-efficiency DEA models for solving infeasibility under non-negative data set’, Lin and Chen develop a super efficiency model that can handle infeasibility caused by the presence of negative and zero data. This allows the application of the super efficiency approach when non-positive data are observed. In an effort to distinguish the performance of efficient DMUs (decision making units) context-dependent DEA is developed to evaluate DMUs in different levels of best practice frontier. However, as noted by Zhu, Sun and Zhang in ‘Context-dependent data envelopment analysis with common set of weights’, the original context-dependent DEA only shows how distinct a DMU under evaluation is from a single specific virtual DMU on the evaluation context and ignores the entire evaluation context. In this paper, the authors incorporate the common set of weights into the original context-dependent data envelopment analysis to address the issue. This improvement ensures evaluation consistency via providing an overall evaluation. Resource allocation is a popular and important issue in the enterprise management. Most existing DEA-based resource allocation focuses on single-stage system or considers the internal production process of the system as a ‘black box’. In ‘Resource allocation of a parallel system with interaction consideration using a DEA approach: an application to Chinese input-output table’, Xiong, Wu, An, Chu and Liang propose a new DEA approach to allocate the resource in a bidirectional interactive parallel system. They consider not only the resource allocation of a certain DMU, but also the resource allocation of all DMUs for a centralized decision-maker through centralized models. In ‘Banks efficiency and productivity in Togo after the financial liberalization: a combined Malmquist index approach’, Zhou, Placca, Jin, Liu and Wu analyse the productivity growth of Togolese banks. Their study period corresponds to the post-financial liberalization in the West African Economic and Monetary Union (WAEMU) zone and the third phase of a changed banking and financial environment. A new Malmquist Index is developed in order to estimate the Total Factor Productivity Growth and its components. The