Title: Estimation of true efficient frontier of organisational performance using data envelopment analysis and support vector machine learning
Authors: Kerry Poitier, Sohyung Cho
Addresses: Department of Industrial Engineering, University of Miami, Coral Gables, FL 33146 USA. ' Industrial and Manufacturing Engineering, Southern Illinois University Edwardsville, Edwardsville, IL 62026 USA
Abstract: Data envelopment analysis (DEA) and stochastic frontier functions (SFF) are two well-known tools for performance and efficiency analysis of profit and non-profit organisations, referred to as decision making units (DMUs). The challenge to traditional DEA is how to account for both managerial and observational errors if present in the analysis, so as to determine true efficient frontiers. This paper proposes a novel methodology to determine true frontiers in a non-parametric environment. Specifically, traditional DEA is integrated with SFF through support vector machine (SVM) learning to provide an adaptive way to estimate true frontiers considering managerial and observational errors. A statistical ratio is utilised to find the true frontiers, and the proposed methodology is applied to a real data set where frontiers are compared to ones obtained by other existing methods. The work in this paper can help organisations to plan a more realistic investment by providing reasonable sense of benchmarking.
Keywords: performance analysis; data envelopment analysis; DEA; support vector machines; SVM; organisational performance; decision making units; DMUs; stochastic frontier functions.
International Journal of Information and Decision Sciences, 2011 Vol.3 No.2, pp.148 - 172
Published online: 26 May 2011 *Full-text access for editors Access for subscribers Purchase this article Comment on this article