Monte Carlo DEA and budget allocation for data collection: An application to measure supply chain efficiency

Abstract

Supply chain operates in a dynamic platform and its performance measurement requires intensive data collection from the entire value chain. The task of collecting data in supply chain is not trivial and it often faces with uncertainties. This thesis is divided into two parts. First, we develop a DEA model to measure supply chain efficiency and provide an alternative way to measure efficiency in stochastic environment, which is Monte Carlo DEA. Second, in the context where data collection is needed and expensive, we provide a way on how to intelligently allocate the resources in data collection in order to get a better estimation of the efficiency score. As there is no explicit model to address this question, this thesis will introduce few methods based on the optimization simulation technique, which are the two-phase gradient technique and the GA (genetic algorithm) based technique to solve the problem.