Selecting location of IT business units using unique DEA-MACBETH-OPA framework

This paper proposes a methodological framework that combines data envelopment analysis (DEA) with measuring attractiveness through a categorical-based evaluation technique (MACBETH) and with the Ordinal Priority Approach (OPA) to evaluate the location problem. This problem is solved in the field of IT business, which has its own specific factors and criteria for location selection. The efficiency evaluation in the presence of multiple different factors and criteria is done by DEA and results heavily depend on their selection, values, and weights assigned. The MACBETH method helps to reduce the number of criteria and to determine the relative importance of such criteria, based on experts’ preferences. OPA method helps in weight restriction and consequently in increasing DEA discrimination power. The aim of the empirical study is to verify the possibility of using a proposed methodological framework for IT business location problems. The authors used a database of 42 potential locations, observed as decision making units (DMUs) for evaluation of ten criteria: seven inputs and three outputs. Based on the DEA results, 25 DMUs have an efficiency index of 1, which is almost 60% of the total number of DMUs. Using MACBETH method for defining the most important criteria and incorporating these results into DEA, the number of relatively efficient DMUs is decreased to 8. The OPA method is applied to aggregate 10 inputs into one weighted input which is included in the DEA analysis and only 4 DMUs with restricted weights using DEA & Rough OPA is obtained. The study results show that in comparison to the single DEA method, the proposed methodological framework has the fewest efficient locations, more precisely four relatively efficient locations for IT business units were identified. These findings emphasize that this framework has a superior ability to discriminate the performance of DMUs. Obtained importance might be a guideline for selecting adequate input and output criteria in DEA model when assessing IT business location efficiency. The originality and validity of applying this methodological framework are shown through the evaluation of potential locations for IT business units. The importance of the proposed framework is particularly seen in its flexibility and adaptability in location problems. The results of experiments that discuss and evaluate different scenarios confirm the efficiency and validity of the proposed framework. Finaly, proposed methodological framework can help to define location strategies based on infrastructure, recruitment issue, ecosystem and costs in real problems.