Estimating relative attractiveness of locations using data envelopment analysis

Competitive facility location problems involve identifying the best location of a facility that can capture maximum market share in the presence of competition. One of the most popular models for competitive facility location, namely the Huff model, considers only two factors for deciding a good location. This model is not considered very realistic and efforts have been made to improve the model by including additional factors. In this paper, an extension of the Huff model to consider multiple factors using data envelopment analysis (DEA) is proposed. Since DEA computes the weights (measuring the influence) of factors implicitly using linear programming, there is no need for externally specifying their values. The proposed model is applied for estimating the market share of selected tourist destinations in the Sultanate of Oman. The results of the proposed model agree with the findings of a tourism survey conducted in the country.