Ensemble decision making for logistics center location

Abstract

Logistics centers are crucial for efficient and environmentally friendly transportation, storage, loading, unloading, maintenance, repair, and handling processes. Proper selection of a logistics center location is essential for efficient trade and economy. Unlike other studies, this study presents an ensemble decision-making approach for logistics center location selection, using four weighting methods-entropy, standard deviation, method based on the removal effects of criteria, and statistical variance; six ranking methods-evaluation based on distance from average solution, measurement of alternatives and ranking according to compromise solution, multi-attribute utility theory, range of value, Technique for Order Preference by Similarity to Ideal Solution, and weighted aggregated sum product assessment; Spearman's correlation coefficient; and Borda methods. One of the main reasons for proposing the approach is that relying on a single method in solving decision-making problems may lead to wrong decisions due to the procedures of the methods. Instead, group decision-making based on different methods, as in this study, may be more dependable. The proposed approach is applied to the real-life problem of Turkiye's logistics center location selection due to its strategic location on trade and transportation corridors, ranking 47th in the Logistics Performance Index, and plans to build 25 logistics centers. Considering the significance of the problem, it is essential to use a reliable model. The integrated results indicate that the total imports criterion is the most important meaning that total imports are the most determinant in the best location. The ensemble ranking results indicate that Istanbul, Ankara, Izmir, Kocaeli, and Konya are the best five alternatives. In general, the results show that different methods provide different rankings for the same problem, which proves the importance of the proposed approach in providing more reliable results and avoiding facing a new decision-making model in terms of choosing the right method.