An information granulation entropy-based model for third-party logistics providers evaluation

We introduce an innovative Information Granulation Entropy method to evaluate third-party logistics providers. Conventional fuzzy evaluation methods are valuable but biased at times. Objective measurements are rational; however its results are often difficult to explain. To take advantage of the strength of both methods, we propose a comprehensive evaluation framework to allow subjective judgment on alternatives, at the same time deriving criteria weights objectively. In the proposed model, experts input fuzzy language to form an evaluation matrix. After defuziffying the matrix, the K-means clustering method is applied to discretise the matrix. An information granulation entropy approach, based on information science theory and data mining technique, is then developed to determine the weights of criteria. Finally, TOPSIS closeness rating method is applied to derive the priorities of alternatives. To demonstrate its validity, we present a real-world application for selecting a third-party logistics provider. The proposed evaluation framework is particularly beneficial when dealing with large-scale, diverse criteria and alternatives.