Pavement performance evaluation using an improved gray clustering

Pavement performance evaluation plays a critical role in a pavement management system as it provides basis for maintenance and rehabilitation method. Essentially, pavement performance evaluation is a complex multi-criteria decision-making process including some data and properties relating to the environment and structure which are usually uncertain and subjective. In order to make the evaluation process more efficient and objective, this paper proposes an improved gray clustering principle of the gray theory, by applying the vectorial angle cosine method to calculate the weight of each index, and then use the values to replace the gray classification coefficient of the traditional gray clustering. In this paper, the principle of the improved gray clustering is introduced and a general algorithm of the performance evaluation is presented. An illustrative example of pavement performance evaluation is used to demonstrate the feasibility and practicability of the proposed model.