Benchmarking service quality from the inside and outside of a service organization by group decision-making

How to benchmark the service quality of a service organization is an important problem. However, there is no standardized or formulized method to measure it. One of the feasible methods is marked by a group of decision makers. This group is composed of decision makers coming from both the inside and the outside of the service organization. The most difficult problem is to allocate the decision-makers´ weight in the group decision. Each decision maker has his or her specific weight, because he or she has different authority and experience or for other reasons. This paper studies the decision-makers´ weight allocating algorithm and then puts forward a new service quality benchmarking model. Prior research methods on the weight allocation of group decision-making include Authority Allocation, Nominal Group Technique method, Entropy Method and Fuzzy Cluster Analysis. The first and the second methods analyze the weight from the subjective aspect and the rest analyze the weight based on the historical decision-making data. Essentially, weight allocation is a complicated problem. These methods unavoidably lead to unilateral conclusion. This paper defines the quantificational preference utility value of a service quality item, puts forwards a new algorithm to determine the decision-makers´ weight in group decision problems, and considers three factors affecting the decision-makers´ weight and two aspects of the service organization in the service quality benchmarking model, including the inside evaluation and outside evaluation. This paper studies the resolving algorithm of weight evaluation based on hierarchical analysis and normal stochastic matrix, analyzes the quantitative relationship between decision maker´s weight and the consistency index and characteristic vector of preference judgment matrix. In the end, the paper puts forward the interactive implementation steps of the decision-making model for service quality benchmarking problems.