Application of Clustering Algorithms, Grey Relational Analysis and TOPSIS Approach in the Performance Evaluation of Global Offices

• Main Authors: Cindy Leou, 柳美珠
• Other Authors: Tien-Chin Wang
• Format: Others
• Language: en_US
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/40328457624175054066

碩士 === 義守大學 === 資訊管理學系碩士班 === 93 === This study aims to identify key factors affecting the ABC Register of Shipping global offices’ performance. Service quality is a composition of various criteria that are difficult to measure, and among ones the most important are survey standards, late reporting, authority levels, plan appraisals, new construction, statutory and report writing, all of which are listed in the Key Performance Indicator Table (KPI). A quick response is of the most critical factor such that a delayed survey report could cause serious embarrassment of the ABC Register of Shipping*, and irritation to their clients. In order to overcome the above problem, this paper adopts a clustering algorithm into the performance measurement. Specifically, we apply grey relational analysis to obtain values of relational coefficients and grey relational grades to rank the 124 ABC Register of Shipping offices within five clusters. At the same time, a Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method of a simple weighted addition approach was applied for processing the performance measurement to rank the 124 offices within five clusters according to the Closeness Coefficient ( ). According to the results of ranking for the benchmark with each office in the same cluster, the previous office is used as a benchmark target to achieve the highest level of service based on its level of efficiency and effectiveness. The results confirm the approach is feasible and appropriate; also the results enable these offices to improve their service quality and meet the training needs of their surveyors.