Productive service demands modularization for CNC machine tools based on the improved AP clustering algorithm

Abstract

The integration of manufacturing and service industry has become the mainstream trend. During the manufacturing process, CNC machine tools, as large-scale and complex equipment, involve a variety of service demands in the life cycle, especially the productive service demands are numerous and scattered. The article established the correlation model between customer demands and productive service demands based on clustering ideas and mathematical statistics theory to complete the modularization of productive service demands for CNC machine tools. A comprehensive correlation coefficient model and an improved AP clustering algorithm were put forward. The comprehensive correlation coefficient model mined the correlation between customer demand and production service demand and the self-correlation of production service demands directly based on the combination weight obtained by the analytic hierarchy process and rough set theory. The improved AP clustering algorithm was the combination of AP and Kruskal minimum tree principle. By this new algorithm, the clustering project of productive service demands for every customer demand could be attained. Finally, the five matrices can be computed in 1083 ms, 1067 ms, 1029 ms, 1149 ms and 1042 ms, respectively, by the improved AP clustering algorithm. However, the original AP method should spend 1122 ms, 1241 ms, 1231 ms, 1383 ms and 1231 ms, respectively. So it was very clear that the improved AP clustering algorithm can improve the data processing efficiency.