سیدیعقوب حسینی
منیژه بحرینی زاده
علیرضا ضیائی بیده

In customer relationship management (CRM) systems, importance and performance of the attributes that define a service is very important. Importance-Performance analysis is an effective tool for prioritizing service attributes based on customer needs and expectations and also for identifying strengths and weaknesses of organization in the market. In this study with the purpose of increasing reliability and accuracy of results, customers are segmented based on their demographic characteristics and perception of service attributes performance and then individual IPA matrixes are developed for each segment. Self-Organizing Maps (SOM) has been used for segmentation and a feed forward neural network has been used to estimate the importance of attributes. Research findings show that mobile subscribers in Yazd province can be categorized in three segments. Individual IPA matrixes have been provided for each of these segments. Based on these results, recommendations are offered to companies providing mobile phone services.

Artificial Neural Networks, Customers Segmentation, Data Mining, Importance-Performance Analysis, Mobile Telecommunication Industry, Self-organizing Maps