Managing valuable Taiwanese airline passengers using knowledge discovery in database techniques

Taiwanese domestic airline passenger loads have declined over the past decade, and the inauguration of the Taiwan high-speed rail service in 2007 has further intensified the competition facing the airlines. Gaining and retaining valuable passengers thus has become essential in maintaining domestic airline profits in the increasingly competitive transportation market. This study develops a knowledge discovery in database process, including the recency, frequency, and monetary model and the C5.0 decision tree. The former is used to assess passenger value, while C5.0 is used to mine demographic, travel behavior, and critical service quality information for valuable passengers. The results identify valuable passenger characteristics that can help Taiwanese airlines maximize their number and retention. (C) 2007 Elsevier Ltd. All rights reserved.