Analyzing Traveler Activity Choice Patterns in an Airport Terminal

• Main Authors: Lu, Kuan-Hung, 陸冠宏
• Other Authors: Chung, Yi-Shih
• Format: Others
• Language: zh-TW
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/fnsh3j

碩士 === 國立交通大學 === 運輸與物流管理學系 === 106 === As air passenger demand continues strong growth these years, the yield of aviation industry has been increasing accordingly. Besides aviation revenue, the yield and proportion of non-aviation revenue has also went up with significance. Hence, it is become more crucial for airport operators and shops to understand how passengers arrange their activities in an airport. Literatures about air passenger behavior mainly focus on the choice of activity type and fewer involve the time allocation on activities. This study introduced multiple discrete-continuous extreme value model (MDCEV) to discuss air passengers’ activities choice and time allocation. Next, we considered static and time-dependent attributes (e.g. remaining time before boarding and activities already done) simultaneously with Bayes classifier to reflect the characteristic that passengers would be affected by these factors when choosing which activity to do next and the time length one would spend on the activity. This study would also utilize Bayes Network to discuss the relation among passenger attributes. Third, this study would use cluster analysis to classify passengers who had similar time allocation on activities. By knowing passengers’ behavior in airport, airport operators and shops could adjust their service and make promotion to cater to the need of passengers with different characteristics.