資料探勘技術應用於旅客自助報到之分析—以C航空公司為例;Using Data Mining Techniques for Passenger Self-check-in - A Case Study of C Airlines

NCU Institutional Repository > 管理學院 > 資訊管理學系碩士在職專班 > 博碩士論文 > Item 987654321/84025

請使用永久網址來引用或連結此文件: http://ir.lib.ncu.edu.tw/handle/987654321/84025

題名:	資料探勘技術應用於旅客自助報到之分析—以C航空公司為例;Using Data Mining Techniques for Passenger Self-check-in - A Case Study of C Airlines
作者:	劉惠雯;Liu, Hui-Wen
貢獻者:	資訊管理學系在職專班
關鍵詞:	資料探勘;關聯規則;Apriori演算法;FP-Growth演算法;Data Mining;Association Rules;Apriori Algorithm;FP-Growth Algorithm
日期:	2020-06-22
上傳時間:	2020-09-02 17:56:46 (UTC+8)
出版者:	國立中央大學
摘要:	桃園國際機場旅客人數年年攀升，早已超過原先兩座航廈的設計容量，在有限的櫃檯數量分配之下，各航空公司在櫃檯配置上如何運用達到最佳化運用及動線的順暢，無疑是門學問。過去的研究指出櫃檯的指派模式、營運成本與自助報到使用率有顯著的關係。本論文將進一步研究使用資料探勘方式來進行實驗分析，並從中找出自助報到的使用率間隱含的關聯規則，以期能運用在航空公司規劃櫃檯配置動線的參考，提供旅客更優化的報到環境及便利性，進而提升旅客的滿意度。本論文以國內航空公司為例，以個案公司歷史航班自助報到服務資料進行分析，資料集期間為2017年至2018年，經整理後分為第一航廈及第二航廈，且同步做了兩個資料及合併資料集，共三個資料集以不同參數設定搭配使用Apriori及FP-Growth演算法進行關聯規則分析。實驗步驟以Weka 3.8.3資料探勘工具進行資料分析，利用資料探勘技術來建構航班旅客人次、航班屬性及飛行航點、時間帶的相關性分析，藉由過去自助報到的使用資料來了解變數因子間的相關性。研究結論「香港航線」及「日本航線」的「自助報到率高於均值」，「飛機機型為738」、「大陸航線」及「北美航線」的「自助報到率低於均值」，實驗分析所產出的關聯規則是有意義值得參考的，以提供機場、航空公司為櫃檯數量配置動線規劃及人力運用的參考方向。 ;The number of passengers at Taoyuan International Airport is increasing year by year, and has already exceeded the limit of the capacity in the original design. Under the limited available number of check-in counters, it is important that how airlines would allocate the counters to achieve the optimal use and perform the passenger check-in process more smoothly. In the past research, it has been pointed out that there is a significant relationship among the counter assignment pattern, the operating cost, and the self-check-in usage rate. This study will further investigate the use of association rule mining methods to conduct experimental analysis, and find out the implicit association rules based on the self-check-in utilization rate. It leads that we may be able to use the rules and the statistics to configure the moving lines at the airline planning counters to provide passengers with more optimized check-in environments and convenient experiences. Therefore, the passenger satisfaction would be improved. This thesis uses a local airline as an example to analyze its historical flight self- check-in data. The data were collected from 2017 to 2018, and are divided into two sets of Terminal one and Terminal two. In addition, one set is constructed by merging the two data sets. Therefore, three data sets in total were used to be analyzed by using association rules algorithms, Apriori and FP-Growth. The experiments were conducted by Weka 3.8.3. and the data exploration technology was employed to perform the correlation analysis between passenger numbers, flight attributes, destinations, and time zones. The experimental results indicate that the self-check-in usage rates of the passengers in the “Hong Kong route” and the “Japan route” are higher than that of the average; the cases of “aircraft type is 738”, “Mainland China route”, and “Americans route” have the lower self-check-in usage rates than that of the average. The association rules produced based on the experimental analysis are meaningful and worthy of reference for the configuration planning. With the help of the self-reported data in the past, the understanding of the variable factors along with their relationship can be achieved and it provides the valuable reference for the airport and the airlines to configure the number of counters along with the line planning and the manpower utilization.
顯示於類別:	[資訊管理學系碩士在職專班 ] 博碩士論文

文件中的檔案:

檔案	描述	大小	格式	瀏覽次數
index.html		0Kb	HTML	115	檢視/開啟

在NCUIR中所有的資料項目都受到原著作權保護.

社群 sharing

資料載入中.....