臺灣博碩士論文加值系統

在運輸、飯店訂房與租車等時效性服務產業中，預訂系統佔有相當重要的位子，過去對於此類產業皆是使用收益管理並以追求最大利潤來解決此類問題，本論文的研究對象為台灣鐵路管理局，在現行的制度之下，並無法使用超額訂購與差別取價，因此主要研究的方向為藉由資源分配以改善其經營績效，我們針對預購票的預訂時間與數量控制為研究重點。
在此研究中，我們假設顧客有三種類型，分別是旅行業者、商務人士與一般人士，我們同時也定義三種顧客在訂票系統中的行為，藉由使用模擬套裝軟體Arena模擬本研究中的各種情境與政策，本研究藉由搭乘人數、退票人次、取消票次與失敗數來建立三種衡量方式分別是最大化收益、最小化浪費次數與總淨值，來衡量該採用何種管理政策。我們考量的放票政策為一開始全額開放(政策一)、第一天開放一半、第八天後全額開放(政策二)與每天多開放10%(政策三)並搭配三種取票時間政策(兩天、三天與五天)共九種管理政策。
本研究的結果顯示，從總淨值價值來看，最佳的管理政策為為取票日五天與放票政策策二，而目前台灣鐵路公司所採用的管理政策在每一個衡量指標下都不是最佳的，若是限定取票日為兩天的情況下，則應該採用放票政策二(即第一周開放一半，第八天後全額開放)；本研究也參考國外的做法，以百分比做為退票的處理成本，而採用此法時，總淨值的最大值落在取票時間五天與放票政策三(即每天開放10%)之中。

Booking system plays an important role in perishable asset such as transportation industry, hotel industry and car rent industry. The concept of Revenue Management solves the problem of perishable asset by using forecasting, allocation, price discrimination and overbooking policy. However, Taiwan Railway Administration’s current system does not allow for price discrimination and overbooking policy. The problem here is resource allocation, we therefore focused on the volume of resource and the timing for the availability of resource.
In this research, we constructed a simulation model and assume there are three kinds of customers namely, travel agent, businessman and normal customers, each with individual booking behavior in the booking system. We have three ticket-get date polices, two days, three days and five days. We also have three resource policies, policy 1 is full allowance in the beginning, policy 2 is half allowance in the beginning and full allowance after a week and policy 3 is extra 10% of resource allowance. We use simulation Arena 11, to simulate different population as our scenario and use four indexes: success booking rate, refund rate, cancel rate and failure rate to describe their conditions. We also use four indexes, revenue, society resource, same weight on revenue and society resource and value to measure the performance in each policy.
The results showed out that on the value side, the best combination of policy is resource policy 2 with 5 days. The current get time policy is 2 days, if we cannot change the get time policy, we would rather choose the resource policy 2 than resource policy 1. We also recommend the regulation for refund that is similar to Japan, China or Korea as percentage of ticket price rather than a fix number. The best combination is also the 5 days and resource policy 3.

Acknowledgements .................................... i
中文摘要 .................................... ii
Abstract .................................... iii
Contents .................................... iv
List of Figures .................................... vi
List of Tables .................................... vii
Chapter One: Introduction and Motivation .................................... 1
1.1 Introduction .................................... 1
1.2 Purpose of this research .................................... 3
1.3 Research structure .................................... 3
Chapter Two: Literature Review .................................... 5
2.1 Revenue management .................................... 5
2.1.1 Demand forecasting .................................... 5
2.1.2 Price discrimination .................................... 6
2.1.3 Seat allocation .................................... 7
2.1.4 Overbooking .................................... 8
2.2 Comparison between airline industry and railway industry .................................... 9
2.2.1 The same characteristics between airline industry and railway industry .................................... 9
2.2.2 The difference in revenue management between airline and railway industry .................................... 10
2.3 Booking policies in different country .................................... 12
2.3.1 Passenger railway system in China .................................... 13
2.3.2 Passenger railway system in Japan .................................... 13
2.3.3 Passenger railway system in South Korea .................................... 15
2.3.4 Passenger railway system in Taiwan Railway .................................... 16
2.3.5 Passenger railway system in France .................................... 18
2.3.6 Passenger railway system in UK .................................... 18
2.3.7 Summary .................................... 20
Chapter Three: Research Methods .................................... 21
3.1 Problem statement and definition .................................... 21
3.2 Research Assumption .................................... 25
3.3 Process of booking system in Arena software .................................... 25
3.3.1 Behavior of travel agent .................................... 25
3.3.2 Behavior of normal customer .................................... 26
3.3.3 Behavior of businessman .................................... 27
3.3.4 The flow of ticket in the system .................................... 28
3.4 Input Parameter .................................... 28
Chapter Four: Simulation and Analysis .................................... 31
4.1 Data collection .................................... 32
4.2 Simulation result and analysis .................................... 32
Chapter Five: Conclusions .................................... 37
5.1 Limitations and suggestions .................................... 39
References .................................... 41

1. Abe, Itaru., (2007),”Revenue Management in the Railway Industry in Japan and Portugal :a Stakeholder Approach,” Master if Science in Technology and Policy, Massachusetts Institute of Technology.
2. Bodily, S. E. and Weatherford, L.R., (1992), “A Taxonomy and Research Overview of Perishable-Asset Revenue Management” Yield Management, Overbooking, and Pricing, Operation Research, Vol.40, No.5, pp.831-844.
3. Clabcimino, A., Inzerillo, G., Lucidi, S., Palagj, L.,(1999), “A mathematical programming approach for solution of railway yield management problem”. Transportation Science 33, pp. 168-181.
4. http://jreast-shinkansen-reservation.eki-net.com/
5. http://www.korail.com/en/rv/pr21100/help.jsp
6. http://www.railway.gov.tw/tw/
7. http://www.thsrc.com.tw/tc
8. http://www.tgv.com/
9. http://www.nationalrail.co.uk/
10. Kraft, Edwin R., Bellur N. Srikar, and Robert L. Phillips.(2000), “Revenue Management in Railroad Application. “ Journal of the Transportation Research Forum 39, no. 1, pp. 157-176.
11. Littlewood, K., (1972), “Forecasting and Control of Passenger Bookings,” AGIFORS Symposium, Proc. 12, pp. 95-117,.
12. Peng-Sheng You(2008), “An efficient computational approach for railway booking problem.” European Journal of Operational Research Vol.185, pp.811-824.
13. Rohit Bharill and Narayan Rangaraj (2008), “Revenue management in railway operation: A study of the Rajdhani Express, Indian Railways.” Transportation Research PartA, pp.1195-1207.
14. Sulistio, Anthony, Kyong, Hoon Kim and Buyya, Rajkumar, (2008), “Management cancellations and no-shows if reservation with overbooking to increase resource revenue”. Eighth IEEE International Symposium on Cluster Computing and the Grid, pp.267-276.
15. Talluri, K.T. and Ryzin, G. V.,(2004), “The Theory and Practice of Revenue Management,” Boston: Mass Kluwer Academic Publishers.