本研究主要目的是建構一鐵路列車超額訂位之收益管理模式，以此模式探討在臺灣高鐵列車乘客不同之隨機到站型態下，使用不同列車超額訂位策略之收益分析。在此超額訂位模式中，應用在於旅客需求與列車座位相差不大之情形，其模式主要考量因素是高鐵乘客到站與否之隨機性，利用二元變數來建構高鐵乘客情境收益式與補償、賠償成本式。本研究使用蒙地卡羅模擬法來處理隨機性之變數，透過模擬探討不同超額訂位策略對於列車收益之影響。本研究亦利用敏感度分析探討在不同乘客未到率下，列車收益最大之超額訂位率。此外，本研究亦推估出列車收益最大前提下 之最適高鐵列車指定席與自由席之車廂分配比例。研究結果顯示：在乘客未到率為 10%~20%之情境假設下，高鐵最適超額訂位率為估計乘客未到率 再加 5%時之列車收益最佳。至於列車最適之指定席與自由席車廂分配比例，在乘客未到率小於 10%時，最適分配比例為 8:3；在估計乘客未到率為 10%~20%時，最適分配比例可為 10:1。

The study aims to establish a railway over-booking yield management model. A binary function is utilized to formulate a high-speed railway formula consisting of benefit, compensation and discount cost. By focusing on the random distribution of high-speed railway passenger’s arrivals and absences, the Monte Carlo method is applied to compute random variables and calculate 1,000 random variations of the HSR over-booking model. Outcome of calculations is normally distributed and the effects of over-booking on railways are analyzed. Sensitivity analysis on different models is conducted to identify the best overbooking strategy and optimal coach distribution of reserved seats and unreserved seats. Results assume a 10%-20% no-show rate while the most reasonable no-show rate is 5%, which closely represents that of the high-speed railway. The optimal coach distribution of reserved seats and unreserved seats is 8:3 when the no-show rate is less than 10%. When the no-show rate is between 10%~20%, the best care distribution is 10:1.