Kasilingam航空貨運收益管現模式之研究

Abstract

目前世界空運正蓬勃發展，尤其是航空貨運市場，依據波音公司1997 年之預測，未來五年航空貨運市場成長率以亞洲之年平均成長13.4%居世 界之冠。加上客運市場競爭激烈、逐漸趨於飽合，貨運市場由原先不受重 視，僅為客運之副產品(by-product )角色，躍升為目前舉世注目之焦點 。以往國內研究航空貨運乃是從企業角度作運輸選擇評估，顯少直接由航 空公司收益面去探討航空貨運；此外，歷年來不論國內外，海運或陸運型 態之貨物運輸相關研究成果顯著，對於目前扮演世界貿易舉足輕重的航空 貨運業研究甚少，Kasilingam在1997年提出航空貨運收益管理模式為首見 ，因研究以航空公司經營貨物運送的收益為題，就超賣水準之決定為主要 研究內容。 本研究所構建之超賣模式以總成本期望值最小、以及 淨收益期望值最大等兩種目標式所成，將航空公司所提供之貨運可用容量 與訂位顧客之貨物出現率均表示成具有不確定性之隨機變數型式，以貨物 實際出現數量小於貨運可用容量時所產生之閒置成本期望值，以及貨物實 際出現量大於可用容量時之超賣成本期望值。假設三種參數，包括平均費 率、閒置成本與超賣成本。本研究並構建出可供求解最佳超賣水準與目標 式值之模式。 本研究首先以不同之貨運可用容量及貨物出現率之機率 分配型式作試算結果比較，驗証超賣模式；並以國內可得資料進行應用試 算，比較航線別、機型別之影響。並對試算結果作一敏感度分析。就成本 參數之敏感度分析歸納，發現閒置成本對目標式之緞響大於超賣成本；在 閒置或超賣成本增加時，總成本期望值亦隨之增加，但超賣水準在超賣成 本增加時，會逐漸降低，在閒置成本增加時，會逐漸提高，目的均是為了 降低超賣或閒置所增加之成本期望值，以達總成本期望值最佳之目標。 就隨機變數參數之敏感度分析結果，在相同參數變動率下，不論對超賣水 準或目標式變化量，兩隨機變數之均數影響力均大於標準差。以均數相比 ，以出現率均數超賣水準變化量、總成本期望值變化量兩者之影響最大， 次為貨運可用容量均數。以兩隨機變數而言，不論對超賣水準或是目標式 ，出現率參數變化影響均比貨運可用容量參數高，可見對航空貨運，市場 需求(出現率)較航空公司供給(可用容量)對收益影響較深。 In Taiwan, much research has been done on assessing what kinds of transportation modes should be used by company or on analyzing efficient competitive strategies to air cargo forwarders and warehousing company, littlework has been done on revenue management about air cargo. In response to this, this research focus on revenue management of air cargo business from airlines who both have passengers and cargo business. In this study, we formulate firms' overbooking model that has two different objective functions: one is minimizing expected value of total cost,the other is maximizing expected value of net revenue. We define cargo available capacity of aircraft and cargo show-up rate as random variables, and assume average fare, spoilage cost, and oversale cost. We use FORTRAN and MATHEMATICA to solve overbooking level. Besides assuming probability distribution of cargo available capacity and cargo show-up rate, we also have collected some available data to decide overbooking level on different type of aircraft or flight line. Then, we design a sensitivity analysis on our practical case. From the sensitivity analysis results, we have a conclusion that spoilage cost has greater impact on objectivefunction than oversale cost oes. As spoilage (or oversale) cost increases, expected value of total costwill increase, but overbooking level will increase (decrease).In the aspect of random variables, if parameters of these two random variables has the same rate of change, mean of random variable has great impact on change of overbooking or of objective function than standard deviation of them does. For means, mean of cargo show-up rate has greater impact on overbooking level and objective function. No matter on overbooking level or objective function, cargo show-up rate is much effected than cargo capacity is. We can conclude that air cargo demand has greater impact on airline's revenue management.