透過您的圖書館登入
IP:3.147.28.12
  • 學位論文

以動態定價模型解決有限供給下的供需失衡

Matching limited supply with spatial demand through price discrimination

指導教授 : 楊曙榮

摘要


本論文目的在於透過動態定價模型和顧客服務水準來解決供需失衡的問題,為此,我們採用微笑單車的資料來進行相關研究。然而,為了衡量消費者對於各個區域和不同收費採計費率在單車租借的效用,我們採用整數規劃下的貝式估計來估計經濟模型的參數,並依照參數估計的結果來呈現消費者對各產品的效用。此外,依據我們所估出的效用取得消費者對於各產品的需求函數,再透過動態定價去了解價格的變動對於微笑單車在顧客服務水準上面的影響。 根據研究結果,我們發現消費者對於公共單車的租借是非常價格敏感的,少量的價格波動就會對需求量產生劇烈的影響。此外,受限於人行道的空間以及自行車架的固定建置,我們很難透過可用自行車的存貨管理來解決供需不均的問題。因此,政府對於公共自行車的轉乘補貼政策對於我們的動態定價策略會產生非常巨大的影響。不幸的是,在我們所能取得的資料中,並沒有顯示有多少比例的消費者是因為政府的補貼政策而選擇使用微笑單車。為此,我們先是透過對上述比例進行敏感度分析來暸解價格調升對於顧客服務水準的影響程度。然而,為了更近一步提升我們對於顧客服務水準在價格調整下的預測的準確度。我們透過引入報童模型來技巧性地解決了補貼比例的問題。 在本篇論文後段,我們假設微笑單車公司採用報童模型來最適化自己的服務水準。此一假設意味著,我們從原始資料模擬出來的顧客服務水準皆是廠商在最佳化下的結果。這樣的假設大大的減少了我們衡量有多少比例的需求是來自於政府的補貼政策的難度。原因在於,在報童模型中為了求出上述之比例,我們需要的變數僅有產品售價、顧客服務水準、剩貨處置價格以及邊際成本。產品銷售價格和剩貨處置費用皆是已知資訊,而顧客服務水準在我們的假設下也以成為已知資訊,邊際成本的部分我們僅需進行簡單的數字假設就能求得上述段落所提及的比例資訊。此創新應用不僅解決了我們資訊不足的問題,也提升了我們動態定價策略在調整價格後對於顧客服務水準預估的精準度。此一方法也避開了需要求得需求的分配的複雜函數估計過。

並列摘要


In this paper, we try to solve two problems, one is the mismatching problem of demand and supply of rental bike caused by the subsidy policy of the government, and the other one is the missing information of the proportion of demand from the discount price in the aggregated panel data set of Youbike. To manage the demand of rental bike, we evaluate the utility of the rental bike users by a general model, proposed by Sanghak Lee and Greg M. Allenby (2013). Based on the estimation results, we find that most of the Youbike users are sensitive to price, which means a subtle increase in price can induce a huge reduction of demand. The inventory, available bikes in each station, can only cause a minor impact on customer service level changes because the maximum inventory level of each station is low and constrained by the station capacity and the sidewalk size. When it comes to the problem of missing information, we introduce a new application of newsvendor model to obtain the missing information, proportion of demand from discount price, which significantly impacts the customer service level of Youbike stations. The outcome helps us to apply the dynamic pricing strategy more precisely. As a result, the problem of missing information from the aggregated panel data set can be easily solved by the newsvendor model. In addition, it is better to manage the demand of rental bike via dynamic pricing strategy rather than inventory management as for the case of public rental bike.

參考文獻


Abrate, G., Nicolau, J. L., & Viglia, G. (2019). The impact of dynamic price variability on revenue maximization. Tourism Management, 74, 224-233.
Anderson, E. T., & Simester, D. I. (2004). Long-run effects of promotion depth on new versus established customers: Three field studies. Marketing Science, 23(1), 4-20.
Beggs, S., Cardell, S., & Hausman, J. (1981). Assessing the potential demand for electric cars. Journal of econometrics, 17(1), 1-19.
Beviá, C., Quinzii, M., & Silva, J. A. (1999). Buying Several Indivisible Goods, Math. Social Sci. 37(1), 1-23.
Bhat, C. R. (2008). The multiple discrete-continuous extreme value (MDCEV) model: role of utility function parameters, identification considerations, and model extensions. Transportation Research Part B: Methodological, 42(3), 274-303.

延伸閱讀