雙占市場下路外停車場之動態預約訂價模式構建

Abstract

台灣近年來私有運具使用率逐步攀升，在有限的停車資源和缺乏彈性的價格策略下，都市商業中心的停車需求並無得到良好的管理。在停車需求尖峰時段，停車業者採取先到先停的方式，使得許多停車消費者願意付出較高的停車費率卻受限於車位資源的限制，因此，如何有效地管理停車需求，並且透過動態預約訂價來得到最佳的停車期望營收是一項重要議題，此外，位於都市商業中心通常不僅存在一家的停車設施，因此，在停車場預約訂價之策略管理上，必須考慮到其他周邊停車設施之影響。 在此基礎上，本研究利用動態規劃、隨機過程和羅吉特選擇模式，於需求不確定性下，推導雙占市場下路外停車場之動態預約訂價模式。本研究考慮顧客平均到達率、各時段剩餘車位數、剩餘預約時間和消費者願付價格，求解各時段動態預約價格，以追求最佳總期望營收。本研究不同於其他大眾運輸系統(例如，航空、鐵路和公路客運)的座位預約模式，其不同在於停車場消費者所欲停放之時間不一致，使的模式推導求解複雜。而本研究考慮兩種市場型態:獨占性市場和雙占性市場兩種。為簡化此問題，本研究研提單小時預約停車模式做為探討，模式係假設所有停車者均停放1小時，而停車消費者若欲停超過一小時可視為分開解決(即每個時段會有不同的停車費率)。因此，每小時之單一車位可滿足一位停車顧客之需求，進而可將停車預約問題依據不同時段切割為各自獨立的車位預約問題。 至於雙占市場競爭模式則假設停車者於欲停車時，比較兩廠商不同因素(例如，價格、步行時間和現場排隊等因素)，進而選擇停車場。為推導模式，本研究假設停車消費者到達率符合卜瓦松分配，而消費者願付價格分配則符合為韋伯分配。並且本研究以一簡例來驗證模式之適切性及可用性。結果顯示出獨占廠商最佳預約價格隨著剩餘時間減少和剩餘車位數的減少而增加，雙占廠商則會根據競爭廠商的價格而降低自身預約訂價。

With limited parking space and inflexible pricing policy, the parking demand in the central business district has not been well-managed. Many parking needs willing to pay higher fee are rejected for parking due to no empty parking lots left. Thus, how to effectively manage parking demand and to generate even more parking revenue through dynamic reservation pricing is important. Additionally, since in the parking market, there is usually more than one parking facility in the central business district, the pricing policy of a park facility should consider that of other neighboring parking facilities. Based on this, this study proposes dynamic reservation pricing models for off-street parking facilities based on dynamic programming, stochastic processes and logit choice model under demand uncertainties. The proposed model aims to maximize total parking revenue by dynamically determining the reservation price according to customers arrival rate, available parking lots, elapsed time for parking, and willingness to pay of drivers. Unlike traditional seat reservation models of other transportation systems, such as airline, railway and intercity bus services, parking demand is defined by various parking durations during the reservation period, resulting in the complexity of the model derivation and solving. Two types of markets are under consideration: monopoly and duopoly. To simplify the model, the one-hour model is developed and applied. This model assumes that all drivers are homogenous with the same parking duration of one hour. Users with more than one-hour parking demand would be view and solved separately (i.e. each of parking hours may have different parking fees). As a result, during a one-hour period, one user can be satisfied with one parking space, and this can further divide the parking space reservation into independent parking space reservation according to different time period. As for the duopoly market model, this model assumes that users choose the parking lots according to their charges, the walking time, and the time required for queuing. To derive the optimal dynamic reservation price, non-homogeneous Poisson process combining by Poisson arrival rate and Weibull distributed willing-to-pay price are also assumed. To demonstrate the applicability of the proposed model, a case study on a numerical example is conducted. The analytical results show that the optimal dynamic reservation price increases as time elapses and the number of remaining parking lots decreases. The reservation price in the duopoly market decreases due to the competition.