評估貨運車隊收集路網旅行資訊之研究

Abstract

路況資訊的收集在先進旅行者資訊系統是相當重要的，使用探針車是收集方式之一，所以各路段所需探針車數目則有必要瞭解。在已知車隊規模與需求分佈之下，本研究使用貨運車輛為兼職探針車，探討其在營運中可否提供各路段所需車輛數目，目的是為有意招募貨運車隊的路況資訊提供者建構一套模式，估算其所欲招募車隊的各路段車輛行駛頻率。在路況資訊的提供存在10%的相對誤差與95%信賴區間條件下，為估算所欲招募車隊的車輛路徑行駛頻率，研究中首先決定每部貨運車輛可以服務的需求點數目，接著進行各部車輛的需求點分配與車輛路徑指派，最後則是統計車輛路徑行駛頻率做為模式建構依據。 以檢定方法驗證得知本研究所建構模式的車輛路徑行駛頻率與標竿範例的車輛路徑行駛頻率沒有顯著的差異，表示本研究所建構模式與標竿範例所得的車輛路徑行駛頻率是一致的。

Collection of the travel information plays an important role in Advanced Traveler Information System (APTS). The information could be collected by probe vehicles, so it’s necessary to know the number of probe vehicles that be assigned. This paper uses the freight vehicles as the part-time probe vehicles and focuses on the frequency that the freight vehicles in operation could support in the situation that the number of the freight vehicles recruited is known and the demand is deterministic. Our purpose is to construct a model that can estimate the frequency the freight vehicles recruited by the planner could support. This model is based on decision of the number of demand nodes that a vehicle need to service, the distribution of demand for each vehicle, assignments and calculation of the frequency with a relative error 10% in providing the information and the 95% confidence interval. Hypothesis testing indicates that the difference of frequency between this model and VRP instances is not significant, this means the frequency from this model and VRP instances are the same.