Estimating Taxi Demand-Supply Level using TaxiTrajectory Data Stream
Taxis provide a flexible and indispensable serviceto satisfy the urban travel demand of public commuters. Un-derstanding taxi supply and commuter demand, especially theimbalance between the supply and the demand, would directlyhelp to improve the quality of taxi service and eventually increasea city’s traffic system efficiency. In this paper, we consider thetaxi demand from a region during a period of time to includetwo parts: satisfied demand, i.e., passengers successfully receivetaxi service during this period of time, and unmet demand,i.e., passengers are still waiting for taxi service. To properlyestimate the demand-supply level (short for “the level of the taxidemand vs. supply imbalance”), we propose a novel indicator thatreflects how fast an available taxi is taken in any given region.Accordingly, we design and implement a taxi analytics systemto provide such information in near real time. Finally, we usethe passenger waiting time survey data and the taxi streamingdata to validate the proposed indicator on the built taxi analyticssystem https://srisivasakthitravels.com/
Taxi, providing personalized and efficient point-to-pointservice, is a popular form of public transport in many cities.Especially, in the densely populated Asian cities, taxi service isoften at a low cost as well, and thus many people rely on taxisfor their daily trips, to fulfil their business, social, and familyactivities. The taxi service quality not only affects the levelof satisfaction of locals to the land transportation services, butalso has an impact on the experience of visitors and touriststo the convenience of the transport facilities
Taxis provide a flexible and indispensable serviceto satisfy the urban travel demand of public commuters. Un-derstanding taxi supply and commuter demand, especially theimbalance between the supply and the demand, would directlyhelp to improve the quality of taxi service and eventually increasea city’s traffic system efficiency. In this paper, we consider thetaxi demand from a region during a period of time to includetwo parts: satisfied demand, i.e., passengers successfully receivetaxi service during this period of time, and unmet demand,i.e., passengers are still waiting for taxi service. To properlyestimate the demand-supply level (short for “the level of the taxidemand vs. supply imbalance”), we propose a novel indicator thatreflects how fast an available taxi is taken in any given region.Accordingly, we design and implement a taxi analytics systemto provide such information in near real time. Finally, we usethe passenger waiting time survey data and the taxi streamingdata to validate the proposed indicator on the built taxi analyticssystem https://srisivasakthitravels.com/
Taxi, providing personalized and efficient point-to-pointservice, is a popular form of public transport in many cities.Especially, in the densely populated Asian cities, taxi service isoften at a low cost as well, and thus many people rely on taxisfor their daily trips, to fulfil their business, social, and familyactivities. The taxi service quality not only affects the levelof satisfaction of locals to the land transportation services, butalso has an impact on the experience of visitors and touriststo the convenience of the transport facilities
While there are a large number of taxis on the road, ithappens that passengers have difficulties in finding availabletaxis in certain areas and hours. Fortunately, with advances intechnology in sensor and wireless communications, nowadaystaxis are commonly equipped with GPS (Global PositioningSystem) receivers and wireless communication components(such as GSM, GPRS). With these devices, taxis easily reportsthe location and status (such as FREE, HIRED) information tothe central server, for the taxi operators to explore to improvetheir quality of service, by real-time location tracking andcall dispatching to meet the needs of time-sensitive users,etc. To make taxi drivers and commuters more aware of thereal-time demand and supply, Land Transport Authority (LTA)of Singapore has also recently launched a mobile app calledTaxi-Taxi@SG [11], which displays the locations of availabletaxis to commuters in real time and also enables commutersto broadcast their current locations to drivers.How to match the supply with the demand is of criti-cal importance, for the interest of both passengers and taxidrivers, both taxi operators and land transport authorities.There are many existing research works in this direction, suchas modelling the spatio-temporal structure of taxi services[3], [5], predicting the demand at taxi stands [10], or evenrecommending the next-passenger to the drivers [6]. There arealso literatures focusing on passengers’ perspective, such asto predict where to find empty taxis [12]. However, all theabove works are understanding the number of demand as thenumber of historical pickups, while in realities the histori-cal pickups are constrained by the historical supplies. Therecould be unmet demand (i.e., unsatisfied passenger demandstill waiting for a taxi service) not captured by the pickuprecords. Moreover, the above works are either suggesting taxi’spassenger-look-up strategy based on status of other taxis andhistorical pickups, or suggesting passengers based on locationof free taxis, without focusing on the relationships of the twoparties. Therefore, in this paper, we will endeavour to infer thedemand and supply imbalance, by bearing in mind the demandas both successful pickups and unmet demand, to make a morecomplete picture for the passengers and taxis. According to ourknowledge, this is the first piece of work that deals with thedemand-supply relationship in regions
No comments:
Post a Comment