Exploring time varying shortest path of urban OD Pairs based on floating car data

Traffic state (for instance, link travel time, velocity, volume) in the links of the urban road network should be taken into account while calculating the shortest path in real application situations. However, the traffic state of the urban road network varies in different places and at different time. Constructing a time varying road network is critical to explore the time varying shortest path. The travel time of a road link can be described by some probability distribution functions, or estimated through using real time traffic information and short-term traffic forecast. Therefore, the time varying road network can be built based on the varying travel time captured by various approaches. This paper proposes an effective approach to investigate the time varying shortest path based on floating car data. First, all space-time paths of floating cars are reconstructed after map-matching process of floating car data. Then, to capture the velocity and the distribution of vehicles that reflects traffic density, all space-time paths are sliced at every minute to build up the traffic snapshot of the road network. A congestion index is introduced to quantify the traffic state of each snapshot at any location, which is calculated based on velocity and traffic density. By virtue of these time-series traffic state snapshots, the time-varying transport network was established and shortest path of OD pair at different departure time was calculated by A* shortest path algorithm. A case study with one day floating car data in Wuhan, China was implemented, and the experimental prototype provides the capability of exploring the time-varying shortest paths by time of a day in the time varying transport network.