Spatial and Temporal Analysis of Mismatch between Planned Road Infrastructure and Traffic Demand in Large Cities

As large cities evolve in time, the traffic demand and the road network adapt to the mutually presented changes by each other. As a result of this process, previously planned roads and intersections that were designed according to some optimality criteria at the time, turn out to be sub-optimal when traffic conditions change. This paper presents a method that can be used in order to identify intersections whose capacity is no longer in correspondence with the demand of vehicles on them and the choices agent make at those locations. Using real data from a survey describing the travel patterns of people in the city of Singapore we are able to model the routing choices of commuters and simulate the traffic demands on the road network. After calculating the turning probabilities on every intersection we are able to compare the traffic demand for every turn with the planned physical roads' capabilities. Furthermore we define a measure, which quantifies the deviation of the whole road network from the ideal demand-calculated values. We use these measures to evaluate the temporal and spatial profile of the mismatch between the roads and the demand for them. The measure is designed such that it is universal in nature and invariant to the absolute values of the traffic flows in the city. It can, therefore, be used to compare the proper utilization of road networks among different cities.