Traffic condition estimation for delivery planning in city logistics

In classical Vehicle Routing Problem, algorithms are developed based on travel distance or travel time estimated from an assumed constant travel speed. However, in city logistics, due to a growing amount of traffic and a limited capacity of the road network, travel speed between two locations no longer remains a constant but location and time dependent. In this paper, we propose a framework to visualize the traffic condition at different times and different locations of Singapore through estimating the travel speed between any two locations based on Google traffic data. We first divide Singapore into zones according to the Postal Code System. Then travel speeds within and between zones are estimated by randomly sampling pairs of locations and calculating the travel speed between them. The sample size, estimation accuracy, relationship between average and standard deviation of travel speed and travel distance are studied in details.