Obtaining, analyzing and manipulating traffic data for a vehicle routing application.
Vehicle routing problem concerns the service of a logistics company. How things are delivered from one or more depots in a given set of vehicles which move on a given road network to a set of customers. The solution is a set of routes which minimizes global transportation costs. Solutions are obtained based on known characteristics of routes such as travel times or distances between customers. Even though distances are fixed, travel times fluctuate. They are heavily dependent on punctual occurrences, like weather related events or traffic accidents, but also on the period of the year, month and day. In this talk, I will discuss how travel time data was obtained from a TomTom API, how we overcame API limitations by manipulating data and I’ll provide an analysis of how travel time changes with the period of the year, month and day.