As a smart transport technology partner, our business involves pooling passengers in shared vehicles and generating optimal routes to pick them up and drop them off. A key component of our mobility engine which generates routes are speed maps. Speed maps are used to estimate the shortest path between two locations and its corresponding travel time. For our algorithm to be highly accurate when generating trips, the speed maps used need to account for several things.
Firstly, there is a difference in speeds across various vehicle types and road types. For example, a large bus can’t travel as fast as a van, and also won’t be able to make u-turns. There are also restrictions on certain roads for particular vehicle types. Speed limits are also different in different types of roads.
Next, we have to take into account the traffic conditions depending on the time of day, and the day of the week. The time of the year also matters… For example, the traffic conditions on a public holiday that falls on a weekday tend to be lighter than during a normal weekday during commute hours. As such, speed maps should be derived from reliable vehicle trip logs - be it from GPS or odometric means. At SWAT, the speed maps we use are regularly re-calibrated with the latest data, ensuring the accuracy of our routing engine.