Ride dispatch and matching is the part of a ride-hailing app that finds the nearest suitable driver for each ride request, and predicts demand minute by minute so that drivers are in the right place. Dispatch is the industry's word for choosing which driver gets which ride.
The choice is not only about distance. The software estimates the time each driver needs to reach the rider, using maps and live traffic from location APIs. It also considers whether the vehicle fits the request, for example a cab, an auto or a bike taxi, and whether the driver is likely to accept. A driver who often declines or cancels is less likely to be offered the ride.
Consider two drivers near a railway station in Chennai. One is 500 metres away, but the road between them and the rider is cut by a flyover with no turn, so the drive is 6 minutes. The other is 1.2 kilometres away on the same road as the rider, and needs 3 minutes. A simple "nearest driver" rule would pick the first. A dispatch system that estimates the time of arrival picks the second.
When many requests arrive at once, a good system does not serve them one by one. It looks at all the waiting riders and all the free drivers together, and chooses the set of matches that gives the lowest total waiting time. This is called batching.
The second job is prediction. The software uses past trips, the weather and events to forecast where requests will rise in the next few minutes. This is demand forecasting. It then shows drivers where the demand is, or raises the fare there with dynamic pricing, so that more drivers move into the area.
The quality of dispatch decides how long riders wait and how much drivers earn without driving empty. It is also what makes the network effect work.