Intelligent adaptive taxi dispatching system using smartphone

Abstract

A human taxi dispatcher has to figure out how to make optimal use of the resources at hand while providing clients an excellent service. As the company grows and client demands increase, human dispatching proves to be inconsistent and inefficient, therefore it would be for the best to introduce machine learning algorithms to schedule and optimally distribute idle taxis in statistically popular hotspots. A complete, distributed, efficient taxi dispatching system with selfadaptive features has been implemented which allows the users to communicate with the server using smart phones. A model simulating taxis and client demands is implemented to highlight the difference in efficiency between the human and computerised system.