Intelligent taxi scheduling system

Abstract

The invention provides an intelligent taxi scheduling system. An operational method of the system includes the steps of first, cellphones of the passengers and drivers send call / availability requests (containing GPS information); second, a server runs a people-car matching algorithm based on bipartite graph matching to match drivers and passengers, and sends scheduling arrangements to the corresponding drivers; third, if the drivers take scheduling, the server sends requests of confirmation to the corresponding passengers; if the drivers do not take scheduling, the second step is repeated; fourth, after picking up the passengers, the drivers send request conformation to the server. The intelligent taxi scheduling system has the advantage that taxies can be efficiently scheduled through the people-car matching algorithm and a breach monitoring mechanism.

Description

technical field [0001] The invention relates to the field of communication technology, in particular to an intelligent dispatching system for taxis. Background technique [0002] Taxis have occupied a very important position in modern urban traffic because of their convenience and speed. However, because taxis do not have fixed routes and schedules like public transportation such as buses and subways, taxis are in a state of no passenger or no load in many cases. Statistics show that taxis spend more than 50% of their working time looking for or waiting for passengers, which is longer than the actual loading time. Taxi running empty is not only a huge waste of fuel, but also aggravates traffic congestion. Despite this, passengers often find it difficult to wait for an empty taxi on the road, because passengers and drivers do not know each other's location and may pass each other even if they are only one road away. Therefore, using some scheduling methods to help passenge...

AI Technical Summary

Problems solved by technology

However, there are generally two defects in these systems: First, the system processes passenger requests according to a first-come-first-served policy, that is, assigns a taxi to the passenger who sends the request first, and this scheduling strategy often causes unreasonable allocation. Case

Second, passengers may happen to encounter an empty car after calling for a taxi, and it is difficult for the system to detect such violations and notify the taxi driver in time

[0004] Existing technology is searched and found, Der-Horng Lee etc. published Taxi Dispatch System Based on Current Demands and Real-Time Traffic Conditions on the Journal of the Transportation Research Board in 2004 (based on real-time demand and the taxi dispatch system of traffic condition ) proposed a taxi dispatching system based on the first-come-first-served dispatching strategy. This dispatching strategy cannot achieve the global optimal match. At the same time, the system does not consider the default behavior of passengers, and there are large defects in actual use.

In A Collaborative Multiagent Taxi-Dispatch System published on Automation Science and Engineering, IEEE Transactions in 2010 by Kiam Tian Seow et al., it was proved that pairing multiple passengers with multiple drivers can reduce the Scheduling costs, but the paper does not give an algorithm with polynomial time complexity. There are tens of thousands of taxis running at the same time in a large city, and an algorithm with exponential time complexity is not feasible in practice; in addition, the system does not have Consider passenger default

Images