Reducing Travel Time by Incident Reporting via CrowdSourcing

The contribution of this work is the creation of a novel system that enables motorists who witness incidents to submit reports to our system via the web. These reports are aggregated, validated and verified automatically. Then, they are used to update the road network graph. In this work, we designed and implemented an incident reporting system whereby users can report an incident such as an accident or construction on a road network. We extended the FreeSim simulator to accommodate our incident reporting system. Experimental results showed that our system is capable of reducing the travel time of users. We also presented our verification algorithms that are used to verify that reports are fact. Current approaches to congestion detection such as loop detectors probe vehicles, and video image detection may not be available on arterial streets. These technologies may not be available in countries whose transportation budget is low. Our model is less expensive, easy to implement and can work in any environment (e.g. extreme weather). This work is dependent on people’s incident response input and not from sensor signals converted to traffic measurements. To the best of our knowledge, this approach is the first to consider web-based incident Crowd Sourcing with automatic incident verification. Other driver based models are telephoned based.