A Random Forest Incident Detection Algorithm that Incorporates Contexts

Jonny Evans,Andrew Hamilton,Ben Waterson

doi:10.1007/s13177-019-00194-1

Abstract

A major problem faced by state of the art incident detection algorithms is their high false alert rates, which are caused in part by failing to differentiate incidents from contexts. Contexts are referred to as external factors that could be expected to influence traffic conditions, such as sporting events, public holidays and weather conditions. This paper presents RoadCast Incident Detection (RCID), an algorithm that aims to make this differentiation by gaining a better understanding of conditions that could be expected during contexts’ disruption. RCID was found to outperform RAID in terms of detection rate and false alert rate, and had a 25% lower false alert rate when incorporating contextual data. This improvement suggests that if RCID were to be implemented in a Traffic Management Centre, operators would be distracted by far fewer false alerts from contexts than is currently the case with state of the art algorithms, and so could detect incidents more effectively.

Highlights

Road congestion places a burden on citizens worldwide
It could be seen that holiday features affected detectors throughout the city, but event contexts were only included at detectors on routes into and out of the event location
The following sections evaluate the performance of RoadCast Incident Detection (RCID), and make comparisons to an existing Incident detection algorithms (IDAs), RAID

Summary

Introduction

Road congestion places a burden on citizens worldwide. In 2016 alone, road congestion cost U.S drivers more than $295 billion, U.K. drivers £30 billion, and German drivers €69 billion [1]. A major cause of this congestion is from incidents [2]. The causes of disruption in traffic conditions can be categorised into two main types, incidents and contexts. Incidents are defined as unexpected events that disrupt traffic conditions [3]. Contexts are referred to as external factors that are planned in advance or predictable, and could be expected to influence traffic conditions at a particular time in the future. Examples include planned roadworks, sporting events, rush hours, schools closing and weather conditions. The key difference is that contexts could be expected to occur, but incidents are inherently unexpected

Objectives

Methods

Results

Conclusion