››› Session 31

A General Specification for an Incident Management System

Gregory Smith - Project Sales Manager, Mi Consulting Group (Australia) Pty Ltd

An Incident Management System can be the "brain" behind a range of incident management resources and programs. Like a "brain" the incident management system must recognise and incident, analyse it, determine the best course of action, and put those actions into effect. This paper looks at what makes up an Incident Management System - how it is typically structured, what functions it performs, and how it interacts with other transport and traffic management systems.

››› Session 32

Incident Management - A Tunnel Perspective

Paul  Higgins - Managing Director, Excel Technology Group

The consequential effects of vehicle congestion and deviant vehicle behaviour within the close confines of tunnels, has been a major concern for traffic authorities since tunnels were first promoted as a traffic engineering solution. Having entered the tunnel confines, there is little chance of alternate routing and hence the ability of a vehicle monitoring system to accurately and expediently detect traffic variations, is essential to effective traffic management and the safety of motorists using the tunnel. The performance level expectation is set according to the degree of risk associated with the application. While the desired outcome of performance based on minimising false alarms and maximising incident detection in the shortest possible time applies to a freeway, tunnel based incident management requires additional verification of vehicle behaviour. The difficulties with performance expectation arises when the system has to provide the same level of incident detection performance when the number of vehicles may vary from 5 vehicles an hour to 2500 vehicles per lane regularly within a twenty-four hour period. Volume-occupancy incident management models while satisfactory at high vehicle flow rates, perform less than satisfactorily at very low volumes. A freeway model may not be concerned with low volumes, that is, low volumes will never produce congestion however, within a tunnel, more critical incidents associated with deviant vehicle behaviour emerge at low volumes. It is imperative that under any vehicle flow conditions in a tunnel, that a vehicle travelling contra flow, a stopped vehicle causing congestion or queued traffic beyond the visibility range and a vehicle travelling either considerably faster or slower than the preset tunnel speed cause a predetermined hazard warning response immediately.

››› Session 33

Automated Incident Detection Algorithms

Hussein Dia - School of Engineering, The University of Queensland

The high cost of congestion caused by incidents such as accidents, disabled vehicles, construction work and other events that result in a capacity reduction of roads has prompted a growing worldwide interest in developing efficient and effective incident management programs. The success of incident management programs will depend to a large extent on the development of reliable and efficient automated incident detection (AID) algorithms. These software programs analyse data collected from traffic detection devices to determine whether an incident has occurred and alert operators in the traffic control centre to abnormal traffic conditions. This paper will provide a background to automated incident detection algorithms and describe some of the most widely used freeway AID models. Common measures of performance and criteria used to evaluate the efficiency and effectiveness of these algorithms will then be presented. The traffic detection, software and hardware requirements needed for the development and implementation of these models will also be described. A comparative evaluation between a number of AID algorithms is then presented. The paper will conclude by highlighting current state-of-the-art techniques in the development of AID algorithms on both freeway and arterial roads and the benefits to be gained from the implementation of quick and reliable incident detection systems.

››› Session 34

Acoustic Detection Techniques

Andrew Smith - , Tyco Integrated Systems

Acoustic detectors use varied technologies. Recent techniques are very different from earlier techniques and detectors using these have new capabilities. Some of the features that are closely related to the different technologies are identified and the mechanisms of active, passive, proximity and phase coherent technologies are examined to show how they yield their particular features. The features of the new generation acoustic detectors now make them candidates for incident detection systems which were previously the domain of loops, radar and video detectors.