Use of automatic incident detection algorithms (AIDA) in advanced freeway management systems has been sporadic and scarce. The current study investigates the causes of such limited implementation. A survey was carried out among the system managers, operators and end users, as well as the decision makers who set the operational policies and the priorities for future system enhancements. The survey responses point to a general consensus that the unacceptably high rates of false alarms generated by available incident detection algorithms is the major deterrent. This study not only provides an understanding of the causes of the limited implementation of incident detection algorithms, but also allows a direct comparison between the conventional incident detection technologies and automatic incident detection technology on the basis of their performance. Despite the lackluster performance of AIDA to date, 90% of the survey respondents feel that the current methods of incident detection are insufficient either at present (70%) or will be so in the future (20%). This finding alone motivates a need to redouble research efforts aimed at developing robust and accurate automatic detection methods. In this regard, the paper presents promising directions to overcome the past AIDA deficiencies.
Available at: http://works.bepress.com/daiheng_ni/31/