Loading...
Comparative study of threshold selection for incident detection based on California algorithm
Author(s)
Puangnak, Korn
Date Issued
July 1, 2019
Type
Conference Paper
Abstract
This paper purposes the Comparative Study how to find the threshold for the Automatic Incident Detection by California Algorithm based on the 3 types of the general incidents from the simulations on the 3 lane-highways. Each type of the happening incidents has an effect on the traffic measurement and towards the performance of the incident detection and the alarm rate because the fixed threshold is suitable for each type of an incident. So this comparative study will use Statistic Mathematic Technical to calculate the comparison of the proper threshold and the best threshold by using the breadth-first search Algorithm to make California Algorithm able to detect all types of the happening incidents. The study and test of the proper threshold using Statistic Mathematic methods shows that the Mode Technical can calculate the threshold that enables the automatic incident detection process using California Algorithm to have the Performance Index close to the complete search Algorithm that gains the best performance index. Without calculating the past situation and the situation of the new incidents, the rapid process will be suitable to apply for the real-time situation.
Citation
Proceedings of the 16th International Conference on Electrical Engineering Electronics Computer Telecommunications and Information Technology Ecti Con 2019, 911-914, 2019