Methodology for Simulating Manual Traffic Control

Abstract

Manual traffic control (MTC) is a key part of managing traffic during emergencies and planned special events. Despite its long history, there has been little, if any, research on how to model MTC effectively. It is commonly represented as a version of actuated signal control. Although this method is useful, it has significant shortcomings because it does not adequately represent the variability of police officer control actions under field conditions. This paper presents the results of recent research to develop an MTC model and integrate it into a traffic simulation system. Here, the process of MTC is represented by police officers’ decision making in relation to a system of discrete choice equations (logit models) that compute signal phase length and green-time allocation as a function of demand, directional priority, phase length, and gaps in the approach traffic streams. The MTC discrete choice model was validated with the use of various data sets to show that it computed phase lengths and allocated green time within a 95% confidence level compared with field observation.