The quality of service provided by signalized intersections can be expressed in a number of ways. The volume-to-capacity ratio, the probability of discharge, the load factor, the average delay, the number of stops, the length of queues, or composite measures such as level of service or cyclic flow profiles are frequently used. Among these criteria, the number of stops and the average delay have a special importance for coordinated systems because they are directly perceived by drivers. As a result, they greatly affect public acceptance of the system design. "Delay" is not a simple or exact term. It is defined as the difference between the actual travel time and the unimpeded travel time through a roadway section upstream of (and sometimes including) a signalized intersection. Driver expectations and behavior, speed distributions, and the arbitrary nature of the distance reference are some of the factors influencing the delay. As a result, its measured values are difficult to duplicate. Nevertheless, delay not only remains the most important signal performance criterion but can also be used as an indicator of the quality of signal coordination. The basic idea is simple: when the vehicular progression through a traffic signal is good, most vehicles will encounter no delay. On the other hand, if most platooned vehicles are stopped and delayed, the coordination of signals is suspect. This idea has been employed in a survey system developed at the University of Alberta. Portable field equipment and subsequent computer programs assist in the evaluation of signal coordination by using delays at individual intersections along a coordinated route. The survey method is based on a re-creation of a time-space diagram for a signalized intersection approach lane. The spatial reference extends from an arbitrary point upstream of signal queues to the stopline. The difference between the actual and unimpeded travel times determines individual vehicle delays. Individual vehicle information is summarized in the form of delay distributions and cyclic flow profiles. This paper describes the survey and employs practical examples to illustrate its use.
Abstract