Investigation of Cyclic Vehicle Queue and Delay Relationship for Isolated Signalized Intersections

Delay is the one of the design criteria that has been used for performance evaluation of signalized intersections. In pre-timed control, vehicle delays can be minimized or reduced by proper design of signal timings and phasing. But vehicle delays include many parameters such as signal timing, number of phases, vehicle headways, saturation flow, queueing etc. Among these parameters vehicle queue is formed by unbalanced signal timings or unexpected demand values. On the other hand, headways of vehicles can be effective on queue forming especially in discharging case. It is also related to driver behaviors. In this paper, relationship of cyclic vehicle queue and vehicular delay is investigated considering different signal timings and phase sequencing. The data are obtained from observations that are made at urban intersections of Denizli city, Turkey. Regression analysis is used for relationship and statistical tests are applied. The MuLReD (Multiple Linear Regression Analysis based Delay Estimation) model is developed and significance of the model is proved by statistically. Adjusted R2 value of MuLReD Model is obtained as 0,95. On the other hand, delay values obtained from the MuLReD Model and Akcelik Vehicle Delay Formula are compared with each other. As a result of comparisons, Mean Square Error (MSE) values for Akcelik Equation and the MuLReD Model are determined as about 112 and 7 respectively. These results show that outcomes obtained by the MuLReD Model are closer to field observations. Discussions about the present delay formulas, are also reported in the paper.