---

Delay is one of the principal measures of performance used to determine the Level of Service (LOS) at signalized intersections and several methods have been widely used to estimate vehicular delay. Very few studies only have been carried out to estimate delay at signalized intersections under mixed traffic conditions prevailing in developing countries like India. In the present study, various problems associated with delay estimation under mixed traffic conditions in a developing country (India) and the methods to over come them were discussed and an attempt was made to improve the accuracy estimating the same. Five isolated signalized intersections from a fast developing industrial city located in TamilNadu, India were chosen for the study. Site specific PCU values were developed considering the static and dynamic characteristics of vehicles. Saturation flow was also directly measured in the field for the prevailing roadway, traffic and signalized conditions and expressed in PCU/h. Control delay was also measured following HCM 2000 guidelines. Later, this was compared with that estimated from the theoretical delay model. Even after taking several measures, good correlation between observed and predicted delay could not be obtained. Therefore, in the present scenario field measured control delay was taken into account to define LOS. A new criteria for Indian cities recently published in the literature was used to assign LOS grades of study intersections and found to be better reflecting the field conditions.

---

The gap acceptance behavior of drivers at U-turn median openings is studied by considering the critical gap which cannot be obtained directly by field measurements. A thorough investigation on critical gap estimation is carried out by collecting the data sets from median openings at 4-lane, 6-lane and 8-lane roads of Hyderabad City, India. Wide difference (10% - 42%) among the critical gap values estimated by the application of existing methods shows the limitation to consider for the mixed traffic situations. In order to address this issue, recently developed method Influence Area For Gap Acceptance (INAFOGA) which is based on clearing behavior of drivers at unsignalised intersections is modified and applied considering the merging behavior of U-turn vehicles at median openings and named as ‘Modified INAFOGA’ method. Modified INAFOGA method is compared with probability equilibrium method through paired-sample hypothesis (t-test) and result revealed that difference in mean values 0.009 signifies that both methods are comparable. Difference in critical gap values obtained from the box plots and radar charts indicates that Probability equilibrium method is not suitable to address the behavior of U-turn vehicles at median openings under mixed traffic conditions. These observations coupled with higher critical gap values validate the fact that ‘modified INAFOGA’ method is indeed appropriate under mixed traffic conditions.