Abstract:
Capacity is an important parameter on the basis of which other operational measures are calculated for signalised intersections. The capacity is estimated based on the green time ratio and saturation flow rate. The basic model of traffic operations at signalised intersections is based on an assumption that, when a signal changes to green, the flow across the stop line increases rapidly to a maximum queue discharge rate, also termed as saturation flow rate, which remains constant until either the queue is exhausted or the green period ends. In recent years, this assumption has been challenged by some field observations. These observations depict fluctuations in the queue discharge rate for different queue positions and show a marginal increase in the queue discharge rate along the queued vehicles’ position. This research investigates the reported variations in queue discharge rate at signalised intersections. To verify this proposition, a pilot study was conducted to investigate the queue discharge behaviour at signalised intersections based on video data collected from one of the busiest at-grade signalised intersections in Auckland. A comparison was made between the queue discharge rate observed in the field and the saturation flow rate estimated based on the methodologies proposed in the HCM (2000) and the Australian Road Research Board Report 123. The results from the pilot study conducted in Auckland were similar to those observed in Taiwan and the USA A detailed field study was conducted at six signalised intersections in Auckland to collect data verifying the variable nature of queue discharge rate at signalised intersections. The observed trends are then modelled using empirical models and assessed by comparing against the models developed using the Gene Expression Programming (GEP) algorithm. The proposed model shows compatibility with the existing traffic signal design methodologies for calculating different performance measures. More importantly, it shows prospects to overcome the shortcomings of the traditional method for practical applications. The proposed model is investigated for the purpose of real world applications, which will require calculations relating to capacity and cycle time. The model is implemented to compute uniform delay using a modified formulation for a D/D/1 case. The modified uniform delay model is compared against the existing delay models based on traditional methodologies. The results revealed that the modified model can overcome the deficiencies of the existing ones. Finally, a microscopic car-following model is proposed to replicate the increasing queue discharge behaviour observed at signalised intersections. The proposed model is verified using the GEP algorithm. It exhibits the potential to overcome the shortcomings of the existing car-following models in replicating the queue discharge behaviour observed in the field.