PREDICTIVE MODELLING OF TRAFFIC FLOW AT URBAN ROAD INTERSECTIONS: A CASE STUDY OF AKURE METROPOLIS, NIGERIA

Abstract

Traffic congestion on major roads consequent upon existing bottlenecks at intersections is a major problem especially in metropolitan areas. This research was thus aimed at formulating models that could be used to predict traffic flow and assist in effective traffic management at intersections. A reconnaissance survey was conducted after which seven intersections consisting of three Tee intersections, two Cross intersections and two Rotary intersections were selected for detailed study on the basis that they are critical to traffic flow in the study area. Data on geometric features were collected using odometer (advance distance measuring equipment), while traffic parameters were captured and metered using cine camera placed at an elevated vantage point from the road section during the morning and evening peak period between 7:30-8:30 am and 4:30-5:30 pm and off-peak period between 11:30 am-12:30 pm during week days, and the data collected were later extracted by replaying the cine-camera. An appraisal of the intersections showed that Tee and Cross intersections have an average carriageway width of 7.19 metres which is less than 7.30 metres specified by the Federal Highway Capacity Manual in Nigeria. However, the variation reduces the traffic flow at those intersections which increases the delay. Also, the Level of Service (LOS) of Tee intersection was “E” while that of Cross intersection was “F”. This indicates that travel speeds are substantially restricted and roadway operations are over capacity with extreme delays likely at those intersections. On the other hand, a level of service “C” was found to be what obtained at the Rotary intersections, indicating stable operating conditions with average delays. The Raff and Logit methods of gap acceptance analysis yielded similar results with an average critical gap of 7.4 seconds. The critical gap varies by about 4.23% when compared with the value specified in Highway Capacity Manual (HCM) 2000, The implication is that the driver will have enough gap to maneuver at those intersection. Results of vehicular analysis show that cars, motorcycles and buses constitute 49.82%, 39.87% and 9.37% respectively while 2 and 3 axle load vehicles constitute 0.73% and 0.21% respectively. The implication is that passenger cars and motorcycles are the most frequently used modes. Traffic flow at intersections was expressed as a function of traffic characteristics such as speed, volume, density, headway, gap acceptance parameters, delay and geometric features of the roads. Distance of the next intersections to a major intersection was also found to play a major role in determining flow at that intersection. Shorter distances (less than 300m) lead to increased flow while greater distances (greater than 300m) lead to reduced flow. Models to predict flow at Tee, Cross and Rotary intersections were developed using multiple linear regression technique with the aid of SPSS software. These were validated with empirical data other than those used for model calibration. Peak period R2 values of 0.883, 0.884 and 0.959 for Tee, Cross and Rotary intersections respectively and off-peak period R2 values of 0.939, 0.942 and 0.887 respectively indicate that the flow models are very robust in replicating the observed data. The models have the potential to accurately predict traffic flow at intersections; the findings significantly contribute to knowledge on analysis of traffic flow in a typical African capital city and can be extended to other cities of the world with similar traffic conditions. In order improve the flow of traffic and especially at intersections in Akure metropolis, the following measures are recommended viz-a- viz: provision of traffic signal for effective control mechanisms at the cross intersections, provision of mass transit that will reduce the numbers of car/taxis being used, establishment of continuous counting station to record traffic data and traffic education/ law enforcement