Session: 09-08-01: Distance/Online Engineering Education, Models and Enabling Technologies

Paper Number: 70956

Start Time: Tuesday, 10:25 AM

70956 - Application of Adaptive Neuro-Fuzzy Inference System Model on Traffic Flow of Vehicles at a Signalized Road Intersections 

The increase in private car ownership and the population explosion in cities is making road traffic congestion severe and time-consuming to road users. Traffic congestions are part of human beings' everyday activities, activities including going to work and visiting shopping malls. Even nowadays pedestrians have to stop before crossing the road to adhere to the rule of traffic signal light and zebra crossings because of heavy vehicular traffic flow from both directions of the road. International airports are now getting congested such that aircraft are put into a kind of holding pattern before other airplanes are allowed to land at the terminal. Even the time it will take a passenger to queue at the airport has increased tremendously at an alarming rate. There is an urgent need to develop an intelligent approach for traffic congestion at a signalized road intersection. Several transportation researchers have suggested that autonomous vehicles are the future of transportation, but the world is still years behind from achieving a fully autonomous vehicle. Traffic congestion has become a widely acknowledged very difficult to solve societal problems that have taken their toll on many major urban cities of the world. The primary objectives of transportation researchers and government administrators nowadays are to eliminate the occurrence of traffic congestion and assist urban planners with solutions on how to tackle traffic congestion problems. Although various efforts and incentives have been implemented to reduce traffic congestion in megacities, new traffic congestion problems kept re-occurring and with a high level of unpredictability, especially in developing countries even though some pre-existing traffic problems have been ameliorated. Sometimes, it is difficult to comprehend if the traffic congestion measures that have been put in place by urban planners and transportation engineers will either work or not. Over the last decades, the evolutionary improvement in road transportation has been viewed as one of the primary reasons for the increase in the urban population because of the tremendous boom witnessed in factories and industries. The continuous migration from rural to metropolitan cities has led to the availability of more vehicles on the road, causing severe traffic bottlenecks. This paper aims to develop an algorithm based on signalized traffic flow system to address the constant repetitive traffic congestion problem in South Africa. The proposed algorithm is the adaptive neuro-fuzzy inference system (ANFIS) that can be found in the Matlab environment. The speed of vehicles within the investigation period, the time is taken for vehicles to navigate at the traffic section, and the distance the vehicles covered before reaching the road intersection were used as input and output variables. Membership function for input and output variables were defined, and rules equally developed based on available traffic flow parameters and dataset obtained from the South Africa transportation network. The ANFIS model was applied. The result showed a great improvement in the system, capable of putting an end to traffic congestion at signalized road intersections.

Presenting Author: O. I. Olayode University of Johannesburg

Authors:

O. I. Olayode University of Johannesburg
L. K. Tartibu University of Johannesburg
M. O. Okwu University of Johannesburg