Designing a fuzzy logic controller for a single intersection: A case study in Gaziantep

Management of traffic flow is one of the most significant issues for globalizing world. In city centers, traffic is generally regulated via intersections using traffic lights. Therefore, optimized traffic light durations alleviates the traffic congestion significantly. "Queue length during red light" and "remaining vehicles in line after green light" are two important parameters to determine the green light duration. In this study, a single intersection of Gaziantep is taken into account with above mentioned two inputs of fuzzy logic controller considering each approach one by one. The output variables are chosen as "phase selection" and "extend of green light duration". Unlike other studies, these output variables are combined into one system based on modal distinction. Proposed controller consists of two separated model with morning and evening peak hours based on different defuzzification methods. The model is solved using MATLAB fuzzy inference system. Delay parameter is considered as performance measurement. This real application model can be applied in similar intersections if the necessary data are collected. Finally, proposed model is compared with traditional Webster model, a mathematical model formed by the regression equation and an existing mathematical model solved by GAMS (The General Algebraic Modeling System). Results show that the fuzzy controller can effectively minimize total delay and is superior to compared methods.