Optimizing Traffic Flow Using Fuzzy Logic-Based Control Systems

Modern cities are becoming more and more urbanized, which has increased traffic congestion and made traditional fixed-time traffic signal control systems less effective. These older technologies are not real-time adaptable to changing vehicle density and flow rate since they are based on static timing formulas like Webster’s equation and historical flow patterns. This study presents a control system based on fuzzy logic of the Mamdani type that is intended for dynamic signal optimization at isolated four-way junctions. Traffic Density (D) and Flow Rate (F) are the two main inputs that the system uses real-time sensor feedback to record. By applying 25 expert-defined rules based on these inputs, the fuzzy inference engine addresses the non-linearity and uncertainty present in traffic systems by enabling the system to produce adaptive Signal Cycle Length (C) outputs without the need for pre-programmed thresholds. In contrast to traditional fixed-time systems, the fuzzy controller facilitates rule-based decision-making by modeling ambiguous real-world scenarios using language terms (“Low,” “Medium,” and “High”) for both input and output variables. In comparison to conventional Webster-based systems, the simulation findings show an average cycle length reduction of 17.6%, reduced vehicle idle time, and enhanced traffic throughput. Fuzzy logic provides a lightweight, real-time solution for situations where prompt, understandable, and computationally efficient conclusions are crucial, whereas traditional systems only use mathematical averages and contemporary AI-based deep learning techniques frequently call for sizable labeled datasets. This study demonstrates fuzzy logic’s potential as a bridge technology, providing computational clarity in contrast to black-box AI systems and flexibility beyond conventional models.