APJITM UKM

Article Info

Controlling Traffic for Clean Air and Healthy Cities with Multi-Fuzzy Inference Systems for a Sustainable Future

Muhammad Aria Rajasa Pohan, Jana Utama
dx.doi.org/10.17576/apjitm-2025-1401-18

Page 324 - 338

Abstract

Surging urbanization and vehicle emissions exacerbate traffic congestion in major cities, leading to increased air pollution and high gasoline emissions as traditional traffic signal systems fail to adapt to real-time traffic conditions. To address this critical issue, this research proposes an innovative traffic control strategy using multi-fuzzy inference systems (mFIS) distributed at adjacent intersections in areas experiencing high traffic volumes. These distributed traffic light control systems collaborate to reduce congestion gradually in areas of high congestion. The mFIS system optimizes the traffic light settings based on the length of the vehicle queue, the duration of the vehicle stopping time, the number of vehicles in front of the lane, and the number of vehicles entering the lane. Our simulations demonstrated that the proposed mFIS controller effectively reduced average vehicle delay by up to 23.6% compared to the Proportional-Integral-Derivative (PCT) controller, 19.9% compared to the Variable Speed (VA) controller, and 14.1% compared to the traditional Fuzzy Logic System (FIS) controller. This significant performance improvement was consistently observed across various traffic conditions, including heavy traffic scenarios. The advantages of the proposed algorithm lie in restricting vehicles entering high-congestion lanes, speeding up the outflow of vehicles from congested areas, and using distributed control principles within areas experiencing high traffic volumes. Therefore, the algorithm in this study has the potential to be further developed to help reduce traffic congestion in big cities; thus, it can be a sustainable solution to create cleaner, healthier, and more sustainable cities in the future.