TRAFFIC SIGNAL CONTROL SYSTEM IN URBAN NETWORK USING MAX-PRESSURE AND SCATS ALGORITHM WITH MACROSCOPIC FUNDAMENTAL DIAGRAM (MFD) EVALUATION

• Main Author: Putri Utami, Fani
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/56949
• Institution: Institut Teknologi Bandung

Nowadays, the need to travel has become an inseparable part of everyday life. It is undeniable that congestion is one of the things that can hinder a trip. In addition, congestion is also a very serious problem because it has an impact on the performance and service of a transportation system in Indonesia. Many factors can cause congestion, including an imbalance in the volume of vehicles with road capacity, infrastructure damage, errors in determining the allocation and range of green lights at each traffic light, and drivers' behaviour in driving their vehicles. The dynamic traffic management system is a solution to reduce congestion that occurs, by improving the performance of a network as a whole. The dynamic traffic management system is a system that allocates the use of infrastructure and vehicle fleets temporally and spatially through dynamic signals. One of the strategies contained in the dynamic traffic management system to improve the performance of a network and reduce congestion is traffic signal control. Traffic signal control is a control system used to regulate vehicles through traffic lights to produce traffic conditions with maximum output, queue lengths that do not accumulate, and less travel time. In this study, the development and simulation of traffic light control on the Bandung city traffic network model were carried out using two control algorithms: the Max-pressure control algorithm and SCATS. Each of these control algorithms has a different working system. In the max-pressure slotted time control algorithm, the green light time allocation is obtained by comparing the results of the accumulated max-pressure value of each phase at each intersection. The phase that has the largest accumulated max-pressure value will be given the green light for a one-time slot with the current time interval and will receive a green light priority. Meanwhile, in the max-pressure cycle time control algorithm, the max-pressure weight value for each phase will determine the length of the green light allocation given in the next cycle. In the scats control algorithm, the green light time allocation is selected based on the minimum degree of saturation (DoS) plan value. The minimum DoS plan value will activate the green light plan. Based on the simulation results on MFD diagrams and travel time graphs, the Max-pressure algorithm with Slotted Time and Cycle Time scenarios is proven to be able to reduce queue lengths and reduce congestion on several roads, especially on Jl. Lembong, Jl. Veterans, and Jl. Tamblong. This can be seen in the density value, the maximum vehicle flow value, and the travel time value generated from the three controllers. In the evaluation of the MFD diagram, the maximum density values generated from the Max-pressure Slotted Time, Max-pressure Cycle Time and SCATS algorithms are 316.07 vehicles/km, 327.56 vehicles/km, and 370.59 vehicles/km. Meanwhile, the maximum vehicle flow values generated from the three algorithms are 3231.90 vehicles/hour, 3058.67 vehicles/hour, and 2981.72 vehicles/hour. From the density values and vehicle flow values, it is evident that the Max-pressure control algorithm is able to provide better performance than the SCATS algorithm. From the results of the evaluation of the travel time graph, the control using the Max-pressure algorithm produces a less travel time value than the control using the SCATS algorithm. It can be seen from the time it takes to travel from Jl. Lembong to Jl. Lengkong Besar in the afternoon is 600 seconds (10 minutes), while with the scats controller the resulting travel time is 1000 seconds (16.7 minutes). This proves that the max-pressure algorithm is more adaptive in dealing with changes in demand, especially in the afternoon or when the traffic peaks in the afternoon.