Traffic Signals in Motorcycle Dependent Cities

Traffic signal control was first invented in 1868 in Great Britain. Then, it has quickly spread across many other countries, what are now usually developed countries. Therefore, during a long history of development, traffic signals usually have been dealing with traffic in which four-wheel vehicles play an important role in motorised traffic. However, as a result of an unequal development among countries, many Motorcycle Dependent Cities (MDCs) nowadays still exist mostly in the developing countries such as: Hanoi, Ho Chi Minh City (Vietnam), New Delhi (India), Taipei (Taiwan), Bangkok (Thailand), etc. In these cities, motorcycles play an important role in motorised traffic. Because motorised traffic dominated by motorcycles was born later than car traffic, the complete application of traffic signals from developed countries, which usually deals with four-wheel vehicles, to these specific traffic conditions of two-wheel vehicles is usually ineffective. In practice, it has been causing many traffic problems at traffic signals. To solve problems at traffic signals in MDCs, this study analysed the applicability of the German standard “Guidelines for Traffic Signals” (RiLSA, edition 2009) to establish a draft of Guidelines for Traffic Signals in MDCs. In order to achieve the goals and objectives, some contents of RiLSA needed to be modified, in which the minor modifications are directly written in the draft of Guidelines for Traffic Signals in MDCs. This study, therefore, focuses on the major modifications of the intersection layout engineering design, the signal program design, and the traffic signal control strategies. However, the four criteria: (i) traffic safety, (ii) traffic flow quality, (iii) environmental impacts, and (iv) economics are always considered when modifying any content of RiLSA, in which the first two criteria play an important role in MDCs. And overall, the traffic regulation at traffic signals has to obey the following priority in order: (i) pedestrians, (ii) cyclists, (iii) public transport, and (iv) motorised traffic. For the intersection layout engineering design, it is necessary to take not only motorised traffic but also public transport, pedestrian traffic, and cycle traffic into account as a whole. However, the layouts for cycle traffic and pedestrian traffic can be applied from RiLSA. Therefore, some modifications on the intersection layouts are implemented for motorised traffic and public transport, in which the major aim is firstly to give priority to public transport, and secondly to give priority to motorcycles in motorised traffic. For the signal program elements, unlike some available researches that consider only the equivalent factor converting motorcycles into passenger car units, this study has dealt with a series of the signal program elements such as cycle time, green time, amber time, intergreen time, etc., in which a new concept of saturation flow was used. This concept results from the homogeneous motorcycle traffic saturation flow and the homogeneous car traffic saturation flow, and of course this saturation flow depends on the motorcycle traffic volume as well as car traffic volume on the approach. It means that the saturation flow is not fixed as the homogeneous car or motorcycle saturation flow, but it varies depending on the proportion of motorcycles and cars in the traffic flow. From this new concept of the saturation flow, the formulas calculating the cycle time and the green time were formed. For the last two signal program elements, the amber time and the intergreen time were calculated based on the German method in which the positions of the stop-lines, the speeds as well as the deceleration rate of vehicles are taken into account. For the traffic signal control strategies, according to RiLSA, there are three macroscopic control levels and three microscopic control levels, in which the microscopic control levels are activated from the macroscopic control levels. However, in MDCs, counting the number of vehicles requires too much effort or it seems to be impossible with conventional technologies. Therefore, only the macroscopic control level “time-dependent signal program selection (A1)” can be applied to MDCs. From this macroscopic control level, two microscopic control levels (fixed-time signal program (B1), and signal program adaption including green time adjustment (B2), phase swapping (B3), demand phase (B4), and time-offset adjustment (B5)) are activated in MDCs because the fixed-time signal program is always easily implemented, and the signal program adaption does not require efforts in counting the number of vehicles. Instead, it only requires detecting vehicles, and this is possible for two-wheel vehicles. The last microscopic control level, the signal program formation (B6), should not be applied to MDCs because it requires much effort in collecting traffic data online. After having the results for some modifications, a draft of new “Guidelines for Traffic Signal in MDCs” was compiled. It includes six chapters: (0) Introduction, (1) Basic Principles, (2) Signal Program Design, (3) Interrelationships between Traffic Signal Control and Road Engineering Design, (4) Control Strategies, (5) Technical Design. Besides, it also has three annexes: Annex 1: Details on the Traffic Load, Annex 2: Traffic Flow Quality, and Annex 3: Traffic Engineering Calculation. After testing this study at some signalised intersections in Hanoi and Ho Chi Minh City, these guidelines will be considered to be a foundation for establishing a Vietnamese Standard for Traffic Signals. The necessary “Formal Right Agreement“ between the German Forschungsgesellschaft fur Strasen- und Verkehrswesen FGSV (Road and Transport Research Association) and the Vietnamese Ministry of Transport is in the approval process. Finally, also other MDCs, might consider applying this standard.