Evaluation of Group-based Signal Control through Field Operational Tests

At signalized intersections, green times should be assigned to satisfy the traffic demand for all approaches. Stage-based signal control is the most common and basic control policy worldwide particularly at Japanese signalized intersections. It is defined as a signal control policy in which compatible traffic movements are grouped to move together in a specific time span within a signal cycle, which are referred to stages, and green times are then assigned to each stage depending on the movement with the highest demand in each stage. At intersections with relatively balanced traffic demand in opposite approaches, the stage-based signal control can assign green time reasonably for each stage. However, at intersections with imbalanced traffic demand in opposite approaches, a portion of the assigned green time to the approaches with minor traffic demand will not be used since stage-based signal control assigns equal green time to major and minor traffic demand approaches. Since this control policy has a stable phasing plan and the green time is assigned for an approach not for movements, long cycle length and large delay are likely to occur at such intersections. In order to overcome such deficiencies, group-based signal control was invented and widely used in different countries. It is defined as signal control which directly assigns green times to traffic movements without the need to maintain a specific stage structure. This policy can assign the green time to each traffic movement rather than a stage of different movements, therefore the unused green time which is assigned by stage based signal control to the approaches with minor traffic demand can be avoided. Such control policy provides flexible phasing structure where different non-conflicting movements can be joined together in one phase. This mechanism might reduce the cycle length by saving the unused green time, and furthermore delay as well. Furthermore, the flexible phasing structure is advantageous in avoiding several conflicts between traffic movements which might have positive impact on the safety phases, the flexible phasing patterns which might confuse drivers at the beginning and the end of green phases. This possible negative impact might be the main reason behind the limited application of this policy in Japan. In this study, the operational and safety performance before and after the implementation of group-based signal control at two test intersections are evaluated and compared.