Intelligent Traffic Signal System for Isolated Intersections Dynamic Pedestrian Accommodation

Transportation Research Record: Journal of the Transportation Research Board, No. 2259, Transportation Research Board of the National Academies, Washington, D.C., 2011, pp. 96–111. DOI: 10.3141/2259-09 plex in nature because their scopes have been rapidly expanding far beyond the traditional realm. “A large number of variables are involved; the parametric relationships among them are not well-understood; a large volume of incomplete data is involved, and the goals and constraints are many and the priorities among the stakeholders are unclear” (1, p. 455). Philosophically, “as the complexity of a system increases, our ability to make precise and yet significant statements about its behaviors diminishes, and significance and complexity become almost mutually exclusive characteristics,” which implies that a multitude of transportation problems are profoundly difficult to resolve in traditional approaches (2, p. 363). Artificial intelligence methodologies are proved instrumental to modeling the behaviors of complex phenomena while giving researchers the latitude to “acknowledge some level of ignorance,” which “provides the opportunity to examine the problem from different perspectives and compare the results” (1, p. 455). There has been vastly growing interest in applying artificial intelligence and relevant advanced computing techniques to address the complex issues essentially associated with safety, operations, and other dimensions of multimodal transportation infrastructure systems (3). Perhaps one of the most critical issues that challenge traffic engineers is to comprehensively optimize the performance of urban signalized intersections where motorized and nonmotorized travelers are busily transported in dynamic operations. The signalization improvement has been recognized as one of the most cost-effective ways of mitigating roadway congestion and ameliorating multimodal transportation safety (4). In traffic engineering, intersection control pursues dual principal objectives: “(a) ensure safety for all intersection users; (b) promote efficient movement of all users through the intersection” (5). Alternatively, the dominant considerations in the operation of an isolated intersection are safe and efficient traffic movement, limiting vehicle delay, and increasing intersection capacity (6). To achieve those goals is a difficult task because generally safety and efficiency are mutually competing (or even conflicting), rather than reinforcing or complementary, issues (5). Fuzzy logic control (FLC) has been proved more effective, flexible, and robust than traditional controls in tackling complex systems in which conflicting objectives, subjective perception, imprecise data, and vague decision-making criteria play critical roles (1, 7). Therefore, the variability and complexity in intersection signalization can be effectively modeled in an FLC-based approach for manifold improvement purposes. U.S. safety data reveal that a pedestrian is killed or injured in a traffic accident every 120 or 8 min, respectively (8). Intersection crosswalks are particularly hazardous to seniors and children. In 2008, 35% of pedestrian fatalities among people 60 and older occurred at intersections, compared with 20% for those younger Intelligent Traffic Signal System for Isolated Intersections Dynamic Pedestrian Accommodation