The three fundamental parameters of a traffic signal system are split, cycle and offset.  This paper describes the use of high-resolution data to identify time periods when split parameters that might be too short and assist the engineer with identifying opportunities to reallocate split time. A case study of 7 corridors with 47 intersections is presented. A drill-down approach was developed to find movements where there is opportunity to reallocate split times to improve performance. A heuristic that can reallocate up to 5 seconds of under-utilized green time on a competing phases is applied to these corridors. For the selected phase identified in this study, the adjustment was shown to reduce split failures by an average of 40% while also decreasing yellow occupancy by an average of 40% and red light violations by an average of 66%.  The paper concludes by recommending central systems to implement drill-down split failure dashboards, similar to those shown in this paper to facilitate sustainable practices of monitoring split failures and easily identifying opportunities to reallocate green time when there is slack capacity in other phases.

This paper presents a Distributed-Coordinated methodology for signal timing optimization in connected urban street networks. The underlying assumption is that all vehicles and intersections are connected and intersections can share information with each other. The novelty of the work arises from reformulating the signal timing optimization problem from a central architecture, where all signal timing parameters are optimized in one mathematical program, to a decentralized approach, where a mathematical program controls the timing of only a single intersection. As a result of this distribution, the complexity of the problem is significantly reduced thus, the proposed approach is real-time and scalable. Furthermore, distributed mathematical programs continuously coordinate with each other to avoid finding locally optimal solutions and to move towards global optimality. We proposed a real-time and scalable solution technique to solve the problem and applied it to several case study networks under various demand patterns. The algorithm controlled queue length and maximized intersection throughput (between 1% to 13% increase compared to the actuated coordinated signals optimized in VISTRO) and reduced travel time (between 18% to 34% decrease compared to actuated coordinated signals) in all cases.

Making inappropriate driving decisions at signalized intersections is one of the major reasons causing accidents. In this paper, we present an integrated in-vehicle decision support system to help making better stop/go decisions as the vehicle is approaching an intersection. The system integrates and utilizes the information from both the vehicle and intersection, supported by the Vehicle-to-Infrastructure (V2I) communications in the era of the Internet of things (IoT). Its effective decision support models (DSM) are realized with the probabilistic sequential decision making process (PS-DMP) which combines a variety of advantages gained from a set of decision rules, where each decision rule is responsible to specific situations and does not require complete information for right decisions. We extract decision rules from the existing models of the indecision zone problem. We also design new decision rules to utilize the key inputs from vehicle motion, vehicle-driver characteristics, signal timings, intersection geometry and topology, and the definitions of red light running (RLR). The performance of the decision support system is empirically evaluated with simulation experiments. The results show that the system enhances both the safety and the mobility of the vehicles approaching an intersection. The increase in safety and efficiency can be achieved by changing the inputs of the system in a large variable space, not only from the benefits of reducing the perception-reaction time, increasing the green countdown time, extending the yellow interval, and extending the all-red interval, but also from a seamless integration with the RLR definition and specific driver behavior.

The effects of realistic wireless communication are important for the modeling and evaluation of emerging connected and automated vehicle applications. This paper presents the development of a novel closed-loop Connected Vehicle Analysis System (CONVAS) interlinking Vissim microscopic traffic simulation and ns-3 wireless communication simulation. In CONVAS, ns-3 utilizes Vissim’s vehicle positions for wireless modeling and in turn Vissim utilizes the packet reception from ns-3 to modify vehicle behaviors. We propose an intelligent dilemma zone avoidance (IDZA) application to demonstrate the use of the platform for modeling connected and automated vehicles and illustrate the impacts of communications on the application performance. A simulation experiment is conducted using an isolated high-speed fixed-time signalized intersection test bed with different communication model configurations. The evaluation results show that the effectiveness of the application as measured by successful clearing of vehicles out of the dilemma zone depends greatly on wireless communication performance.

Traffic congestion is becoming an increasingly serious issue in urban road networks across the United States and around the world. Over many solutions to traffic congestion, operational improvement through traffic signal optimization is preferred because a properly designed and fine-tuned signal timing plan can significantly improve travel delay of motorists, progression of traffic flow, vehicle emissions, fuel consumption and operating costs. In this paper, a traffic signal coordination algorithm, named Network-wide Traffic Signal Coordination Algorithm (NeTSCA), is developed aiming to improve the operational performance of traffic network. The developed algorithm consists of two optimization levels: corridor level and network level. At the corridor level, NeTSCA maximizes the bandwidth along each corridor, taking unbalanced directional traffic flow, residual queue and start-up delay into account. At the network level, NeTSCA optimizes total network bandwidth under the major intersection constraints, while taking corridor importance level into account. Offset fine-tuning functionality is designed into NeTSCA as a post-processing optimization step to deal with vehicle speed variation and other complex practical situations. Micro-simulation is utilized to evaluate the effectiveness of NeTSCA on improving network operational performance in comparison with SYNCHRO and a baseline condition under various test scenarios. Statistical analyses on performance measures obtained from VISSIM micro-simulation platform validate that NeTSCA can effectively improve network-wide operational performance with significantly decreased average delay, increased average speed and reduced average number of stops under most test scenarios. 

This study proposes an integrated driving control strategy by taking advantage of the automated driving technology at the individual vehicle level and the traffic signal preemption strategy at the traffic infrastructure level. This aims to to facilitate an automated driving-based emergency vehicle control and ultimately to achieve efficient and safe control of emergency vehicles. To this end, this study developed the integrated emergency vehicle control logic, implemented the logic in the microscopic traffic simulation environment using the simulation software’s application programming interface capability, and evaluated the impacts of the proposed emergency vehicle control logic in the aspects of mobility and safety with different driving aggressiveness and preemption initiation settings. The results show that the proposed emergency vehicle control logic achieved benefits on mobility and safety within the normal driving concept that does not allow special right-of-way rules while it did not produce significant side-impact to adjacent vehicles. In addition, conservative or intolerant lane changes appeared no benefit especially in terms of safety, and these lane change schemes were even worse than the Base scenario, which does not consider both the signal preemption and the automated driving control. The study results emphasize that the benefit of emergency vehicle control strategy can be maximized when the signal preemption and the automated driving control operate in collaboration.

In this paper, a new methodology is proposed to establish practical guidelines for four incremental left turn treatments: (1) permissive single left turn lane, (2) protected/permissive single left turn lane, (3) protected dual left turn lanes with equal lane utilization, and (4) protected dual left turn lanes with unequal lane utilization.  Decision boundaries were derived based on equilibrium at which the delays of two incremental treatments are equal.  The benefits and costs, such as safety impact, construction and maintenance costs, associated with different left turn treatments were also considered, which effectually shift the boundary curves for more realistic decision making.

Traffic signal cycle lengths are traditionally optimized to minimize vehicle delay at intersections using the Webster formulation. This study continues previous work to develop new formulations to compute the optimum cycle length, considering measures of effectiveness such as vehicle fuel consumption and tailpipe emissions. Additionally, it validates the Webster model against simulated data. The microscopic simulation software, INTEGRATION, was used to simulate two-phase and four-phase isolated intersections over a range of cycle lengths, traffic demand levels, and signal timing lost times. Intersection delay, fuel consumption levels, and emissions of hydrocarbon (HC), carbon monoxide (CO), oxides of nitrogen (NO_x), and carbon dioxide (CO₂) were derived from the simulation software. The cycle lengths that minimized the various measures of effectiveness were then used to develop the proposed formulations. The first effort entailed enhancing the Webster model by incorporating an additional intercept to the new formulations. The second effort entailed recalibrating the Webster model to the simulated data to develop an updated delay formulation. The proposed model produced better traffic signal timings and was calibrated against four study intersections to minimize delay, fuel consumption, and emission levels. The proposed model considers fuel consumption and greenhouse gas emissions, and its estimated cycle lengths were longer than cycle lengths optimized for vehicle delay only. Secondly, the simulation results had shorter cycle lengths for high vehicle demands when optimized for delay compared with the Webster and proposed formulations.

This paper proposes a connected vehicle–based approach to improve the throughput at signalized intersections and ultimately increase the mobility of transportation systems. Connected vehicle technology demonstrates tremendous potential for improving safety and mobility, as it enables the real-time sharing of vehicle data, including position, speed, acceleration, etc., not only among vehicles but also between vehicles and infrastructure. The proposed approach takes advantage of such real-time data to develop a strategy that maximizes throughput of a signalized intersection locally. Accordingly, the problem is formulated as a two-step centralized responsive control. There are two main processes in this method: responsive control for vehicles in motion, and responsive control for stopped vehicles. The first step maximizes the intersection throughput of vehicles in motion using advisory acceleration. The second one minimizes the total delay of stopped vehicles by adjusting the positions at which vehicles stop. Simulations are presented to demonstrate the efficiency of the proposed approach, which improves the traffic flow throughput of a signalized intersection and reduces the total delay of all vehicles. The simulation results show that we can increase the throughput by the responsive control of the vehicle in motion and stopped vehicles in most traffic conditions.

Despite the increasing applications of Superstreets on congested arterials, a reliable method to design its signal plan remains a research issue.  To account for the unique geometric features of a Superstreet, this study presents a two-stage signal optimization method, where its first stage determines the optimal common cycle length and the second stage optimizes the offset for each sub-intersection. Using the Mixed-Integer-Linear-Programming (MILP) technique for model formulations, the first stage has the objective of maximizing the total traffic throughput while the second stage aims to maximize the bandwidths and minimize the delay experienced by minor road drivers. To assess the proposed model’s effectiveness, the study has further conducted a simulation evaluation with the field data from a Superstreet in Maryland. The results of extensive experiments reveal that the proposed model can effectively offer the desirable signal progressions to both heavy through and left-turn movements, and prevent the potential queue spillback among a Superstreet’s four closely-spaced sub-intersections.

Signal Progression has been proven as an effective way to improve the operational efficiency of traffic signals at local arterial corridors. Conventional two-way progression models have shown their promising in providing green bands to both inbound and outbound through traffic along the arterial. However, they may not be capable of offering an effective progression plan when a long arterial contains a large number of intersections. Under such condition, it is critical to firstly divide the arterial corridor into a set of subgroups for progression design while a reliable tool for such a need is not available yet. To tackle this issue, this study develops a decomposition model that can concurrently determine the arterial decomposition strategy and optimize the resulting signal progression plan within each subgroup. With an integrated control objective function, the proposed model is able to minimize the required number of subgroups but still satisfy the operational need (i.e., a minimal bandwidth is required). In addition, the proposed model is formulated with a mixed-integer-linear-programming (MILP) technique that can guarantee a global optimal solution. Numerical example on a field arterial corridor which consists of fifteen signalized intersections proved the effectiveness of the proposed model.

This paper presents a discrete-time optimal control model to dynamically adjust signal green times for single-destination networks. The dynamic traffic signal re-timing strategy is designed to minimize the negative impact of signal timing and lead to a reduced total travel time during the transition period. The underlying day-to-day flow dynamics in the proposed optimal control model applies flow splitting rates at nodes to avoid route enumeration and preserve the flow conservation in the absence of route flow variables. The proposed discrete-time nonlinear programming problem can be solved efficiently by optimization software. A numerical example is used to investigate the traffic network performance when the proposed day-to-day dynamic signal timing strategy is deployed. The numerical results demonstrate that the proposed day-to-day dynamic signal timing strategy can reduce the total travel time under network disequilibrium, and emphasize the importance of understanding travelers’ route choice in the design of traffic signal re-timing plans.    

This paper develops a perimeter control strategy for an urban road network affected by special traffic events (STEs) to maximize network throughput. The proposed control strategy features its effectiveness from two aspects: (i) The control target is considered as a variable determined based on the Macroscopic Fundamental Diagram (MFD) that is affected by the STEs. A data-driven approach is proposed to dynamically update the control target. (ii) The proposed control strategy applies the proportional–integral–derivative (PID) regulator, which is easy to implement according to the adaptively updated control targets. A simulation study investigates the effectiveness of the proposed control strategy in improving the network throughput when STEs occur. The simulation results also demonstrate the effectiveness of the proposed control strategy in terms of reducing the adverse impacts of STEs and minimizing the total delay.

This study analyzed the theoretical dilemma zone (DZ) and option zone (OZ) under three different scenarios: normal weather (S-1), foggy weather (S-2), and foggy weather with in-vehicle Advance Yellow Warning (AYW) system (S-3), and it investigated the impact of AYW message on the length and location of DZ and OZ. The data was collected from a driving simulator test of participants with different ages and gender. Results indicated that, with an in-vehicle AYW system under foggy weather conditions, the intersection safety was improved in terms of minimizing the DZ effect through placing the drivers in the OZ, and the DZ started far from intersection depending on the approaching speed compared to the normal and foggy weather conditions. Analysis of the differences between the vehicles distance to intersection and two critical distances (i.e. stopping and passing) at yellow onset indicated that in-vehicle AYW system increased the intersection safety in terms of minimizing the unsafe maneuvering distances. Two variables were introduced for the DZ driving behavioral analysis: short of minimum stopping zone ( X _sms) and cut of possible passing zone ( X_cpp ). By applying a binary logistic regression, it was concluded that at any speed under foggy weather condition, the probability of falling into these unsafe zones were higher compared to the other two scenarios.

With growing demand for travel, most Indian cities face problems of congestion, acute delay and pollution. There are different approaches adopted by transportation engineers and planners to address such problems. One approach is to efficiently co-ordinate the signals in a traffic network using automated methods. Such automated methods make use of knowledge of traffic's progression from an upstream signal to a downstream signal. The signal directs the movement of vehicles in groups during the release of traffic at green light that are called vehicle platoons. A very simple and accepted notion for signal coordination is that the green light should start when the platoon from the previous signal reaches the signal so that average delay is minimized. For efficient signal coordination, the effect of platoon dispersion should be taken into consideration. Among the different platoon dispersion models, the Robertson's model is widely used in automated signal co-ordination. In this paper, Robertson's model was modified to account for the heterogeneity observed in Indian traffic conditions, and an optimization approach was taken to estimate model parameters in a methodical and time-efficient way. The results from the proposed models were compared with the traditional Robertson's model that was calibrated for Indian conditions. The results showed that the performance of proposed models was almost on par with the calibrated Robertson's model. Also, the proposed models would be more apt for real time application given the traditional trial and error method of estimating the parameters that is more time consuming.

Arterial traffic signal coordination makes traffic flow more effective and safer. This paper presents a partition-enabled multi-mode band (PM-BAND) model that is designed to solve the signal coordination problem for arterial with multiple modes, i.e., passenger cars and transit vehicles. The proposed method permits the progression bands to be broken if necessary, and optimize system partition and signal timings coordination in one unified framework. The impacts of traffic demand of passenger cars and transit vehicles as well as the geometry characteristics of the arterial are taken into account. PM-BAND model is formulated as a mixed-integer linear program and can be solved by the standard branch-and-bound technique. Numerical examples results have shown the proposed method can reduce average stops and delay, moreover, the progression bands generated by the PM-BAND model have a higher reliability and effectiveness.

The performance of signalized intersections affects the performance and productivity of the entire road network. Hence, design for optimal signal control strategy is always a traffic engineer’s concern. Delay is a crucial parameter in developing this optimal strategy which is a function of saturation flow. Further, saturation flow is subjected to stochastic variability due to the random composition of vehicles in heterogeneous or in other words mixed traffic conditions that generally prevails in cities of developing countries like, India. This paper proposes an analytical methodology which takes into account the stochastic variability of saturation flow in delay determination and thereafter in the optimization process which is done using a standard nonlinear optimization solver: LINGO. Results show that the delay values calculated based on the proposed probabilistically estimated saturation flow are more accurate and closer to the field observed values. When this delay model is used in the optimization process it yields green timings for the intersection which reduces the intersection delay by 29.71%, the evaluation of which is done in microscopic simulation environment using VISSIM.

Recently, connected vehicle (CV) technology has gained great attentions with active efforts of pilot deployment, considering its potential of improving the effectiveness and efficiency of the transportation system. At traffic intersections, CVs pertain particular potential to serve as virtual and mobile detectors, providing opportunities of creating detector-free signal operation. So far, researchers have proposed innovative signal control with CVs through extending actuated signal or proposing new adaptive signal control strategies. However, the efforts mainly focus on scenarios in which penetration rates of CVs achieve certain levels, e.g. 25%, which are not feasible in the near future. It remains untapped regarding using CVs in real world to improve existing traffic signal operation. To fill in this gap, this work develops an innovative approach of using CV data with low penetration rates (<10%) to estimate traffic volumes, key inputs to traffic signal operation. Essentially, we model traffic arrivals at signal intersections as time dependent Poisson process accounting for impacts of signal coordination. Based on the Poisson process, an expectation maximization (EM) procedure is derived to estimate volumes using signal status data and CV trajectory data from the roadside equipment (RSE). For validation, a case study is conducted using real-world CV data from the safety pilot model deployment (SPMD) project, with good accuracy found for the estimation. Considering existing deployment scales of CV pilot systems, the proposed approach could help paving the way of utilizing CVs for improving traffic signal operation in real world.

Several research studies have been conducted on the effectiveness of Controller Actuated Beacons (CABs) near isolated high-speed signalized intersections, indicating that CABs are generally effective in reducing accidents{Formatting Citation}. However, while drivers approaching the intersection during flashing CABs might anecdotally accelerate or decelerate depending on their time/distance to the stop bar, the literature lacks guidance on quantifying this behavior and accounting for it in the use of DZ protection systems in the presence of CABs. This paper focuses on analyzing and identifying key variables that can result in a change in a vehicle trajectory as it approaches intersections with flashing CABs, and develop guidelines for the use of DZ-protection systems in these cases. The findings of this work include identifications of conditions when drivers accelerate and get trapped in DZ due to the CAB operation and provide guidance on placement of CABs and change in DZ-protection boundaries to avoid these situations.

This paper presents a mixed integer linear programming (MILP) model to optimize vehicle trajectories and traffic signals in a unified framework for isolated urban intersections in a connected and automated vehicle (CAV) environment. A rolling horizon strategy is applied to conduct the optimization. All vehicle movement directions such as left-turning, right turning and through movements are considered. Phase sequence, green start and duration of each phase, and cycle lengths are optimized over the time horizon together with vehicle arrival time, vehicle platooning, and lane changing behaviors with the objective to minimize vehicle delays. The trajectories of platoon leading vehicles are then optimized based on the optimal arrival time for the minimization of fuel consumption and emissions. The trajectories of following vehicles are modelled by Newell’s car-following model, which makes the proposed model extendable to mixed traffic conditions where not all vehicles are CAVs to increase the intersection capacity. Simulation results show that the proposed model outperforms actuated signal control, especially in terms of delay and emission reduction. Sensitivity analysis suggests that length of minimum green time has a significant impact on the performance of the model.

To improve the safety of people walking at particular signalized intersections, traffic signal engineers may implement leading pedestrian intervals (LPI) to provide pedestrians with a walk signal for a few seconds prior to the parallel vehicular green indication. Previous research using before-after studies and simple economic analyses shows that LPIs are low cost tools that can reduce vehicle-pedestrian conflicts and crashes at some signalized intersections. Despite this evidence, there is little guidance for municipalities on when to implement LPIs. This paper develops a marginal costs and benefits framework using quantitative metrics, extending the concept of traffic conflicts and marginal safety-delay tradeoffs to analyze the appropriateness of implementing an LPI at specific signalized intersections. The guidance provided by this method helps quantify the probability of a conflict happening, and provides direction on whether or not to implement an LPI at a given location based upon macroscopic level inputs, including turning movement counts, crash data, and geometry. A case study with sample data indicates that an LPI is cost effective for the scenario presented. 

ABSTRACT

At coordinated signals, a signal may goes to transition due to oversized pedestrian timing, transit signal priority (TSP) or emergency vehicles. At these situations, signal tries to bring back disrupted coordination to its original plan by a transition option. There are several options for transition method, including Dwell, Max Dwell, Add, Subtract, and Short way. To have a quantitative tool to analyze different transition options, mathematical models of transition methods were developed. The mathematical models of transition options were implemented into software called TOD (Transition Options’ Delays), which is an open source program available for anyone who is interested in using it. With this software, practitioners simply input the common timing parameters such as cycle length, vehicle and pedestrian volume, and required pedestrian timing. The software will then give a recommendation based on the overall delay. The mathematical models and the related TOD software are expected to provide much improved recommendations when dealing with pedestrian crossings at coordinated signal systems.

Keywords: Pedestrian timing accommodation; Transit signal priority, Emergency vehicle, signal coordination; transition methods; TOD

The flashing yellow arrow (FYA) signal display creates an opportunity to enhance the left-turn phase with a variable mode that can be changed on demand.  The previously developed decision support system (DSS) in phase I facilitated the selection of the FYA left-turn mode, changing by time of day at intersections.  There was a need to continue to refine the interactive framework to improve its service.  However, the ultimate objective of the continued research of phase II was to demonstrate the ability to execute the automation of the process.  Phase II of the FYA project provided additional intersection data that refined the model. Virtual testing of the DSS was first conducted using VISSIM application programming interface (API) before the field testing environment. A Custom communications software was developed to retrieve instantaneous channel input data, synchronize opposing thru green phase, analyze traffic information, provide the algorithm decision, and generate a real-time log recording the events to determine whether it would be optimal to switch the red arrow to a flashing yellow arrow. The algorithm determines the time interval between the successive arrivals of vehicles and computes the corresponding headway for each lane by cycle on a second-by-second basis. The DSS was ultimately tested at two different intersections in Seminole County. The FYA 4-section configuration provides the opportunity for a fully adjustable system and provides the TMCs with more tools to operate the intersections as efficiently as possible at peak and off-peak times.

Saturation flow is one of the most important parameters to assess the efficiency of the urban road network. In order to bring a heterogeneous traffic stream into a homogeneous one consisting of passenger car, the concept of Passenger Car Unit (PCU) factor was introduced. Historically, several methods have been developed for the estimation of PCU values. Among all, the headway and regression-based methods are the most commonly used techniques for PCU estimations at signalized intersections. The present study computed PCU values for different vehicle categories present at signalized intersections under heterogeneous traffic condition using these two methods. While headway method involves complex equations along with the difficulty related to obtaining vehicle type based headway combinations, regression method shows unrealistic PCU values for certain vehicle categories at few approaches. Consequently, the study introduced the concept of Stream Equivalency Factor (SEF) to avoid the profound understanding of the PCU estimation. Its equation has been developed using data from six different study approaches. Later, data from five other approaches are used to validate the developed model. It was observed that the estimation of saturation flow in PCU/hr using SEF does not vary much from the saturation flow achieved through the actual use of PCU values. While the SEF values decreased with the increase of two-wheeler proportion, the proportion of bus increased the SEF value. The proposed SEF can be used to estimate the saturation flow at signalized intersections in terms of PCU/hr, without going through the complicacy of PCU estimations.