An Evaluation of Light Rail Transit Signal Control Options

As part of the Central Phoenix/East Valley Light Rail Transit Project, the Cities of Phoenix, Tempe and Mesa, Arizona are proposing a new light rail line, part of which will follow Central Avenue, a major north-south corridor in the City of Phoenix. The service is expected to initially operate with approximately 6-minute headways in both directions. The LRT tracks will be in the center median and will require the incorporation of either train preemption or priority into the signal control system. This paper reviews the evaluation of three distinct LRT signal control strategies based on a microscopic simulation analysis of future traffic and transit operations through the Central Avenue corridor from Earll Drive to McDowell Road. The individual simulation models include a detailed emulation (in one case an actual interface) of the LRT priority or preemption logic of current off-the-shelf signal control equipment. The three analyzed strategies include: • NEMA TS/2 with Railroad Preemption • NEMA TS/2 with Transit Priority (Early Green and Green Extension) • Type 2070 Predictive Priority The three LRT signal control strategies are evaluated based on measures of effectiveness such as LRT travel time, general purpose traffic travel time, average intersection delay and queue lengths at critical approaches such as left-turns across the LRT tracks. This detailed operational comparison using a variety of LRT and general purpose measures of effectiveness provides a good reference for traffic engineers faced with the task of selecting an appropriate LRT signal control strategy. METHODOLOGY AND APPROACH OVERALL PROJECT WORK PLAN This simulation analysis is a task of the overall work plan, which includes the following tasks: 1. Conceptual Engineering Phase • SYNCHRO analysis of LRT intersection crossings on alignment, intersections within station areas, and intersections on parallel streets affected by diversion of traffic. • Qualitative analysis of impacts. 2. Preliminary Engineering Phase • SYNCHRO and CORSIM analysis of intersection geometric requirements. • CORSIM analysis of coordination versus LRT priority control of signals. • VISSIM analysis of priority control options for demonstration segment (subject of this paper). 3. Final Design Phase • VISSIM analysis of intersections on alignment and on intersecting and parallel arterials. • Signal timings and geometric refinements. DESCRIPTION OF SIMULATION TOOL – VISSIM VISSIM is a microscopic simulation model developed to model urban traffic and public transit (including railroad) operations. The program can analyze traffic and transit operations under constraints such as lane configuration, traffic composition, traffic signals, transit stops, etc., thus making it a useful tool for the evaluation of various alternatives based on transportation engineering and planning measures of effectiveness. VISSIM’s traffic flow model is a discrete, stochastic, time step based microscopic model, with driver-vehicle-units (DVU) as single entities. The model contains a psychophysical car following model for longitudinal vehicle movement and a rule-based algorithm for lateral movements (lane changing). The model is based on the continuing work of Wiedemann (1974, 1991, 1999) at the University of Karlsruhe, Germany. Vehicles follow each other in an oscillating process. A faster vehicle approaching a slower moving vehicle on a single lane has to decelerate. The action point of conscious reaction depends on the speed difference, distance and driver-dependent behavior. On multi-lane links moved-up vehicles check whether they improve their operation by changing lanes. If so, they check the possibility of finding acceptable gaps on neighboring lanes. Car following and lane changing together form the traffic flow model, being the kernel of VISSIM. The simulation system itself includes first, the traffic flow model and, second, the signal control model (see Figure 1). The traffic flow model is the master program, which sends second-by-second detector values to the signal control program. The signal control uses the detector values to decide on the current signal state. Signal control itself can be performed by the programmable signal control software VAP (Vehicle Actuated Phasing), an external controller (e.g., interface to NEMA TS-2 or Type 170 controller or actual Type 2070 controller software (“virtual controller”). VS-PLUS, actual Type 2070 controller software, was used for this simulation analysis. Figure 1. VISSIM Model Architecture The basic element of the modeled network is a single or multilane link. The network is composed of links and connectors. A connector can be placed at any position on a link. Connectors are valid for all vehicles, certain types (i.e., buses) or a set of vehicles (i.e., only right turning vehicles). Cross section markers are used to model routing decision points for O/D modeling (large network) or turning movement percentage (single intersection). Signal control is modeled by placing the signal heads at the positions of the stop lines. Detectors collect traffic flow data for the signal control (i.e., presence, gap, and occupancy) and for microscopic and macroscopic measurements (i.e., speeds, volumes, and travel times). Semi-compatible movements are modeled with a gap acceptance model. A transit route is defined in VISSIM as a sequence of stops along routes. Transit routes can operate either on exclusive right-of-way or in mixed-flow lanes. Transit stops are either on the link or adjacent to it in case of a bus pullout. Transit vehicles enter the network according to their scheduled arriving time. Modeling of random “lateness” is accomplished by modeling transit stops (dummy stations) at the entrance to the network. VISSIM Urban Traffic Simulator Microscopic traffic and transit network simulation Delay, travel time, queues, emissions, signal performance, etc. Fixed-time, VAP, external controller, virtual controller Signal controller Detector values Signal status per phase VISSIM generates arrival time and route (destination) for every vehicle arriving at the entry points of the network. The arrival profile is entered as hourly values for the PM peak period. Within one time interval VISSIM assumes a Poisson arrival distribution. MODEL ASSUMPTIONS AND CHARACTERISTICS