Jihong Cao, PE, Parsons Brinckerhoff Arnab Gupta, PE, Parsons Brinckerhoff Jay Yenerich, PE, Valley Metro

Jihong Cao, PE, Parsons Brinckerhoff Arnab Gupta, PE, Parsons Brinckerhoff Jay Yenerich, PE, Valley Metro

Outline Background Predictive Priority Algorithm Simulation Analysis Measure of Effectives (MOE) Selected Transit Signal Priority Strategy Discussion

Light Rail Transit in Mesa Currently 4 Intersections are part of the 20 mile starter line Main Roosevelt Main Dobson Main Park & Ride Main Sycamore

3.1 mile LRT alignment extends the existing Valley Metro line to downtown Mesa 25+ intersections to be added New LRT Signal System should seamlessly integrate into the existing system

Goals To provide good light rail travel time Minimal stopping delay at intersections. Minimal impact to cross street coordination for major North South arterials Dobson Road Alma School Road Country Club Drive Mesa Drive

Purpose of Study To develop an appropriate Transit Signal Priority strategy and traffic signal timing for Mesa Extension Offline testing the developed timings through traffic simulation Preliminary controller timing will be used as the basis for opening day signal timing.

Predictive Priority Algorithm

Features of LRT Predictive Priority Advanced Predictive Priority logic developed using ASC/3 logic processor. Early Green/Left Swapping Transit Phase Extension Transit Phase Insertion

Phasing and Ring Structure

Predictive Priority Operations LRT Phase LRT Phase Phase 2 & 6 Phase 3 & 7 Phase 4 & 8 Phase 1 & 5

Simulation Analysis

Simulation Analysis Tool VISSIM Micro Simulation Program Transit Pedestrian Signal Controllers ASC/3 software in the Loop package The ASC/3 software in the loop package is able to replicate the exact controller features and the advanced Transit Signal Priority logic embedded in the ASC/3 logic processor.

Model Update Preliminary VISSIM model provided by Valley Metro Updated geometry Updated volume data for opening year, 2016 Added pedestrian traffic volumes Added 8 dedicated signalized pedestrian crossings Replaced previous RBC controller using the developed ASC/3 controller datasets

Model Parameters Vehicular Traffic Speed Dobson Sycamore: 35 50 MPH Sycamore Country Club: 30 45 MPH Country Club Mesa Dr: 20 35 MPH Mesa Dr Stapley: 30 45 MPH LRT 34 36 MPH to follow 35 MPH travel speed 10 minutes headway Station Dwelling Time: 15 seconds Pedestrians Cross walk on the side of adjacent LRT station: 100 peds/hour Other cross walks: 40 peds/hour

Analysis Approach 3 Tiered Analysis Tier I Analysis Evaluation of Alternative Cycle Lengths AM Peak PM Peak 90s, 100s, 110s, 120s, 120 60s,120 90 60s 90s, 100s, 110s, 120s, 120 60s,120 90 60s Two Best Cycle Length Alternatives selected for further evaluation Two Best Cycle Length Alternatives selected for further evaluation Tier II Analysis Offsets Modified to favor LRT progression AM Peak Most Optimal Signal Timing Plan selected for further evaluation PM Peak Most Optimal Signal Timing Plan selected for further evaluation AM Peak Phase 1: Modified logic applied to pedestrian cross walks only Phase 2: Modified logic applied to Pedestrian Cross walks and minor intersections. Phase 3: Modified logic applied to all intersections. Tier III Analysis TSP Improvement PM Peak Phase 1: Modified logic applied to pedestrian cross walks only Phase 2: Modified logic applied to Pedestrian Cross walks and minor intersections. Phase 3: Modified logic applied to all intersections.

ASC/3 Virtual Controller working with VISSIM simulation

Measures of Effectiveness

Travel Time Main Street Vehicular Traffic Travel Time Along Main Street (Dobson-Stapley) AM Peak Main EB Main WB 20.0 16.0 Travel Times (min) 12.0 8.0 4.0 0.0 Base Plus (No LRT) T1: B1 (100 sec) T2: B1 (100 sec) T3: Phase1- B1(100 sec) T3: Phase2- B1(100 sec) T3: Phase3- B1(100 sec) Main EB 13.7 12.7 13.1 16.4 13.2 12.9 Main WB 12.9 13.4 13.5 16.0 13.2 15.1 Alternatives

Travel Time Main Street Vehicular Traffic Travel Time Along Main Street (Dobson-Stapley) PM Peak Main EB Main WB 20.0 16.0 Travel Times (min) 12.0 8.0 4.0 0.0 Base Plus No LRT T1: H2 (120-60 sec) T2: H2 (120-60 sec) T3: Phase1-H2 (120-60 sec) T3: Phase2-H2 (120-60 sec) T3: Phase3-H2 (120-60 sec) Main EB 17.3 14.8 17.4 19.1 16.2 16.5 Main WB 14.5 14.1 14.1 15.7 15.4 15.4 Alternatives

Travel Time LRT LRT Travel Time Along Main Street (Dobson - Robson) AM Peak EB LRT WB LRT 16.0 14.0 12.0 Travel Times (min) 10.0 8.0 6.0 4.0 2.0 - T1: B1 (100 sec) T2: B1 (100 sec) T3: Phase1-B1(100 sec) T3: Phase2-B1(100 sec) T3: Phase3-B1(100 sec) EB LRT 14.8 12.7 11.5 10.2 8.1 WB LRT 13.8 13.1 11.1 9.8 8.3 Alternatives

Travel Time LRT LRT Travel Time Along Main Street (Dobson - Hobson) PM Peak EB LRT (minutes WB LRT (minutes) 16.0 14.0 12.0 Travel Times (min) 10.0 8.0 6.0 4.0 2.0 - T3: Phase1-H2 (120- T3: Phase2-H2 (120- T3: Phase3-H2 (120- T1: H2 (120-60 sec) T2: H2 (120-60 sec) 60 sec) 60 sec) 60 sec) EB LRT (minutes 12.3 11.9 9.5 10.4 8.4 WB LRT (minutes) 13.4 12.7 11.7 11.2 8.3 Alternatives

Travel Time Alma School Rd 6.0 5.0 AM Peak General Traffic Travel Time Along Alma School Rd (University - Broadway) Alma School SB Alma School NB Travel Times (min) 4.0 3.0 2.0 1.0 0.0 T1: B1 T2: B1 T3: Phase1-B1 T3: Phase2-B1 T3: Phase3-B1 Alma School SB 2.4 2.5 2.5 2.4 2.7 Alma School NB 2.5 2.5 2.4 2.3 2.7 Alternatives

Travel Time Alma School Rd PM Peak General Traffic Travel Time Along Alma School Rd (University - Broadway) Alma School SB Alma School NB 6.0 5.0 Travel Times (min) 4.0 3.0 2.0 1.0 0.0 T1: H2 T2: H2 T3: Phase1-H2 T3: Phase2-H2 T3: Phase3-H2 Alma School SB 2.4 2.5 2.5 2.4 2.5 Alma School NB 2.4 2.4 2.5 2.4 2.4 Alternatives

LRT Stop Possibility at Signal AM Peak

LRT Stop Time at Signal AM Peak

LRT Stop Possibility at Signal PM Peak

LRT Stop Time at Signal PM Peak

Time Space Diagram AM Peak Tier 3 (100 seconds Cycle Length)

Time Space Diagram PM Peak Tier 3(120 60 seconds Cycle Length)

VISSIM Simulation Video 2016 AM Peak with Tier III 100 seconds Cycle Length

Selected Signal Priority Strategy AM Cycle length: 100 seconds PM Cycle length: 120 seconds for Major and Minor Intersections 60 seconds for pedestrian crossing intersections LRT phase extra hold for all the intersections

Next Steps Test operation in Signal Shop Implement traffic signal timing in field controllers. Implement TSP and fine tune timings

Summary Advanced Transit Priority Control Logic developed using ASC/3 virtual controllers. The developed control strategies tested through VISSIM simulation. Operational efficiency quantified and evaluated by various Measures of Effectiveness. Core modeling team has extensive experience in modeling and development of transit priority solutions for various BRT and LRT systems.