Integration of Dynamic Traffic Assignment in a Four-Step Model Framework A Deployment Case Study in PSRC Model 13TH TRB National Transportation Planning Applications Conference By: Robert Tung, PhD With: Yi-Chang Chiu, PhD (U of Arizona) Sarah Sun (FHWA) WSDOT PSRC Motives Static trip based macro model is limited in solving modern transportation issues. Activity Based Model (ABM) is promising by may be costly to implement. DTA tools are increasingly sophisticate and efficient in handling large multimodal network. Combination of 4-Step model and DTA is potentially a Low-Hanging Fruit & cost-effective approach to add temporal dynamics to static trip based models. Tung & Chiu : Integration of DTA in a 4-Step Model Framework 2 Objectives Implement a full DTA feedback mechanism in a static 4-step trip based model framework (PSRC) Document the findings and issues learned from the process. Focus on network development, calibration and validation, scenario analysis, and computing resources. Deriving insights from comparing the proposed DTAembedded approach with the existing method. Understand the cost and benefit of integrating DTA in the 4-step process. Tung & Chiu : Integration of DTA in a 4-Step Model Framework 3

Multi-Resolution Modeling (MRM) MACRO MICRO O/D DTA Static/Instantaneous Paths Region Wide Zonal Trips Analytical Equilibrium Demand Driven Planning/Forecasting MESO Dynamic/Time Varying Paths Subarea / Corridor Vehicle Platoons Static Paths Corridor/Intersection Individual Vehicles Simulation One-Shot Supply Driven Operational Simulation Equilibrium Supply Driven Planning/Operational Tung & Chiu : Integration of DTA in a 4-Step Model Framework 4 MRM Issues Macro-Micro Approach: Pros: Widely used in practice. Many tools are available. Cons: Macro demand are not consistent with micro network. No temporal dynamics on demand slices. No feedback.

Macro-Meso-Micro Approach: Pros: Meso demand are more consistent with micro network. Demand reflect temporal dynamics. Cons: Learning curve for planners. Require more computing resource. Mostly auto only. No feedback. Tung & Chiu : Integration of DTA in a 4-Step Model Framework 5 DTA Primer STA DTA MICRO Analytical Meso Sim Micro Sim Shortest Path Instantaneous Time Dependent Instantaneous Route Choice FW/OBA/TAPAS GFV Logit/MSA Connectivity

Link Link/Lane Lane/Turn Resolution Hour Minute Second UE DUE Non-UE Unique Non-Unique Non-Unique Static Average Time Varying Time Varying Flow Model VDF Speed-Density Car Following Arrival Time Profile No

Yes Yes Loading Solution Convergence Speed Tung & Chiu : Integration of DTA in a 4-Step Model Framework 6 DTA Integration in PSRC Land Use Trip Generation Trip Distribution Modal Choice Time of Day Trip Assignment DTA Auto Skims Tung & Chiu : Integration of DTA in a 4-Step Model Framework 7 DTA Integration Concept Land Use Land Use Generation Generation Distribution Distribution

Modal Choice Modal Choice Assignment DTA Tung & Chiu : Integration of DTA in a 4-Step Model Framework 8 Task Outline Network Conversion & Enhancement Intersection Controls Time-of-Day Model and 24-Hour Demand Interface between DTA and TDM 24-Hour Continuous DTA Simulation & Assignment Calibration and Validation Scenario Analysis (HOT, Tolling, Work Zone) Tung & Chiu : Integration of DTA in a 4-Step Model Framework 9 Network Conversion Centroids: From single point to multi-point loading Use arterial links as trip generation and apply loading weights Use standard nodes as trip destination Links/Nodes: Maintain realistic connectivity and GIS shape Nodal orientation is important Controls: Use actuated signals as default if real data are not

available Use reasonable max and min green times Tung & Chiu : Integration of DTA in a 4-Step Model Framework 10 Demand Conversion Use temporal (departure) profile derived from survey or TDM with directionality and peaking characteristics retained Assemble 24-hour demand from time varying period O-D tables Use smaller time interval as possible (15-minute) Separate demand by mode and purpose PSRC 2006 Diurnal Profile 0.0600 PSRC 2006 Auto Demand by Period 0.0500 Period 0.0400 0.0300 0.0200 0.0100 0.0000 HOV Truck 983,292 176,292

104,580 1,264,164 Mid-day 1,721,472 536,416 201,042 2,458,930 PM 1,124,537 382,502 116,784 1,623,823 Evening 888,251 410,576 57,164 1,355,991 Night 490,499 105,715 44,203 640,417

Daily 5,208,051 1,611,501 523,773 7,343,325 AM SOV HOV SOV Tung & Chiu : Integration of DTA in a 4-Step Model Framework Total 11 DynusT Simple , lean and easy to integrate with macro and micro models Developed since 2002, tested (in test) for 20 regions since 2005 Used in several national projects Memory efficient Capable of large-Scale multimodal 24-hr simulation assignment Fast simulation/computation Multi-threaded Realistic microlike mesoscopic traffic simulation Anisotropic Mesoscopic Simulation (AMS) Managed Open Source in 2010/2011 Tung & Chiu : Integration of DTA in a 4-Step Model Framework

12 DynusT Algorithmic Structure Iteration n Traffic Simulation TD O-D Time-dependent OD, network Initial/Intermediate Vehicle Paths Generated Vehicles with Assigned Attributes TD Network Information Strategy Initial Path Anisotropic Mesoscopic Simulation (AMS) AMS Simulation Model MoEs Evacuation Time, Exposure Level, Casualty, etc. TD SP Method of Isochronal Vehicle Assignment n=n+1 Epoch k Traffic Assignment Time-Dependent Shortest-Path Algorithm Assignment Gap Function Vehicle Based Traffic Assignment Algorithm k=k+1 Convergence No

All Epochs Assigned? Assignment Converged? No Yes Stop Tung & Chiu : Integration of DTA in a 4-Step Model Framework 13 Anisotropic Mesoscopic Simulation (AMS) Stimulus-response model Uninterrupted Flow l Net influence for speed adjustment primarily comes from traffic in the front (SIR) Can define different average traffic conditions to model uninterrupted and interrupted flow conditions Speed Influencing Region SIRi left lane Vehicle i right lane Interrupted Flow l Speed Influencing Region SIRi Vehicle i

Tung & Chiu : Integration of DTA in a 4-Step Model Framework right lane 14 AMS q-k-v Curves Modified Greenshields model: Flow Density Curve 2500 2000 Flow (q) 1500 1000 500 0 1 20 40 60 8 0 0 0 2 0 4 0 6 0 8 0 0 0 2 0 4 0 6 0 1 1 1 1 1 2 2 2 2 Density (k) Speed Density Curve 70 60 50 40 30 20 10 0 Speed (v) Speed(v) Speed-Flow Curve 0 500 1000

1500 2000 2500 70 60 50 40 30 20 10 0 1 20 40 60 80 0 0 2 0 4 0 6 0 8 0 0 0 2 0 4 0 6 0 1 1 1 1 1 2 2 2 2 Flow (q) Tung & Chiu : Integration of DTA in a 4-Step Model Framework Density (k) 15 AMS Examples =3.35 Jam Density = 200 Density Breakpoint = 25 Free Flow Speed = 60 Minimum Speed = 6 Speed Intercept=92 70 60 50 40 30 20 10 0 Fre ew ay 1

9 1 7 25 3 3 41 4 9 57 6 5 73 8 1 89 97 0 5 1 3 2 1 2 9 3 7 4 5 5 3 6 1 6 9 7 7 8 5 9 3 0 1 1 1 1 1 1 1 1 1 1 1 1 1 2 Density (k) AMS Flow Density Curves 2000 1600 1200 Flow (q) Speed (v) AMS Speed Density Curves 800 400 0 1 9 1 7 2 5 3 3 41 4 9 57 6 5 7 3 81 89 97 0 5 1 3 2 1 2 9 3 7 4 5 5 3 6 1 6 9 7 7 8 5 9 3 0 1 1 1 1 1 1 1 1 1 1 1 1 1 2 Density (k) Tung & Chiu : Integration of DTA in a 4-Step Model Framework 16 AMS Examples continued Jam Density = 200 Density Breakpoint = 25 Free Flow Speed = 60 Minimum Speed = 6 AMS Speed Density Curves 70 60 AMS Flow Density Curves

40 2500 = 3. 75 30 20 2000 10 1500 0 1 4 7 10 13 16 19 22 Density (k) 25 28 31 34

37 40 Flow (q) Speed (v) 50 = 3. 75 1000 500 0 1 4 7 1 0 13 16 19 22 2 5 28 31 3 4 3 7 40 Density (k) Tung & Chiu : Integration of DTA in a 4-Step Model Framework 17 Compare BPR to AMS 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 0.0 BPR Flow-Density Curve 2.5 2.0

V/C Ratio Travel Time Factor BPR Volume-Delay Curve t = t0[1+0.15(v/c)4] 1.5 1.0 0.5 0.0 0.5 1.0 1.5 2.0 0 2.5 50 70 70 60 60 50 50 40 40 30

20 10 10 0 150 250 2.0 2.5 30 20 100 200 BPR Speed-Flow Curve Speed Speed BPR Speed-Density Curve 50 150 Density V/C Ratio 0 100 200

250 0 0.0 0.5 Density Tung & Chiu : Integration of DTA in a 4-Step Model Framework 1.0 1.5 V/C Ratio 18 BPR Examples BPR Travel Time Curve 25.0 BPR Speed Curves 70 15.0 =0.15 =4.0 =0.72 =7.2 =0.60 =5.8 10.0 50 40 5.0 0.0 60

Speed Travel Time Factor 20.0 0.00.10.20.3 0.40.50.6 0.70.8 0.91.01.11.21.3 1.41.51.6 1.71.8 1.92.02.1 =0.15 =4.0 =0.72 =7.2 =0.60 =5.8 30 20 V/C Ratio 10 0 0.00.10.20.30.40.50.60.70.80.91.01.11.21.31.41.51.61.71.81.92.02.1 V/C Ratio Tung & Chiu : Integration of DTA in a 4-Step Model Framework 19 STA vs. DTA Comparison Simple Network Example BPR: =0.6 =5.8 AMS: =3.35 Tung & Chiu : Integration of DTA in a 4-Step Model Framework 20 STA vs. DTA Comparison Simple Network Example Average Trip Time

by Demand Level 200.0 STA DTA 250x3 2.8 2.2 350x3 3.1 2.4 450x3 4.6 6.6 550x3 8.7 14.0 650x3 18.5 21.7 750x3 38.9 29.1 1,000x3

194.7 47.8 1,500x3 2,017.7 85.0 180.0 160.0 140.0 Avg Trip Time Demand 120.0 100.0 STA DTA 80.0 60.0 40.0 20.0 0.0 250 350 450 550 650 750 1,000 1,500 Demand

Tung & Chiu : Integration of DTA in a 4-Step Model Framework 21 Time Dependent Shortest Path The key feature in DTA Able to produce Experienced travel time and route that is far more realistic than Instantaneous travel time and route produced in STA. Experienced travel time is affected by vehicles departing earlier and later Experienced travel time can only be realized after the trip is completed (Arrival Time Profile) Tung & Chiu : Integration of DTA in a 4-Step Model Framework 22 PSRC Time of Day Model Discrete Logit Choice Model by 30-Minute Interval Aggregated to five periods: AM, MD, PM, EV & NI Uijkpm = ak + c1kDijk + c2kDijkSE + c3kDijkSE2 + c4kDijkSL + c5kDijkSL2 + v + d Where: i = Production zone j = Attraction zone k = Time interval p = Purpose (HBW, HBO, HBShop) m= Mode (SOV, HOV) D = Delays SE = Shift early factor SL = Shift late factor V = Socio-demographic variables

d = Dummy variables 0.3000 0.2500 0.2000 0.1500 A-P P-A 0.1000 0.0500 0.0000 -0.05000 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 :0 1:0 2:0 3:0 4:0 5:0 6:0 7:0 8:0 9:0 0:0 1:0 2:0 3:0 4:0 5:0 6:0 7:0 8:0 9:0 0:0 1:0 2:0 3:0 1 1 1 1 1 1 1 1 1 1 2 2 2 2 Tung & Chiu : Integration of DTA in a 4-Step Model Framework 23 PSRC Time of Day Model Tung & Chiu : Integration of DTA in a 4-Step Model Framework 24 Time of Day Choice Model Pros & Cons Comparing to static TOD model, choice model adds temporal dynamics that enables peak spreading The Shift variables can reasonably spread peak trips over shoulder periods The model is sensitive to changes in delays or generalized costs that is crucial for congestion relief studies Because TOD was calibrated based on base year HH survey and skims data, the model coefficients become questionable for future years of much higher demand and congestion, and resulting TOD

profiles are often unrealistic. Variations of TOD Profiles by Period AM MD PM EV NI Tung & Chiu : Integration of DTA in a 4-Step Model Framework 25 DTA Based TOD Model Baseline Year Model Development: Start from initial departure time profile Delay calculated by DynusT can be fed back by 30 min increment to the TOD model TOD model will adjust the departure time profile Iterative process until convergence Consistency between TOD and DTA is established Future Year Development Considerations: Departure or arrival time profiles based on trip purposes Minimizing total schedule delay + travel time based on trip purposes Decisions applied to future years Tung & Chiu : Integration of DTA in a 4-Step Model Framework Time of Day Model 24-Hour Temporal 24-Hour DTA Time Varying Skims 26 DTA Based TOD Model Average Trip Time by Departure Time 140.00

120.00 SOV_06 TRK_06 HOV_06 SOV_30 TRK_30 HOV_30 Trip Time 100.00 80.00 60.00 40.00 20.00 0.00 1 7 1 0 13 16 19 2 2 2 5 28 31 3 4 37 4 0 43 46 4 9 52 55 5 8 6 1 6 4 67 70 7 3 76 7 9 82 85 88 91 94 9 7 4 15-Minute Interval TOD Shares Departure & Arrival Time Profiles 0.0200 0.0180 0.0160 0.0140 0.0120 0.0100 0.0080 0.0060 0.0040 0.0020 0.0000 Departure_06 Arrival_06 Departure_30 Arrival_30

1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85

89 93 97 15-Minute Interval Tung & Chiu : Integration of DTA in a 4-Step Model Framework 27 Next On-going research project funded by FHWA to investigate the costs and benefits of integrating DTA in a 4-step framework. Results are pending in 2012. Findings of this project will be shared with modeling community. Contact Robert Tung [email protected] for more information. Thank you ! Tung & Chiu : Integration of DTA in a 4-Step Model Framework 28