integrated macro-micro highway demand/operational analysis case study: cross bronx expressway...

Integrated Macro-Micro Highway Demand/Operational Analysis Case Study: Cross Bronx Expressway Corridor, Bronx, NY

Presented at the 15th TRB Transportation Planning Applications Conference

Authors: Bernard Alpern, Paul Balmacund, AECOM

Presentation Overview

• Introduction/Purpose of Analysis• Approach• Macro-level Modeling• Micro-level Modeling• Results

The Cross Bronx Expressway is a Highly Congested Highway

Inrix

Cross Bronx Expressway in a Regional Context

LONG ISLAND

NEW ENGLAND

NEW JERSEY, WEST & SOUTH

Introduction/Purpose of Analysis

• Planning-level Study performed for NYS DOT

Cross Bronx Expressway:

- CBE only has Partial, Discontinuous Service Roads

- Lack of Alternative, Efficient E-W Routes for Local Travel

• Concept: - Create a continuous E-W “Connector Road” alongside the CBE

- Reconnect N-S Streets that were made discontinuous at time of the CBE construction

- Improve Local Traffic Circulation

- Involves the Construction of New Streets and the Use of Existing Streets

• Estimate Travel Demand and Traffic Operational Feasibility, and Benefits

Limits and Configuration of Connector Road in CBE Corridor

Configuration Details of Segment 1 of CBE Connector Road

University Ave

Hugh GrantCircle

GrandConcourse

University Ave

Modeling Approach

• Use a combination of “macro-level” static assignment model and microsimulation

• Macro-level model used to project growth, trip re-routing, and as the source for “seed” trip matrices to the microsimulation model

• Microsimulation model used to test operational feasibility of proposed Connector Road, and to identify locations for further analysis

• Both macro and microsimulation models were calibrated to measured traffic conditions

Modeling Challenges for the CBE Corridor

• Extremely-Congested and Constrained Corridor during lengthy AM and PM peak-periods

• Several Major and extremely Complex Interchanges are located along the Corridor, with very substandard geometric features

• Very aggressive driving behavior in the Corridor

• Microsimulation must correctly reflect impacts of these factors on capacity

Macro-Level Static Assignment Model

• Subarea network and Auto/Truck trip-tablesextraction from NYMTC MPO Best Practices Regional Model

• Modeled Periods: 6-10 a.m., 4-8 p.m. Weekdays

• Successfully calibrated to Y2009 CBE Mainline and

Ramp peak-period volumes

Macro-Level Static Assignment Model Network

NORTH

Microsimulation Network

Henry Hudson Parkway

George Washington Bridge

Harlem River Drive

I-87Major Deegan Expwy

I-895SheridanExpwy

Bronx RiverParkway

NORTH

HamiltonBridge

WashingtonBridge

Microsimulation Network

Henry Hudson Parkway Harlem

River Drive

I-87Major Deegan Expwy

Bronx RiverParkway

I-895SheridanExpwy

NORTH

Microsimulation Model

• CBE, Connector Road, Street intersections

• Modeled periods: 6-10 a.m., 4-8 p.m. weekdays

• Successfully calibrated to 2009 volumes, travel-times, queues, following FHWA/WISDOT guidelines.

Microsimulation Model – Travel Time CalibrationAM Peak

PM Peak

Obs 5-Run Average Difference Difference %EB Section 1 TME Inner Road 2:23 2:43 0:20 14%

Section 2 TME Outer Road 1:24 1:35 0:11 13%Section 3 CBE 8:37 8:01 -0:36 -7%EB Time 12:24 12:18 -0:06 -1%

WB Section 3 CBE 12:36 12:10 -0:26 -4%Section 2 TME Outer Road 1:21 1:35 0:14 17%Section 1 TME Inner Road 1:21 1:47 0:26 33%WB Time 15:18 15:32 0:14 1%

Obs 5-Run Average Difference Difference %EB Section 1 TME Inner Road 1:34 3:32 1:58 125%

Section 2 TME Outer Road 1:29 1:33 0:04 5%Section 3 CBE 16:31 14:14 -2:17 -14%EB Time 19:34 19:16 -0:18 -2%

WB Section 3 CBE 22:55 23:26 0:26 2%Section 2 TME Outer Road 1:53 1:21 -0:32 -28%Section 1 TME Inner Road 1:24 1:14 -0:10 -12%WB Time 26:12 25:59 -0:13 -1%

Traffic Growth to Year 2035

• Traffic growth for the Macro-level model was derived from the regional model

• Traffic Growth for the Microsimulation model was derived from the Macro-level model

• Due to extreme peak-period capacity constraints on the CBE, the growth in CBE traffic derived from the regional model needed to be moderated.

• A uniform growth rate (for each peak period) based on both projected CBE and local travel growth was applied to all origin zones to create the AM and PM peak-period trip tables for assignment to the Microsimulation network

Results – CBE Mainline

Vol. Change % Change Vol. Change % ChangeEastbound SegmentUniversity Ave to Jerome Ave -680 -2.8% -1,090 -5.0%Jerome Ave to Clay Ave -530 -2.3% -890 -4.5%Clay Ave to Arthur Ave -270 -1.3% -510 -3.0%Arthur Ave to Southern Blvd -270 -1.2% -770 -3.4%Southern Blvd to Bronx River Ave -280 -1.4% -770 -3.8%Westbound SegmentBronx River Ave to Southern Blvd -190 -1.2% -170 -1.3%Southern Blvd to Arthur Ave -280 -1.3% -530 -3.0% Arthur Ave to Clay Ave -150 -0.9% -310 -2.2%Clay Ave to Jerome Ave -150 -0.7% -410 -2.4%Jerome Ave to University Ave -120 -0.5% -240 -1.3%

AM (6-10)Peak-Period Volume Change from No-Build

PM (4-8)

Projected FY 2035 Peak-Period, Cross Bronx Expressway Mainline Traffic Volumes

Results – Connector Road

AM PM(6 - 10) (4 - 8)

Eastbound SegmentUniversity Ave to Jerome Ave 2,190 2,310Jerome Ave to Clay Ave 1,360 2,020Clay Ave to Arthur Ave 990 3,290Arthur Ave to Southern Blvd 940 3,080Southern Blvd to Bronx River Ave 1,160 3,270Westbound SegmentBronx River Ave to Southern Blvd 820 1,120Southern Blvd to Arthur Ave 110 1,480 Arthur Ave to Clay Ave 1,030 1,560Clay Ave to Jerome Ave 910 1,140Jerome Ave to University Ave 1,260 2,870

Projected FY 2035 Peak-Period, Connector Road Traffic VolumesPeak-Period Volume

Recap

• Connector Road improves east-west mobility

• Connector Road provides for improved traffic circulation and connectivity

• A number of locations were identified for further analysis in the microsimulation to assure that the Connector Road will work operationally

• Successful calibration – “if you can model this corridor, you can model any corridor”