Statewide Modeling and Corridor Management CE 451/551 d reference: http://onlinepubs.trb.org/onlinepubs/circulars/ec011/ec
Jan 17, 2016
Statewide Modeling and Corridor Management
CE 451/551
A good reference: http://onlinepubs.trb.org/onlinepubs/circulars/ec011/ec011.pdf
Presentation Outline Statewide Models Managing Access in
Commuting Corridors
What is a Statewide Model?
Very similar to a Metropolitan Area Travel Demand Model. Wider in scope!
Many use the same sequential 4-Step Process Trip Generation Mode Split Trip Distribution Traffic Assignment
Truck Model Freight Model Intercity Bus Air and Barge
Why Model Statewide? We are seeing the benefits modeling brings
to urban areas. Technical basis behind big $ projects. In the rural area there is no such technical
tool to support decision making. Currently use old fashioned hand-drawn
methods to forecast traffic.
Why Model Statewide? Traffic forecasts affect every aspect of a State
DOT’s core business activities. Planning, finance, programming, design, construction
and maintenance. A useful, cost-effective tool
System, corridor, project level analysis for evaluating needs and alternatives
Consistent methodology Better than current DOT method. Two forecasters-two
forecasts.
Why Model Statewide? Policy questions State Gas Tax Increase Project Prioritization Statewide VMT by geographic area to determine
where dollars should be spent State Plan (like MPO Plan) Bypass Studies Corridor Studies Rural detour studies
Why Model Statewide? One major construction project costs
millions of dollars. Shouldn’t it be based on the best available data?
Project Example US 218 – Charles City, Iowa Area. Avenue of the Saints.
St. Louis, Missouri and St. Paul, Minnesota Systems Planning did a traffic forecast for
pavement design. Old fashioned trend line growth analysis. We totally missed on the Truck percentages. Route experienced significant increases in truck
volumes.
Maps from www.iowadotmaps.com
Project Example What happened?
Asphalt pavement mix substandard for those truck weights.
Result: Significant rutting. Too many dynamics taking place.
Statewide Model likely benefit. Could have saved a lot of $.
Truck trends
0%
10%
20%
30%
40%
50%
60%
70%
Urban- IS Urban-NHS
Rural- IS Rural -NHS
1998
2020
Percentage of Segments with Over 10,000 Trucks, Comparison of 1998 to 2020.Percentage of Segments with Over 10,000 Trucks, Comparison of 1998 to 2020.
National Trend
Opportunity for Modelers Large Scale Plans require Huge Investments Need Computer Models to Help Decisions Statewide Model new in Iowa - MI had one in
the 1960s see: http://onlinepubs.trb.org/onlinepubs/circulars/EC011/NELLET.PDF
Built from Urban Knowledge Base Severely Lacking Intercity Data Lacking in Data Overall
Survey Data
Data Collection is Key Data or Theory? Shortcomings in Both Behavioral Surveys are needed Origin Destination Surveys are needed Focused to Specific Corridors Economic Interaction Surveys Commodity Flow Surveys
Other States Louisiana Maine Virginia Texas Michigan Ohio Wisconsin Florida California
New Hampshire North Carolina South Carolina Oregon Mississippi Vermont Missouri Kentucky Now Iowa
Statewide Modeling Capabilities Passenger Car Models Truck Models Freight Models Rail Models Intercity Bus Air Barge Intermodal Connections Time of Day Trip Purpose
Ohio Example
Nationwide Coverage
Coded Area Type
Virginia Example
County Level Zones
Virginia Example
County Level Zones to Larger Areas
Virginia Example Network
Why is this road needed?
Kentucky Zone Example
Kentucky Zone Example
Iowa’s Model will simulate this structure.
BEABusiness Economic Areas
Trip Generation Data Population Employment CTPP National Survey Data Reebie Data (now Global Insight):
Transsearch database
Trip Purposes Home Based Work Home Based Non-Work Urban Non-Home Based (2,500 +) Rural Non-Home Based Tourist Truck and External Trips
Special Generators Intermodal Centers Tourist Attractions Etc.
Trip Distribution Even though there are issues the Gravity
Model is the common choice Issue – Rural vs Urban Trip Lengths
Trip Distances are Different Why we have both urban and rural trip
purposes
Trip Assignment All or Nothing Capacity Restraint – Traditional Pre-Assignment of Trucks
Ensures trucks stay on through routes Don’t use Equilibrium Assignment
Calibration vs Validation No Widely Accepted Standards Lenient FHWA standards apply. +/- 4,000 ADT
Typical Cost and Time Depends on Multimodal Detail Depends on Data Available Are Travel Surveys Needed? Range to Expect $300,000 - $2 million Or $6 million for Activity Based like Ohio Typically, a multi-year effort
Process to Identify High Priority Corridors for Access Management Near Large Urban Areas in Iowa
David Plazak and Reg SouleyretteCenter for Transportation Research and Education
Iowa State University
Iowa DOT Access Priority Ratings (English Conversion)
Source: Iowa DOT.
Rating Description
1 Access points at interchanges only
2 Access points spaced at minimum 2625 ft
3 Access points spaced at minimum 984 ft rural, 656 ft urban
4 Access points spaced at minimum 656 ft rural, 328 ft urban
5 Iowa DOT has minimum access rights acquired
6 Iowa DOT has no access rights acquired
Research Project Goals
Address current and future access management problems on state highway routes located just outside urban areas that serve as major routes for commuting
There were two basic goals for the project: Develop a ranking system for identifying high-priority segments
for access management treatments on primary highways outside metro and urban areas.
Focus efforts on routes that are major commuting routes at present and in the future
An example commuting corridor: US 6 to the west of the Des Moines metropolitan area (on next slide)
US 6, In The Waukee Area
Technologies Used Arc View GIS was used to integrate various
Iowa DOT databases, including roadway characteristics, traffic, and crash records
A 2940 zone traffic model was developed to estimate and forecast commuting activity on all Primary routes
Known model weakness: border metro areas
Iowa Forecast Population Growth By County
Source: Woods and Poole Economics
Traffic Model Zone Structure
Traffic Analysis Zones (TAZs) were developed from US Census Block Groups.
ArcView was used to find the center of each TAZ and then create a centroid at this location
The centroids were then connected to the network at breaks in the existing network
To reduce the number of segments and nodes between intersections, ArcInfo was used to combine multiple segments into a single link
ArcView was then used to create node and link files formatted for use in TRANPLAN
Base-year Estimated Traffic Model Commuting Trip Volume
5-Year Estimated Traffic Model Commuting Trip Volume
Forecast Absolute Change In Commuting, 1999-2004
Forecast Percentage Change In Commuting, 1999-2004
Access-Related Crashes Included In The Analysis
Collision Type Description
4 Rear-end/right-turn collision
5 Rear-end/left-turn collision
12 Broadside/right-angle collision
13 Broadside/right-entering collision
14 Broadside/left-entering collision
Ranking Factors Used Frequency Rate Loss/Severity Percentage
Access Related
Access-Related Crash Losses on Commuter Routes by Iowa DOT District
District
Crash Loss
Access Crash Loss
Access Related Ratio
District to State Ratio
1$19.3B
$4.1B21.45% 54.52%
5$9.1B $1.7B
18.51% 22.41%
6$5.8B $830M
14.23% 10.93%
2$1.8B
$490M26.26% 6.43%
4$1.3B $350M
25.66% 4.65%
3$750M $80M
10.63% 1.05%
Total$38.2B $7.0B
20.00% 100.00%
Red circles indicate pilot projects
“Pro-Active Corridors” Some corridors may not have access-
related crash problems today, but could have in the future
“Pro-Active Corridors” were identified based on the following factors: Forecast commuting traffic growth Proximity to metro and large urban areas Access priority ranking of 3, 4, 5, 6, or none Driveway access density
Most Access Management Problems Happen Incrementally
Poorly managed corridors don’t happen overnight; they happen over many years
They often happen one decision at a time A series of decisions is usually involved in
degrading a corridor What can one more median opening hurt?
One more commercial driveway? One more traffic signal?
The problem is that all the small, bad decisions cumulate into one large problem—a “hairball”, to use computer programming slang
For another example of a specialty statewide model, see the second half of last week’s freight lecture.