1. Transportation leadership you can trust. TRB Planning Applications Conference May 18, 2009 Houston, TX A Recommended Approach to Delineating Traffic Analysis Zones in Florida

2. Presentation Overview Purpose and Need Reviewers Recommendation Checklist Delineating TAZs Base-Year Models Future-Year TAZ Considerations Splitting TAZs Topics Requiring Further Research Where to Find the Full Paper Acknowledgements and Discussion

3. Purpose and Need Purpose Provide modelers within Florida guidance on delineating TAZs Applicable both to new travel demand models and refining existing models Need Better represent transit access in Cube Voyager’s Public Transport (PT) program Improved model accuracy for both highway and transit models Consistent approach for Florida’s modeling and planning community

4. Reviewers FDOT Federal Highway Administration (FHWA) Florida Model Task Force (MTF) MTF Model Advancement Committee MTF Data Committee Other interested parties outside Florida

5. Recommendation Checklist Persons per TAZ should be greater than 1,200, but less than 3,000 for all years (base and future) Each TAZ should yield less than 15,000 person trips in all years The size of each TAZ should be between 0.25 and one square mile in area Achieve reasonable number of intrazonal trips in each zone based on mix and density of land use Minimize irregular-shaped TAZs Centroid connectors should load less than 10,000 to 15,000 vehicles per day in all years

6. Recommendation Checklist (Cont’d) Study areas should be large enough so that nearly all (>90%) of trips begin and end within study area The TAZ structure should be compatible with highway and transit networks for all years The centroid connectors should represent true access points to highway network Transit access should be realistically represented TAZ structure should be compatible with Census, physical, political, and planning district boundaries Special generators and freight generators should be isolated within their own TAZ

7. Delineating TAZs in the Base-Year Model Zone size and quantity Boundary compatibility Socioeconomic data Access Centroid connectors Existing transportation facilities TAZ numbering Figure 3.3 Blocks versus Block Groups versus Census Tracts

8. Delineating TAZs in the Base-Year Model Zone Size and Quantity Size of zones depends on intended use of model Larger-sized zones for statewide and systems planning −Statewide models −TAZs can be larger than the arterial grid −Up to approximately 20,000 persons per zone Medium-sized zones for major investment planning −Regional or MPO models −Two or more arterials should not traverse the TAZ in any direction −Up to approximately 4,000 persons per zone Small-sized zones for corridor analysis −Corridor level or subarea models −TAZs in a corridor or subarea should be more refined −Should be greater than 1,200, but less than 3,000 persons per zone

9. Delineating TAZs in the Base-Year Model Boundary Compatibility Physical geography Census geography Political geography Planning district/ sector boundaries Irregular zone geography Figure 3.1 Delineating TAZs Consistent with the Physical Geography Figure 3.2 Delineating TAZs Based on Permanent Barriers

10. Delineating TAZs in the Base-Year Model Socioeconomic Data Special generators Trips per zone Developments of regional impact Figure 3.10 Delineating TAZs Based on Clusters of Similar Land Uses

11. Delineating TAZs in the Base-Year Model Access Transit access Freight/ intermodal facilities Figure 3.11 Over-Estimating Walk Access to Transit

12. Delineating TAZs in the Base-Year Model Centroid Connectors Represent realistic roadway and transit access Should not cross man-made or natural barriers Include sufficient number of centroid connectors to avoid over-loading to one roadway network link Do not connect centroid connectors at intersections or directly to interstate ramps When two centroid connectors are connected to the same roadway segment, access points should be separated by a certain distance Figure 3.20 Placement of Centroid Connectors Relative to Intersections and Interstate Ramps The Wrong Way The Right Way

13. Delineating TAZs in the Base-Year Model Existing Transportation Facilities Rail lines Limited access highways Arterial streets and roadways Collector streets and roadways Figure 3.14 Delineating TAZs Based on Transit Access

14. Delineating TAZs in the Base-Year Model TAZ Numbering Recommended that TAZ numbers begin with one and be consecutive Regional models may set aside a specific range of TAZ numbers for both TAZs and dummy zones for each county Create TAZ equivalency tables to provide list of new zone numbers split from original zone numbers

15. Future Year TAZ Considerations Planned transportation corridors Developments of regional impact Population and trips per zone Figure 4.1 Defining the TAZ Structure Based on Future Corridors Figure 4.2 Defining the TAZ Structure Based on Future DRI

16. Future Year TAZ Considerations Population and Trips per Zone Greater than 15,000 trips per day or 1,200 to 3,000 people per TAZ exceeds recommended zone size The Census Bureau is currently considering increasing the minimum number of persons per block group (and potentially TAZs) to 1,200 persons (previously 600) or 480 households. The existing maximum thresholds of 3,000 persons and 1,200 households per block group remain unchanged Often times, a TAZ may not have as many people or generate that many daily trips in the base-year However, with future-year land use, it may push population or daily trips over the thresholds noted above

17. Splitting TAZs Network data impacts Socioeconomic data impacts Boundary shifts and zone aggregations Zones beyond model boundary Revalidation Figure 5.1 Splitting Household Data into New TAZs

18. Topics Requiring Further Research Mixed land uses Delineating TAZs based on homogenous land uses is not always feasible CBDs and neo-traditional developments make it difficult to isolate employees from households Dynamic subzoning Potential solution to mixed land use issue Includes multiple centroids within one zone; one to represent employment and another to represent households Allows different access points onto the network by having one set of centroid connectors for employment and another for households Would require further research, including the numbering of subzones

19. Where to Find the Paper Full White Paper is on-line at www.fsutmsonline.netwww.fsutmsonline.net

20. Acknowledgements Florida Department of Transportation (FDOT) Vidya Mysore, Systems Planning Office (SPO) Terry Corkery, SPO Cambridge Systematics (CS) Keli Paul, AICP Rob Schiffer, AICP AECOM Consult Dave Schmitt

21. Discussion Questions or Comments? Keli Paul, AICP Cambridge Systematics Senior Associate Tallahassee, FL 32301 850.219.6388 kpaul@camsys.com Terry Corkery FDOT, Systems Planning Office Senior Transportation Modeler Tallahassee, FL 32399 850.414.4900 Terrence.Corkery@dot.state.fl.us