trip distribution and assignment modeling methods
TRANSCRIPT
Traffic Impact AnalysisModeling Methods
• Why– Most manual distribution and
assignment techniques include numerous subjective inputs
– Models offer an MPO-adopted tool to aid in distributing and assigning traffic
Modeling Methods• FSUTMS–Florida Standard Urban
Transportation Model Structure– Florida’s standard model used in all 26
MPOs
– FSUTMS uses Cube Voyager software engine
• Effective for site impact analysis when combined with manual techniques– At a minimum, FSUTMS trip generation
outputs must be verified with manual analysis (ITE)
Travel Demand Modeling• Four-Step Process
– Other auxiliary modules include network building and calculating zone-to-zone travel times.
Trip Generation
Trip Assignment
Mode Split
Trip Distribution
SFDU MFDU Hotel Employment1 xxx xxx xxx xxx xxx xxx 2 xxx xxx xxx xxx xxx xxx 3 xxx xxx xxx xxx xxx xxx 4 xxx xxx xxx xxx xxx xxx
ZONEDATA
Special Generators2 xxx xx xxx xx3 xxx xx xxx xx
SPECGEN
Generation
TAZ PRODS ATTRS 1 xxxx xxxx 2 xxxx xxxx 3 xxxx xxxx 4 xxxx xxxx
FSUTMS Trip Generation
Trip Generation• Inputs:
– Number of homes and employees (ZONEDATA file) in each traffic analysis zone (TAZ)
• Outputs:– Number of Production trip ends (generated by
homes) and Attraction trip ends (generated by employment sites) for each zone
Note: The model adjusts the number of A’s (usually downward) until number of P’s and A’s in the entire MPO are equal.
Distribution
PRODS
ATTRS
Person-Trip Table T O______ 1 2 3 4 1 xx xx xx xx 2 xx xx xx xx 3 xx xx xx xx 4 xx xx xx xx
F R
O M
__
FSUTMS Trip Distribution
Trip Distribution• Inputs:
– P’s and A’s calculated in trip generation
• Outputs:– The number of person-trips traveling between each
zone pair in the MPO area
• Production trip ends are matched up with attraction trip ends from throughout the network to form person-trips, using a gravity model.
Person-Trip
Table
Mode Split
Vehicle-Trip Table T O______ 1 2 3 4 1 xx xx xx xx 2 xx xx xx xx 3 xx xx xx xx 4 xx xx xx xx
F R
O M
__
FSUTMS Mode Split
FSUTMS Mode Split• Inputs:
– Person-trip table from trip distribution
• Outputs:– Vehicle-trip table
• Vehicle occupancy factors or transit usage equations are applied to convert person-trips to vehicle-trips– Vehicle occupancies differ by trip purpose (e.g., work
trips generally have fewer passengers than recreational trips).
Vehicle-Trip
Table
Assignments
xxx
xxx
xxx
xxx
xxx
xxx
xxx
xxx
xxxxxx
xxx
xxx xxx
xxx
xxx
xxx xxx
xxx
xxx
xxx
xxx
xxxxxx
xxx
xxxxxx
xxx
xxx
xxx
xxxxxx
xxx
xxx
xxx
xxx
xxx
xxx
xxxxxx
Network Link VolumesFrom ToNode Node Volume1001 1002 xxxx1001 1003 xxxx1002 1004 xxxx
FSUTMS Highway Assignment
Highway Assignment• Inputs:
– Vehicle-trip table
• Outputs:– Number of trips upon each roadway link
• Vehicles are routed along specific roadways to arrive at the zone they desire to reach.
FSUTMS vs. ITE Trip GenWhich is better for Site Impact Analysis?
• Intended Purpose– FSUTMS
• Designed to replicate MPO areawide travel patterns
• Calibration = Avg Link Volume-To-Count comparisons
• NOT calibrated to zone-specific traffic volumes
– ITE• Designed to find correlation between land
use quantities and number of trips entering and exiting specific types of developments
FSUTMS vs. ITE Trip GenWhich is better for Site Impact Analysis?
• Application– FSUTMS
• Does not calculate “Trips” for employment sites. Rather, the model calculates “Attractions,” which are only measures of relative attractiveness to draw Productions to the zone.
– ITE
• Calculates the average number of trips generated by varying quantities of specified land uses.
So what is FSUTMS good for?
• Key strengths:– Trip distribution and assignment
– Non-development trips
• Therefore…– The best modeling methodology for traffic
impact analysis:
• Uses the gravity model and capacity-restrained assignment
• But, calculates the number of development trips consistent with ITE
Replicating ITE Results in FSUTMS(For the proposed development’s TAZs)
• Two alternative methods:
– Special Generator method• uses model’s SPECGEN input file
• requires running the model and adjusting several times
– Link Distribution Percentages method • uses conventional ZONEDATA input file
• single model run indicates percentage of total external development traffic traveling on each link
Special Generator Method• Uses SPECGEN input file
• Requires several assumptions to factor the development’s ITE-generated trips:– Vehicle-trips to person-trips
(using vehicle occupancy factors)
– Percent trips by purpose (work, shopping, social-recreational, non-home-based, etc)
– Percent productions versus attractions
Disney Logo / Arena in Miami
Special Generator Method• Requires trial-and-error attempts to match
FSUTMS to ITE– Check final trip table after FSUTMS trip attraction
balancing and vehicle occupancy factoring
– Compare to ITE calculations
– Adjust SPECGEN inputs and rerun model until traffic loadings to/from development zones match development’s ITE external trip generation
Link Distribution Percentages Method
• Insert dwelling unit and employment estimates into conventional ZONEDATA input file– Requires converting square feet to
industrial, commercial, and service employment
• Execute full FSUTMS model run
Employees per 1,000 sq ft–Typical Rates
Office: 4 Industrial: 2
Hospital: 3 Warehousing: 1
Retail <200KSF: 3 Hotel: 1 per room
Large-scale retail: 2 Source: ITE, ULI
Link Distribution Percentages Method
• For each roadway segment, calculate development traffic percentage:
Development traffic on link .
Total external generation of site (modeled)= Link %
Apply ITE total external generation to link percentages
Development trips for each roadway link =Link % x Total External ITE Generation
External ITE = 15,000
15,000
15,000x 30% 4,500
15,000x 70%10,500
3,000 7,000
10,000
FSUTMS Output(Development Trips)
30% 70%
DistributionFactors
(manually calculated)
ITE-GeneratedDevelopment Trip Loadings
Link Distribution Percentages Method
Link Distribution Percentages Method Benefits
• Easier application
• Generally similar results as SPECGEN method
• If development’s land use quantities change slightly, no need to rerun model– CAUTION: Significant land use changes
may affect distribution patterns, requiring a new model run
FSUTMS Highway Assignment• Based on a capacity-restrained,
equilibrium assignment routine• Capacity restraint accounts for route diversion
that occurs when drivers encounter congestion
– FSUTMS decreases speeds on congested roadways after each iteration of the assignment, until:
• Equilibrium is achieved when all trips in the MPO area have found the least congested, shortest-time path to their destination
FSUTMS Site Impact AnalysisHIGHWAY ASSIGNMENT
• Uses the Selected Zone analysis tool– Single assignment tracks total trips and
development trips throughout the capacity-restrained roadway network
– Background traffic = Total minus development trips
Selected Zone Analysis Procedures
• Input proposed development’s land use into zonal data
• Run FSUTMS• Display traffic that enters/exits development
zone(s) on the loaded network using the traffic assignment path file
• Save development traffic as a new link attribute for further analysis
• Check for reasonableness
Using MOCFModel Output Conversion Factor
• Theory behind use– Land use inputs reflect peak-season
occupancy rates
– Base-year traffic counts used during validation represent peak-season volumes
– Therefore, traffic volumes calculated by the model reflect peak-season weekday average daily traffic (PSWADT).
MOCF• AADT needed to calculate peak hour traffic volume:
– AADT x K100 = 100th Highest Hourly Volume
• MOCF converts PSWADT to AADT (always reduces model output volumes)
• MOCF source: Florida Traffic Information disc Peak Season Factor reports
• CAUTION: Use MOCF for appropriate part of county (e.g., tourist area vs. rural); do not reduce development trips
Internal Trips & FSUTMSQ . Is FSUTMS the best method for
estimating the number of internal trips?
A . FSUTMS is only a tool that may help in a manual determination of internal trips.
Caution: Size (land area) of TAZs and length of centroid connectors are the prime determinants of intrazonal trips in FSUTMS. (Longer centroid link = more intrazonal trips.) The model does not consider whether the specific land use types in the proposed development are compatible and thus, conducive to internal trip making.
Selected Zone Versus “With & Without”
• Selected Zone Analysis– Single model run with total trips and development
trips
– Development trips are tracked within the total trip assignment paths
– Retains the capacity-restraint trip diversion that occurs in the assignment of total trips
Show with Show without
• “With & Without” Methodology– Two separate model runs, one with proposed
development in place, the other with development’s zonal data zeroed out
– Link volumes for “without” run subtracted from “with” run, yielding net impact of development on each roadway link
Selected Zone Versus “With & Without”
• “With & Without” methodology presents a problem:– Equilibrium highway assignment capacity-restraint
equation diverts trips, often resulting in virtually no change in traffic volumes
– Developer: “So, what! Diversion occurs in the real world. I should only be required to mitigate for net impacts of the development.”
Selected Zone Versus “With & Without”
“With & Without” MethodFSUTMS Total Trips(With Development)
50,000 45,000
10,000
FSUTMS Total Trips(Without Development)
48,000 48,000
0
“Net” Impact???
50,000 45,000- 48,000 -48,000 2,000 - 3,000 or Ø
10,000
Selected Zone Versus “With & Without”
• Court’s Ruling:– DRI process requires accounting for ALL trips caused by
development, NOT net impact resulting from displacing existing trips to other roadways
– Rationale: if all developers used the argument that trips are diverted and net impact is negligible, no developer would be responsible for mitigation
Westinghouse Gateway Communities, et al. v. Lee County Board of County Commissioners Case Nos: 90-2636DRI and 90-2638DRI, Jan 14, 1991.
Westinghouse Gateway Communities, et al. v. Lee County Board of County Commissioners Case Nos: 90-2636DRI and 90-2638DRI, Jan 14, 1991.
“With & Without” Method
FSUTMS Total Trips(With Development)
50,000 45,000
10,000
FSUTMS Total Trips(Without Development)
48,000 48,000
0
“Net” Impact???
50,000 45,000- 48,000 -48,000 2,000 - 3,000 or Ø
10,000UNACCEPTABLE
Distribution/Assignment Summary
• Definition: Identifying traffic volumes on each roadway link– Total traffic
– Development traffic
– At future analysis years
• Next step after assignment: Determine if any links are: – Operating at adverse LOS
AND
– Carrying a significant volume of development traffic