DEVELOPMENT OF A STATEWIDE MODEL FOR HEAVY

1
DEVELOPMENT OF A STATEWIDE MODEL FOR HEAVY TRUCK
FREIGHT MOVEMENT ON EXTERNAL ROAD NETWORKS
CONNECTING WITH FLORIDA PORTS     
Haitham Al-Deek, Ph.D., P.E. Associate Professor,
Associate Director of CATSS for ITS Programs, and
Director of Transportation Systems
Institute  Jack Klodzinski, Ph.D. Associate
Research Engineer  Ahmed El-Helw Pradeep
Sarvareddy
2
Introduction

• Four Phases
• Phase I
• OBJECTIVE Develop truck trip generation and
  forecasting models for the Port of Miami
  (completed 1998)
• Phase II
• OBJECTIVE Develop truck trip generation and
  forecasting models for 4 more Florida Ports
  (completed July 2001)
• Ports included Palm Beach, Everglades, Tampa,
  Jacksonville

3
Introduction

• Four Phases
• Phase III (in progress)
• OBJECTIVE Develop truck route assignment models
  for two Florida Ports.
• Ports included Tampa, Canaveral
• Phase IV (under review)
• OBJECTIVE Development of statewide truck model
  for transportation networks connecting with deep
  seaports in Florida and forecast truck volumes on
  key interstate routes.

4
Phase II Overview

• Investigate the potential of Artificial Neural
  Networks (ANN) for forecasting inbound and
  outbound heavy truck trips at seaports
• Develop a methodology for forecasting future
  truck volumes at seaports

PHASE II
5
Phase II Overview (Artificial Neural Network
Development)
Run ANN Model
PHASE II
6
Phase II Overview

• ANN used to model all ports
• Palm Beach, Everglades, Tampa, Jacksonville.
• Separate models developed Each port has unique
  characteristics.
• Forecast freight data using historical data for
  input to ANN models
• Port Everglades example

PHASE II
7
Phase II Overview (Port Everglades site
specific data)

• Imported petroleum is very significant (barrels)
• Exported petroleum is insignificant (barrels)
• High containerized cargo activity
• Three main access roads (Eller Dr., Spangler
  Blvd., Eisenhower Blvd.)

PHASE II
8
Phase II Overview(Port Everglades ANN Data)

• Input Data
• Date
• Daily Containers (import/export)
• Daily Tonnage (import/export)
• Monthly Barrels (import)
• Saturday or Sunday

• Output Data
• Date
• Daily Trucks (inbound/outbound)

PHASE II
9
Phase II Overview(Port Everglades Forecast
Results)

• Trucks Daily Average Increase of 991 Trucks (31)
  From Year 2000 to 2005

PHASE II
10
Phase III (in progress) Objective Develop a
predictive truck route assignment model for
adjacent highway networks connecting to Florida
ports

Jacksonville Port
Tampa Port
Palm Beach Port
Everglades Port
Miami Port
PHASE III
11
Phase III Overview(Port of Tampa)

• Three main port locations/five main access roads
• Network Definition
• 21 Nodes
• Geometry, turning movements, traffic control
  signal timing
• 30 Links (road segments)
• Geometry, AADT data (FDOT), phase II truck
  counts, phase III truck counts on selected links
  (master links)
• Highways (I-4, I-75, I-275, SR-41, SR-618, SR-92
  and SR-60)
• Arterials (for example 22nd street)

PHASE III
12
Port of Tampa
Node 9
Node 10
Node 11
Node 8
8
10
9
Node 12
11
25
28
27
7
30
24
Node 7
15
12
16
29
23
Node 16
6
Node 13
Node 6
Hookers Point
17
14
Node 5
13
22
20
5
Node 18
4
Port Sutton
18
21
1
Pendola Point
3
Node 1
Node 14
2
Node 4
Node 3
PHASE III
13
Phase III Overview(Simulation Packages)

• FSUTMS
• 2 network models available for the Tampa Bay
  Region
• CORSIM
• Well known and widely used by government agencies
• VISSIM
• Explore new software and verify results

PHASE III
14
Phase III Overview(Port of Tampa)

• Origin/Destination (O/D) Matrix Definition
• FDOT data
• Trucking company information
• Field truck counts (34 days)
• Model Calibration (23 days)
• Model Validation (11 days)
• Forecasting
• Tampa ANN Model truck volume forecast results
  used for input

PHASE III
15
Phase III Overview(CORSIM and VISSIM)

• No significant difference at the 95 confidence
  level between field truck counts and model output.

PHASE III
16
Phase III Overview(Port of Tampa Modeling
Results)
PHASE III
17
Phase III Overview(Port Canaveral)

• Define initial highway network
• Develop ANN model
• Forecast port truck volumes
• Model Port Canaveral highway network
• Current year
• Forecasted year

PHASE III
18
Phase III Overview(Port Canaveral)
Orlando
7 Nodes 17 Links Highways I-95, US 1, SR 3, SR
520, SR 528
PHASE III
19
Port Canaveral(Daily Truck Counts for ANN Model)
114 Average Daily Trucks
12 Inbound Average Daily Trucks
221 Average Daily Trucks
PHASE II
20
Port Canaveral(Vessel Data) Commodities by
Unit Type
PHASE II
21
Port Canaveral(Daily Vessel Data)
PHASE II
22
Port Canaveral(Historical Vessel Data)
PHASE III
23
Port Canaveral(Network Modeling)

• Origin/Destination (O/D) Matrix Definition
• FDOT data
• Tennant Listing (for truck route info)
• Field truck counts
• Network Forecasting
• Future truck volumes from ANN model used
• Historical vessel data generates future vessel
  data
• Future vessel data input to ANN model

PHASE III
24
Port Canaveral(Network Truck Counts)
US 1 on off ramps
334 Average Daily Trucks
I-95 on-ramp
SR 3 on off ramps
A1A
SR 3
SR 520
PHASE III
25
Phase IV (under review)

• Determine Origin/Destination of trucks
• Accomplish this by
• Phase II ANN application
• Phase III network definition and modeling
• GPS tracking
• Database information
• Interviews with trucking companies

PHASE IV
26
Phase IV (under review)Determine truck routes on
external road network and apply methodology to a
low tonnage port.

Jacksonville Port
Pensacola Port
Panama City Port
Tampa Port
Fort Pierce Port
Palm Beach Port
Everglades Port
Miami Port
PHASE IV
27
Thank You  
28
Phase II Overview(Port Everglades ANN Results)

• No significant difference at the 95 conf. level

PHASE II
29
Phase II Overview(Port Everglades Forecast
Results)
PHASE II
30
Phase III Overview(Port of Tampa) Total Port
Generated Truck Volumes by Percentage on
Interstate Highways

• Year 2005
• Interstate 4
• Inbound 22
• Outbound 37
• Interstate 275
• Inbound 15
• Outbound 22

• Year 2001
• Interstate 4
• Inbound 22
• Outbound 38
• Interstate 275
• Inbound 11
• Outbound 22

PHASE III
31
Port Canaveral(Outbound Network Truck Counts)
Off-ramp 175 trucks On-ramp 10 trucks
Off-ramp 183 trucks
316 Average Daily Outbound Trucks
Off-ramp 89 trucks
PHASE III