Development of Long-Distance Multimodal Passenger Travel Modal Choice Model

Prepared for:
Office of Highway Policy Information
Federal Highway Administration
1200 New Jersey Avenue, S.E.
Washington, DC 20590

July 23, 2012

PDF Version 1.1 MB

You will need the Adobe Acrobat Reader to view the PDFs on this page.

Table of Contents

• Mathematical Techniques for Long Distance Passenger Travel Modal Choice Modeling
• Data Sources Used for Long-Distance Passenger Traveler Modeling
• Factors That Influence Mode Choice
• Mathematical Models for Predicting Long-Distance Passenger Mode Choice

ABBREVIATIONS

1.0 INTRODUCTION

• 2.1 Background
• 2.2 Methodology
• 2.3 Literature Review Findings
  • 2.3.1 Mathematical Techniques for Long-distance Passenger Travel Modeling
  • 2.3.2 Data Sources used for Long-distance Multimodal Passenger Travel Modal Choice Modeling
  • 2.3.3 Factors that Influence Long-distance Passenger Travel Modeling
• 2.4 Practice Review Findings
  • 2.4.1 Metropolitan Transportation Commission in California
  • 2.4.2 State of Michigan
  • 2.4.3 American Automobile Association (AAA)
• 2.5 Discussion

• 3.1 2001 National Household Travel Survey
  • 3.1.1 Trip Purpose and Mode Choice
  • 3.1.2 Prediction Factors from 2001 NHTS
• 3.2 Data Sources to Supplement the 2001 NHTS
  • 3.2.1 Economic Factors
  • 3.2.2 Availability of Transportation Infrastructure
• 3.3 Summary of Predictive Factors Used in Mode Choice Modeling
• 3.4 Descriptive Analysis
• 3.5 Statistical Modeling Methodology
  • 3.5.1 Multinomial Logit Model
  • 3.5.2 Model Estimation
• 3.6 Model Estimation Results
  • 3.6.1 Full Prediction Models
  • 3.6.2 Reduced Prediction Models
  • 3.6.3 Model Limitations
• 3.7 Discussion

• 4.1 Methodology
• 4.2 Results

5.0 CONCLUSIONS

6.0 REFERENCES

List of Tables

• Table 2-1. Major National-Level Intercity Travel Demand Models for the United States
• Table 3-1. Summary of Predictive Factors Used in Mode Choice Modeling
• Table 3-2. Number of Long-Distance Trips Used in Modeling by Trip Purpose and Travel Mode
• Table 3-3. Descriptive Statistics (Means and Standard Deviations) for Each Model Factor by Trip Purpose and Transportation Mode
• Table 3-4. Coefficient Estimates (Standard Errors) for Full Set of Multinomial Logit Models of Travel Mode Choice
• Table 3-5. Marginal Effects (Standard Errors) Estimates for Full Set of Multinomial Logit Models of Travel Mode Choice
• Table 3-6. Marginal Effects (Standard Errors) Estimates for Full Set of Multinomial Logit Models of Travel Mode Choice
• Table 3-7. Marginal Effects (Standard Errors) Estimates for Reduced Set of Multinomial Logit Models of Travel Mode Choice
• Table 4-1. Comparison of Actual and Model-Predicted Aggregate Mode Shares by Trip Purpose

List of Figures

• Figure 1-1. Percent of Long-Distance Passenger Trips by Mode Share Choice According to 2001 NHTS
• Figure 2-1. Visualization of Simple Multinomial Logit Model
• Figure 3-1. Plot of Air Travel Price Index from 1995 to 2011
• Figure 3-2. Locations of Large, Public-Use Airports from NTAD2011
• Figure 3-3. Locations of Amtrak Stations from NTAD2011
• Figure 3-4. Locations of Light Rail Stations from NTAD2011
• Figure 3-5. MTA NYC Subway and LIRR Stations
• Figure 3-6. Locations of Large Bus Stations
• Figure 3-7. Number of Transportation Hubs Within 25 Mile Buffer for Each Survey Respondent
• Figure 3-8. Visualization of Multinomial Logit Model
• Figure 3-9. Fitted Polynomial Trend of Route Distance vs. Predicted Travel Mode Choice – Business Travel
• Figure 3-10. Fitted Polynomial Trend of Route Distance vs. Predicted Travel Mode Choice – Pleasure Travel
• Figure 3-11. Fitted Polynomial Trend of Route Distance vs. Predicted Travel Mode Choice – Personal Business Travel
• Figure 4-1. Distribution of Predicted Aggregated Mode Shares Relative to Observed Aggregated Mode Shares Across Iterations by Mode Choice (Business Trips)
• Figure 4-2. Distribution of Predicted Aggregated Mode Shares Relative to Observed Aggregated Mode Shares Across Iterations by Mode Choice (Pleasure Trips)
• Figure 4-3. Distribution of Predicted Aggregated Mode Shares Relative to Observed Aggregated Mode Shares Across Iterations by Mode Choice (Personal Business Trips)