Chapter 8 Part I Travel Demand and Traffic Forecasting

1
Chapter 8Part I Travel Demand and Traffic
Forecasting

• From Principles of Highway Engineering and
  Traffic Analysis
• Third Edition
• Fred Mannering, Walter Kilareski and Scott
  Washburn

2
Travel Demand Traffic Forecasting

• Necessary understand the where to invest in new
  facilities and what type of facilities to invest
• Two interrelated elements need to be considered
• Overall regional traffic growth/decline
• Potential traffic diversions

3
Traveler Decisions

• Four key traveler decisions need to be studied
  and modeled
• Temporal decisions the decision to travel and
  when to travel
• Destination decisions where to travel (shopping
  centers, medical centers, etc.)
• Modal decisions how to travel (auto, transit,
  walking, biking, etc)
• Route decisions which route to travel (I-66 or
  Rt 50?)

4
(No Transcript)
5
(No Transcript)
6
Trip Generation

• Objective of this step is to develop a model
  which can predict when a trip will be made
• Typical input information
• Aggregate decision making units we study
  households not individual travelers typically
• Segment trips by type three types 1) work trips
  2) shopping trips and 3) social/recreational
  trips
• Aggregate temporal decisions trips per hour or
  per day

7
(No Transcript)
8
Trip Generation Model

• Typically assume linear form
• Typical variables which influence number of trips
  are
• Household income
• Household size
• Number of non-working household members
• Employment rates in the neighborhood
• Etc.

9
Typical Trip Generation Model
10
Trip Generation Model Example Problem

• Number of peak hour vehicle-based shopping trips
  per household
• 0.12 0.09 (household size) 0.011(annual
  household income in 1,000s) 0.15 (employment
  in the households neighborhood in 100s)
• A household with 6 members annual income of
  50k current neighborhood has 450 retail
  employees new neighborhood has 150 retail
  employees.

11
Trip Generation with Count Data Models

• Linear regression models can produce fractions of
  trips which are not realistic
• Poisson regression can be used to estimate trip
  generation for a given trip type to address this
  problem

12
Poisson Regression Model
13
Estimating Poisson Parameter
14
Example 8.4

• Given
• BZi -0.35 0.03 (household size)
• (0.004) annual household income in 1,000s
• 0.10 (employment in households neighborhood in
  100s)
• Household has 6 members income of 50k lives in
  neighborhood with 150 retail employment what is
  expected no of peak hour shopping trips? What is
  prob household will not make peak hour shopping
  trip?