Incorporating%20Uncertainty%20Analysis%20into%20Forecasting - PowerPoint PPT Presentation

About This Presentation
Title:

Incorporating%20Uncertainty%20Analysis%20into%20Forecasting

Description:

Title: Revised Evaluation Criteria for New Starts Author: RyanJ Last modified by: kenneth.cervenka Created Date: 8/28/2001 10:27:28 AM Document presentation format – PowerPoint PPT presentation

Number of Views:44
Avg rating:3.0/5.0
Slides: 7
Provided by: Ryan1404
Learn more at: https://ampo.org
Category:

less

Transcript and Presenter's Notes

Title: Incorporating%20Uncertainty%20Analysis%20into%20Forecasting


1
Incorporating Uncertainty Analysis into
Forecasting
  • Group Discussion
  • AMPO Travel Modeling Work Group
  • Chicago, Illinois
  • October 1, 2009

2
Why Bother?
  • A new way of thinking about forecasting
  • Insights, not just model mechanics
  • Range of possible outcomes
  • Another approach for quality control checks
  • Information for decision makers
  • Upper/lower bound, most likely
  • Honesty in presentation of forecasts
  • Predicted-versus-actual outcomes

2
3
Example Forecast Build-up
  • Series of forecasts for
  • Today
  • Plus future transit network
  • Plus new transit behaviors
  • Plus future trip tables
  • Plus future highway congestion
  • Plus future parking costs
  • Plus alternative land use (?)

Choice riders Park/ride
etc. Guideway effects
4
Example Honolulu Rail Study
Rail-trip Build-up Forecast Forecast Forecast Forecast Forecast Forecast Forecast Forecast
Attribute 1 2 3 4 5 6 7 8
Person trips -- 05 18 18 30 30 30 30
Highway speeds -- 05 18 18 05 05 30- 30
Bus speeds 30 30 18 30 30 05 30 30
Transit network 30 30 18 30 30 30 30 30
Transit demand 05a 05p 18 18p 30h 30b 30c 30
Rail trips per day 60k 73k 73k 77k 86k 72k 118k 87k
Notes - Transit demand 05a is the 2005 on-board
survey. - Bus speeds 05 are based on highway
speeds from the assignment of 2005 person trips
onto the 2030 highway network. - Highway speeds
30- are from the assignment of 2030 person trips
onto the 2005 highway network.
5
Example Specifications
6
Discussion Points
  • Model documentation
  • Assessment of model plausibility
  • What it does/does not know well
  • Markets, modes, behaviors
  • Results of forecast tests
  • Consistency with similar projects
  • Key drivers of the forecasts
  • Demographic/network inputs
  • Alternative assumptions

6
Write a Comment
User Comments (0)
About PowerShow.com