Transcript Focus Model Class 2 - Denver Regional Council of Governments

FOCUS MODEL OVERVIEW
CLASS TWO
Denver Regional Council of Governments
June 30, 2011
Notes




http://www.drcog.org/index.cfm?page=FocusTechni
calResources
Next week’s class is in Independence Pass from 2-3
PM; New Go To Webinar next week
On-going classes through August 4; Thursdays 2-3
PM in Monarch
Tentative Schedule: Model Steps
July 7
How to Run the Model
Theoretical Underpinning
SQL Database
?????
July 14
July 21
July 28
August 4
Review of General Concepts








1. Logit Models are models that make assign probabilities to
a set of choices for an individual from a list of discrete
choices.
2. The actual choice is made using a monte carlo process.
3. Travel in the model is made on a tour-level, and then a trip
level.
4. We can divide the model into four stages.
5. We use four types of code in the model: T-SQL, C#,
GISDK, and Java.
6. Much of the input and output data is stored in SQL Server.
7. We still have to run parts of our old GISDK code for path
building, skimming and assignment.
8. We are doing this because we can get much finer detail
and answer planning questions better using the model.
Focus Model Flow: 28 Steps
GISDK called from C#:
For DIA, I-E, E-E and Commercial Trips
GISDK called from C#:
GISDK Preprocess
Java:
3. Population Synthesizer
C#
4. PopSyn Output Processor
5. Size Sum Variable Calculator
1. DRCOG Multi-Period Highway Preprocess
2. DRCOG Multi-Period Transit Preprocess
3. DRCOG Transit Preprocess
4. Trip Generation
5. Highway and Transit Skimming
6. Trip Distribution
7. Mode Choice
FEEDBACK
C# Regular Trips
8. Regular Work Location Choice
9. Regular School Location Choice
10. Auto Availability
11. Aggregate Logsum Generation
12.Daily Activity Pattern
13. Exact Number of Tours
14.Work Tour Destination Type
15.Work-Based Subtour Generation
19 . Tour Main Mode Choice
20. Tour Time of Day Choice
21. Intermediate Stop Generation
22. Trip Time of Day Simulation
23. Trip Time Copier
24. Intermediate Stop Location
25. Trip Mode Choice
26. Trip Time of Day Choice
16. Tour Time of Day Simulation
17. Tour Primary Destination Choice
18. Tour Priority Assignment
27. Write Trips to TransCAD
GISDK called from C#:
28. Highway and Transit Assignment
Focus Model Flow: Stage 1
STAGE 1: Make
Population
And Network
STAGE 2:
Run GISDK to
Mode Choice
FEEDBACK
STAGE 4: GISDK
Assignment
STAGE 3: C#
Logit Models to
Create Trips
STAGE 1: Make Population and
Network
Java: Population Synthesizer
 C# to process in database: Size Sum Variable
Calculator; PopSyn Output Processor
 GISDK called from C#: GISDK Preprocess
Creating networks for example

Population Synthesizer
ACS or PUMS
Disaggregate Data
Questions?
Aggregate Data that
We Need to Match:
Economic Forecasts,
Land Use Forecasts
Disaggregate
Population With the
Right Portions Matching
the Economic and Land
Use Forecasts
People come out as disaggregate unique entities
with many characteristics.
Example Family:
Mother, Age 33
Part Time Service
Worker
Father, Age 34
Full Time Education
Worker
Son, Age 4
Pre-School Student
Family Income :
$61,000
Data Gets Processed into a Database
Structure
Database Structure Created and Filled
People
Database Structure Created and Filled
Places
What are:
Tours, Half Tours, Half Tour Stops, Trips
WORK
HOME
STORE
Database Structure Created and NOT
Filled YET
Travel
Household Monte Carlo

PopSyn assigns the zone they live in. Then the
model randomly assign the households to a point
within the zone.
Give Each Household a Housing Unit to
Live in. A disaggregate X-Y point
GISDK:
Preprocess
Highway Network
Inputs
Transit
Network
Inputs
Socio-economic
Inputs
Network Processing
& Data Preparation
Area Type
Trip Generation
Highway
Transit
Skimming
Skimming
Trip Distribution
Parking Cost
Mode Choice
Time-of-Day
Highway
Assignment
Transit Assignment
Stage One is done. Everything outside
feedback loop is done.
Now we have :
 1. A synthesized population
 2. A database filled with point locations and
people, model variables
 3. A set of highway and transit networks ready for
use.
Focus Model Flow: Stage 2
STAGE 1: Make
Population
And Network
STAGE 2:
Run GISDK to
Mode Choice
FEEDBACK
STAGE 4: GISDK
Assignment
STAGE 3: C#
Logit Models to
Create Trips
Highway Network
Inputs
GISDK:
For Skimming
and External/
Commercial Trips
Transit
Network
Inputs
Socio-economic
Inputs
Network Processing
& Data Preparation
Area Type
Trip Generation
Highway
Transit
Skimming
Skimming
Trip Distribution
Parking Cost
Mode Choice
Time-of-Day
Highway
Assignment
Transit Assignment
Stage 2:
GISDK Through Mode Choice



All C#, trip-making components use travel time and
distance skims for highway and transit
Internal-External, External-External, trips destined
to DIA and Commercial Trips created from
Compass GISDK
Later these trips get combined with the regular
Internal-Internal Trips from C# into matrices by
time-of-day by mode
Skim matrices(distance)
Time Periods for Skims












TIMES OF DAY
Highway Times of Day
AM1: 6:30 – 7:00 AM;
AM2: 7:00 – 8:00 AM;
AM3: 8:00 – 9:00 AM;
OP2: 9:00 – 11:30 AM;
OP3: 11:30 AM – 3:00 PM;
PM1: 3:00 – 5:00 PM;
PM2: 5:00 – 6:00 PM;
PM3: 6:00 – 7:00 PM;
OP4: 7:00 – 11:00 PM.
OP1: 11:00 PM – 6:30 AM;
Transit Times of Day
AM: 6:30- 9:00 AM
MD: 9:00 AM- 3:00 PM PM
PM: 3:00 PM -7:00 PM
EL: 7:00 PM – 6:00 AM
Trip Tables (this is commercial)
GISDK Through Mode Choice



So we run trip generation, trip distribution, and mode
choice for the funky trips
This is run from C# calling GISDK.
The C# pops open a TransCAD window and calls
the macros that have been specified to run.
Stage 2 is done.





Now we have a LOT of matrices:
All highway and transit skims
A set of commercial and external trips O-Ds
A set of DIA trips O-Ds and modes
And all stage one outputs: population, networks, a
ready database.
Focus Model Flow: Stage 3
STAGE 1:
PREPROCESS
STAGE 2 :GISDK
Through Mode
Choice
FEEDBACK
STAGE 4: GISDK
Assignment
STAGE 3: C#
Logit Models to
Create Trips
Talking time: Let’s talk about ourselves

What are the set of choices that make you travel
like you do?
 What
is the highest priority?
 What is unique about you that guides your choices?
 What are the smaller choices you make each day?
 What strange behaviors to you have that would be
really hard to model?
The steps in Stage 3. Mostly Logit
Models. The heart of the model.











8. Regular Work Location Choice
9. Regular School Location Choice
10. Auto Availability
11. Aggregate Logsum Generation
12.Daily Activity Pattern
13. Exact Number of Tours
14.Work Tour Destination Type
15.Work-Based Subtour Generation
16. Tour Time of Day Simulation
17. Tour Primary Destination Choice
18. Tour Priority Assignment
19 . Tour Main Mode Choice
20. Tour Time of Day Choice
21. Intermediate Stop Generation
22. Trip Time of Day Simulation
23. Trip Time Copier
24. Intermediate Stop Location
25. Trip Mode Choice
26. Trip Time of Day Choice
27. Write Trips to TransCAD
Long Term Choices
8. Regular Work Location Choice
 9. Regular School Location Choice
 10. Auto Availability
 11. Aggregate Logsum Generation

Long Term Choices:
Regular Workplace Location
Where will I work?
Final Choice: Two nests- Work at Home, Work Outside
Home
Zone and X,Y Location of Work
Type of Model: Nested Logit
Inputs: (What do you think predicts?)
Number of Jobs by type in in a zone
Distance from Home to Work
Full Time Worker or Part Time Worker
Job Sector
Accessibility of Work Location from Home

Here’s how the choice looks, sent back
to the database Persons table:
Long term choice 2:
Regular School Location
Final Choice:
Zone and X,Y Location of School
Type of Model: Multinomial Logit
Inputs: (What do you think predicts?)
Grade Level in School
Distance to School from Home
Income Group
Older Sibling’s School