HomeMy WebLinkAboutWORK ORDER - RFP - P745 TRAVEL DEMAND MODELCity of Port Collins
Admi rative Services t
Purchasing Division
WORK ORDER FORM
PURSUANT TO AN AGREEMENT BETWEEN
THE CITY OF FORT COLLINS
AND
Kimley-Horn and Associates, Inc
DATED April 13, 2000
Work Order Number 1
Purchase Order Number
Protect Title Travel Demand Modeling
Commencement Date May 1, 2000
Completion Date December 31, 2000
Maximum Fee (time and reimbursable direct costs) 1 i $7,600 00, 2) $65,300,
3) $ 37,000 00 Not to exceed One Hundred Nine Thousand Nine Hundred Dollars
($109,900 00)
Protect Description Travel Demand Modeling
Scope of Services 1) Evaluate Work done to date, 21 Create Base Year Model, 3) Perform
Modeling work for the MPO's next LRTP
Acceptance
Professional agrees to perform the services
identified above and on the attached forms
in accordance with the terms and
conditions contained herein and in the
Professional Services Agreement between
the parties In the event of a conflict
between or ambiguity in the terms of the
Professional Services Agreement and this
work order (including the attached forms)
the Professional Services Agreement shall
control
Pgrofe o
Dat0
CC Purchasing
��4 +12-flao
1
User
The attached forms consisting of
pages are hereby accepted and
incorporated herein, by this reference, and
Notice to Proceed is hereby give
NFRT&AO and C t of FIRT.U011jins
By yJ
Ronald Phillips, Executiv Director
B O'Neill II, CPPO
r of Purchasing and Risk
Date;
1
256 IN MountainAvenue • PO Box 580 • Fort Collins, CO 80522-0580 • (970) 221-6775 • FAX (970) 221-6707
destination survey as a basis for establishing time -of -day rates by trip purpose
Next, we suggest using TransCAD's derauit hourly factors QtranscadltabUrourlybin),
which are based on National Cooperative Highway Research Program Study #187,
for filling the gaps and verifying data from the origin -destination survey. After these
steps have been taken, the model can be simulated incrementally to produce time -
of -day results for use in air quality conformity analyses
■ Considerable research has been conducted by the modeling community on volume -
delay equations used in the assignment process. This research includes
experiments with alternate forms of equations that are different from the classic
Bureau of Public Roads formula established more than 40 years ago We believe
this model, given its wide range of areas, should have the capability of supporting
varying forms of volume -delay equations by functional classification (e g., collector,
arterial, freeway, etc.) and area (e g., Central Business District, suburban, rural, etc.)
Using the model expectation guidelines identified in the study design, and depending
upon the involvement role chosen by the MPO, we estimate this effort will take between 9
and 12 months This timeline allows for development of the ongin-destination survey and
time -of -day modeling
C Perform Modehng WOrIz for the MPO's Next LRTP
The initial model simulation performed in this area of development is known as the
'Existing Plus Committed" (E+C) alternative This includes land use and socio-economic
data representing a 20-year horizon, and a network that contains all existing
transportation facilities plus those committed improvements Results from this initial
simulation form the basis for establishing the alternatives for the region's Long Range
Transportation Plan (LRTP).
After the results from the 20-year E+C simulation have been processed, Kimley-Horn's
approach to this area of modeling is to clearly define the alternatives This definition also
includes identifying the expected performance reporting (e g , vehrcie-miles of travel,
vehicle -hours of travel, etc.) needed from the simulations for effective evaluations We
propose developing a standard library of line -by-line reports and GIS graphics illustrating
model output for agreed -upon variables.
We estimate this effort to take between 2 and 4 months, depending upon the number of
alternatives established for the LRTP and Kimsey-Horn's involvement in this area of the
project As with other areas of mode{ development, we can perform the alternative
simulations, or Kimley-Horn could coach the MPO with their development, or our
involvement could be a blending of the two,
Scope of Services
The Request for Proposals identifies six key modeling areas in which our team may
assist NFRTAQPC in developing the regional MPO model While the task scoping in
these six areas has not formally been identined at this time, Kimley-Horn has
prepared the following approaches to developing these six areas
A Evaluate Toile Done to Date
Kimley-Horn has worked with the MPO over the past few months to initially evaluate
where the regional model is in As translation from MfnUTP to TransCAD Our
preliminary evaluation has been that while the model data has been successfully
translated and conflated to actually match map lines, work still needs to be done to
calibrate the models (trip generation, trip distribution, mode split, and trip assignment)
and validate the results to a base year We also noted that some modeling
advantages in the TransCAD software were not being used in the present job stream
and that these advantages may improve model results
Based on this evaluation, we discussed with NFRTAQPC the options they may want
to consider for implementing TransCAD While there are present day projects that a
TransCAD model may benefit (e g , Mason Street Corridor Study in Fort Collins), it is
our understanding that the MPO is primaiily interested in readying the model to
support development of NFRTAQPC's Long Range Transportation Plan (LRTP)
Our approach to this area of the scope wdf be to briefly revisit our discussions and
conclusions with the model stakeholders During this visit, we want to make sure this
path is Sill desired by the MPO, and that there have been no significant events to
change this direction We recommend this discussion take place during project
initiation
After we've had this discussion, we will work with NFRTAQPC to determine the
remaining effort required to develop the MPO model in TransCAD. As this document
is developed, we will want to focus our efforts on the following -
First, we will need to determine the remaining datasets that will need to be
collected to properly calibrate the models and validate the results. There are
common "everyday" data available, such as traffic ceun's and land use /socio-
economic information, that will need to be collected from either the MPO or
current member jurisdictions and catalogued Some of this information was
mentioned in the RFP, including the 1998 Mobility Report Card and information
from the City of Fort Collins' modeling effort for the Mason Street Corridor As
part of the study design, We will work with NFRTAQPC to identify those datasets,
determine how the data should be obtained (i.e., formats), and agree upon
collection responsibilities
■ Next, we will need to tackle the unique model datasets and their treatment For
example, trip generation and trip drsinbubon models rely heavily upon Household
Travel or Origin -Destination surveys for their validation During our past
conversations with the MPO, we discussed this survey in particular and noted its
importance to this model —particularly for validating city-to-abj travel between
Greeley, Fort Collins, and Loveland The problem with these datasets is that
they are difficult and often expensive to obtain, however, we believe there are
several opportunities to consider here, Through our arscussions, we can devise
the best plan for actually developing these datasets, borrowing from other
models, or making data assumptions
■ Thud, we believe it is necessary to discuss expectations for the model's results
Frankly, when consultants are involved in travel demand model development, the
expectations about a model's performance are often a source of conflict with a
client Often these conflicts are related to the expectations during the validation
phase of model development and revolve around a particular link not producing
"acceptable" simulated volumes when compared to the ground count, While
models are not perfect predictors of human travel behavior, they can be
developed to provide the MPO with a reliable tool for identifying transportation
investments in the NFRTAQPC area Let's eliminate the potential for a conflict
situation in the beginning by talking about expectations in the study design
phase
■ Fourth, the study design should address how much effort should be spent in
developing the model's four steps Specifically, our concern is with the mode
spirt task The existing travel oemand model has this task; however, it may not
be functioning effectively in the TransCAD environment it this is the case, it may
be necessary to initially work around this step to keep the model on a prompt
schedule for developing the work necessary to implement the LRTP,
■ And finally, we want to discuss the model development roles —for both MPO and
consultant staff We consider this very ;mportant—we want NFRTAQPC
involvement rn all phases of model development At the conclusion of this
project, it will be the MPO staff's responsibility to continue modeling on projects
other than the LRTP Truthfully, the only way NFRTAQPC staff can continue with
the TransCAD is if they have a working knowledge of the model's development
Thus, the study design must clearly identify the involvement that the MPO wants
in this process Involvement can range from a "hands -off" approach in which
everything is summarized at the model development's conclusion, to a process in
which the MPO staff does the development effort and the consultant acts as a
"coach" in the process, to somewhere in the middle
The product of this effort will be a model development study design document that will
guide,the balance of the project Aber meeting with the model stakeholders, Kimley-
Horn estimates the study design can be developed within the following two to three
weeks
B, Create Base Year Model
The study design document developed upon project initiation will form the basis for
Kimley-Horn's approach to creating the base year model for the NFRTAQPC region
As the model is created, there are varying factors that will need to be addressed
Incorporating additional TransCAD features that are not used in present job
stream For example, the initial translation from MuiUTP relies heavily upon that
program's coding techniques These techniques typically generalize descriptions
about the network to fit the model into the outdated DOS operating system
TransCAD, on the other hand, gives the modeler considerable flexibility in
network coding that can improve the overall calibration and validation We
recommend taking advantage of this Flexibility and will advise the NFRTAQPC
staff about the additional variables and attributes that may be introduced into a
new job stream in TransCAD
■ Working with the MPO to establish a current Origin -Destination Survey to help
validate the model's trip generation, trip distribution, and mode split steps
Conversations with NFRTAQPC staff have suggested that this survey is required
for a successful validation of the TransCAD model We have made some initial
recommendations to the MPO about how this survey could be economically
completed with assistance from Kimley-Horn, These recommendations include
assisting with developing the survey instrument, using fraternal organizations
(e g , Kiwanis, Jaycees, Chamber of Commerce, stq as the subjects of the
survey, working with MPO and student staff to reduce and geocode the data, and
then developing the datasets for use in the TransCAD model,
■ Since our initial evaluation of the TransCAD model, the MPO staff has identified
the need for time -of -day modeling to be incorporated into the job stream This is
to accommodate NFRTAQPC's need to post -process the results for air quality
conformity analyses We recommend a mult)-faceted approach to developing this
model First, we recommend using data from the aforementioned origin-
destination survey as a basis for establishing time -of -day rates by trip purpose
Next, we suggest using TransCAD's default hourly factors ( transcadltab\hourly bin),
which are based on National Cooperative Highway Research Program Study #187,
for filling the gaps and verifying data from the origin -destination survey. After these
steps have been taken, the model can be simulated incrementally to produce time -
of -day results for use in au quality conformity analyses
■ Considerable research has been conducted by the modeling community on volume -
delay equations used in the assignment process This research includes
experiments with alternate forms of equations that are different from the classic
Bureau of Public Roads formula established more than 40 years ago We believe
this model, given its wide range of areas, should have the capability of supporting
varying forms of volume -delay equations by functional classification (e g , collector,
arterial, freeway, etc ) and area (e g , Central Business District, suburban, rural, etc,)
Using the model expectation guidelines identified in the study design, and depending
upon the involvement role chosen by the MPO, we estimate this effort will take between 9
and 12 months This timeline allows for development of the origin -destination survey and
time -of -day modeling
C Perform Modeling Work for the MPO's Next LRTP
The initial model simulation performed in this area of development is known as the
'Existing Plus Committed" (E+C) alternative This includes land use and socio-economic
data representing a 20-year horizon, and a network that contains all existing
transportation facilities plus those committed improvements Results from this initial
simulation form the basis for establishing the alternatives for the region's Long Range
Transportation Plan (LRTP).
After the results from the 20-year E+C simulation have been processed, Kimley-Horn's
approach to this area of modeling is to clearly define the alternatives This definition also
includes identifying the expected performance reporting (e g , vehicle -miles of travel,
vehicle -hours of travel, etc.) needed from the simulations for effective evaluations We
propose developing a standard library of line -by-line reports and GIS graphics illustrating
model output for agreed -upon variables
We estimate this effort to take between 2 and 4 month;,, depending upon the number of
alternatives established for the LRTP and Krmley-Horn; involvement in this area of the
project. As with other areas of model development, we ran perform the alternative
simulations, or Kimsey -Horn could coach the MPO with their development, or our
involvement could be a blending of the two
Admf6rative Services
Purchasing Division
WORK ORDER FORM
PURSUANT TO AN AGREEMENT BETWEEN
THE CITY OF FORT COLLINS
AND
Kimley-Horn and Associates, Inc
DATED April 13, 2000
Work Order Number 1
Purchase Order Number T > / 2, �16—�-)
Project Title Travel Demand Modeling
Commencement Date May 1, 2000
Completion Date December 31, 2000
Maximum Fee (time and reimbursable direct costs) 1) $7,600 00, 2) $65,300,
3) $37,000 00 Not to exceed One Hundred Nine Thousand Nine Hundred Dollars
($109,900 00)
Project Description- Travel Demand Modeling
Scope of Services 1) Evaluate Work done to date, 2) Create Base Year Model, 3) Perform
Modeling work for the MPO's next LRTP
Professional agrees to perform the services
identified above and on the attached forms
in accordance with the terms and
conditions contained herein and in the
Professional Services Agreement between
the parties In the event of a conflict
between or ambiguity in the terms of the
Professional Services Agreement and this
work order (including the attached forms)
the Professional Services Agreement shall
control
P�rofes�sio
Date
CC Purchasing
Acceptance
11
User
The attached forms consisting of
pages are hereby accepted and
incorporated herein, by this reference, and
Notice to Proceed is hereby give
NFRT&A C and City of rt C Ilins
By
Ronald Phillips, E�xecut e Director
By /,` ¢G��
Jam B O'Neill II, CPPO
Din ctor of Purchasing and Risk
Management
Date C-)l�'
256 W Mountain Avenue • PO Box 580 • Fort Collins, CO 80522-0580 • (970) 221-6775 • FAX (970) 221-6707
Scope of ,Services
The Request for Proposals identifies six key modeling areas in which our team may
assist NFRTAQPC in developing the regional MPO model While the task scoping in
these six areas has not formally been identified at this time, Kimley-Horn has
prepared the following approaches to developing these six areas
A Evaluate Worlx Done to Date
Kimley-Horn has worked with the MPO over the past few months to initially evaluate
where the regional model is in its translation from MinUTP to TransCAD. Our
preliminary evaluation has been that while the model data has been successfully
translated and conflated to actually match map lines, work still needs to be done to
cahorate the models (trip generation, trip distribution, mode split, and trip assignment)
and validate the results to a base year We also noted that some modeling
advantages in the TransCAD software were not being used in the present lob stream
and that these advantages may improve model results
Based on this evaluation, we discussed with NFRTAQPC the options they may want
to consider for implementing TransCAD While there are present day projects that a
TransCAD model may benefit (e g , Mason Street Corridor Study in Fort Collins), it is
our understanding that the MPO is primarily interested in readying the model to
support development of NFRTAQPC's Long Range Transportation Plan (LRTP)
Our approach to this area or the scope will be to briefly revisit our discussions and
conclusions with the model stakeholders During this visit, we want to make mre this
path is still desired by the MPO, and that there have been no significant events to
change this direction We recommend this discussion take place during project
initiation
After we've had this discussion, we will work with NFRTAQPC to determine the
remaining effort required to develop the MPO model in TransCAD. As this document
is developed, we will want to focus our efforts on the following
First, we will need to determine the remaining datasets that will need to be
collected to properly calibrate the models and validate the results There are
common "everyday" data available, such as traffic counts and land use /socio-
economic information, that will need to be collected from either the MPO or
current member jurisdictions and catalogued Some of this information was
mentioned in the RFP, including the 1998 Mobility Report Card and information
from the City of Fort Collins' modeling effort for the Mason Street Corridor As
part of the study design, we will work with NFRTAQPC to identify those datasets,
determine how the data should be obtained (i e , formats), and agree upon
collection responsibilities.
Next, we will need to tackle the unique model datasets and their treatment For
example, trip generation and trip distribution models rely heavily upon Household
Travel or Or gin -Destination surveys for their validation During our past
conversatlons with the MFO, we discussed this survey in particular and noted its
importance to this model —particularly for validating city -to -city travel between
Greeley, Fort Collins, and Loveland The problem with these datasets is that
they are difficult and often expensive to obtain; however, we believe there are
several opportunities to consider here Through our discussions, we can devise
the best plan for actually developing these datasets, borrowing from other
models, or making data assumptions
a Third, we believe it is necessary to discuss expectations for the motleys results
Frankly, when consultants are involved in travel demand model development, the
expectations about a model's performance are often a source of conflict with a
client. Often these conrlicts are related to the expectations during the validation
phase of model development and revolve around a particular link not producing
"acceptable' simulated volumes when compared to the ground cow While
models are not perfect predictors of human travel behavior, they can be
developed to provide the MPO with a reliable tool for identifying transportation
investments in the NFRTAQPC area Let's elimmare the potential for a conflict
situation in the beginning by talking about expectations in the study design
phase.
® Fourth, the study design should address how much effort should be spent in
developing the model's four steps. Specifically, our concern is with the mode
split task The existing travel demand model has this task; howevai, it may not
be functioning effectively in the TransCAD environment If this is the case, it may
be necessary to initially work around this step to keep the model on a prompt
schedule for developing the work necessary to implement the LRTP
® And finally, we want to discuss the model development roles —for both MPO and
consultant staff We consider this veryimportant—wewant NFRTAQPC
involvement in all phases of model development At the conclusion of this
protect, it will be the MPO staffs responsibility to continue modeling on protects
other than the LRTP Truthfully, the only way NFRTAQPC staff can continue with
the TransCAD is it they have a working knowledge of the model's development
Thus, the study design must clearly identify the involvement that the MPO wants
in this process Involvement can range from a "hands -off" approach in which
everything is summarized at the model development's conclusion, to a process in
which the MPO staff does the development effort and the consultant acts as a
"coach" in the process, to somewhere in the middle
The product of this effort will be a model development study design document that will
guide the balance of the project After meeting with the model stakeholders, Kimley-
Horn estimates the study design can be developed within the following two to three
weeks
B. Create Base Year Model
The study design document developed upon project initiation will form the basis for
Kimley-Horn's approach to creating the base year model for the NFRTAQPC region
As the model is created, there are varying factors that will need to be addressed
Incorporating additional TransCAD features that are not used in present job
stream For example, the initial translation from MinUTP relies heavily upon that
program's coding techniques These techniques typically generalize descriptions
about the network to fit the model into the outdated DOS operating system.
TransCAD, on the other hand, gives the modeler considerable flexibility in
network coding that can improve the overall calibration and validation We
recommend taking advantage of this flexibility and will advise the NFRTAQPC
staff about the additional variables and attributes that may be introduced into a
new job stream in TransCAD
a Working with the MPO to establish a current Ongin-Destination Survey to help
validate the model's trip generation, trip distribution, and mode split steps
Conversations with NFRTAQPC staff have suggested that this survey is required
for a successful validation of the TransCAD model We have made some initial
recommendations to the MPO about how this survey could be economically
completed with assistance from Kimley-Horn, These recommendations mclude
assisting with developing the survey instrument, using fraternal organizations
(e g., Kiwanis, Jaycees, Chamber of Commerce, etc ) as the subjects of the
survey, working with MPO and student staff to reduce and geocode the data, and
then developing the datasets for use in the TransCAD model.
4 Since our initial evaluation of the TransCAD model, the MPO staff has identified
the need for time -of -day modeling to be incorporated into the job stream. This is
to accommodate NFRTAQPC's need to post -process the results for air quality
conformity analyses. We recommend a multi -faceted approach to developing this
model First, we recommend using data from the aforementioned origin-