Skip to Main Content
Peer Review Program
HomeUnited States Department of Transportation, HomeContact UsSite Map
Travel Model Improvement Program - TMIP
About TMIPTMIP ServicesClearinghouse, selectedConferences and CoursesContactsTravel Model DiscussionsTRANSIMSLinks

BROWSE

SEARCH

ORDER DOCUMENTS

NEWSLETTER

Clearinghouse

TMIP Newsletter Archive

TMIPConnection
The Travel Model Improvement Program Newsletter

Issue 5, May 1996

The Travel Model Improvement Program is sponsored by:
U.S. Department of Transportation
Federal Highway Administration
Federal Transit Administration

Table of Contents
Second Travel Model Improvement Conference
Upcoming Conference on Urban Design, Telecommuting and Travel Behavior
TMIP Review Panel Changes
FHWA Launches Advanced Urban Travel Demand Forecasting Course
Upcoming Conferences Related to TMIP
TMIP Program Update
TMIP Provides Technical Assistance
TRANSIMS: An MPO Perspective
Electronic Reports – Making Your Life Easier?
Listing of Available Reports
TRANSIMS Travelogue

Second Travel Model Improvement Conference

The second annual conference sponsored by the Travel Model Improvement Program was held in Daytona Beach, Florida on December 4-6, 1995. The purpose of the conference was to provide an update on the TMIP research and related activities and to review and discuss current modeling issues, problems and potential solutions. Over 125 persons attended the conference.

Each of the three days began with a general session covering important topics of interest to a wide range of professionals. Individual workshop sessions that followed focused on specific travel modeling topics and were formatted to facilitate discussion.

Four workshop sessions convened during the afternoon of the first day of the conference. Those sessions covered the following:

The plenary session on day two of the conference spotlighted TRANSIMS activity and project products. Members of the TRANSIMS team from Los Alamos National Lab (LANL) provided attendees with further insight into the development, capabilities, and requirements of TRANSIMS.

Twelve workshop sessions were held following the plenary session on the second day of the conference. Topics of discussion in those sessions included:

A general session covering model improvements being implemented in other countries and outside of TMIP was convened on the final day of the conference. Presentations in that session focused on travel forecasting activities in Canada and Sweden as well as an activity based disaggregate model system and the activity/ tour based models from Portland, Oregon.

Following adjournment of the general conference on Wednesday, a meeting with potential vendors of new software was held to discuss computer hardware and software issues and needs relative to dissemination of existing and future travel demand models.

Table of Contents

Upcoming Conference on Urban Design, Telecommuting and Travel Behavior

In recent years, alternative designs for suburban areas have been proposed and carried out in developments around the country. These developments, known variously as neo-traditional design, transit oriented design and pedestrian oriented design, provide an opportunity to reduce automobile travel while maintaining accessibility. These designs seek to bring some of the amenities found in central cities to suburban areas, including the ability to walk and bike to daily activities and to access regional transit facilities. Some research has been carried out on measuring the characteristics of these neo-traditional developments and in assessing the effects of these new urban design patterns on changing travel behavior. To date, there has been only limited application of these results in planning practice.

In a similar manner, telecommuting programs are being planned and implemented to reduce automobile travel as well as improve the flexibility in workers schedules and worker productivity. Telecommuting programs from home, telework centers, and mobile offices offer the opportunity to replace vehicle trips with electronic trips. There have been a number of research studies on the effects of telecommuting on travel behavior. Some work has examined the manner in which housing and community designs might be changed as telecommuting expands. However, few transportation planning agencies address telecommuting options in their studies.

A conference has been scheduled to examine developments in urban design and telecommuting and identify their potential impacts on reducing automobile travel, congestion, air pollution, accidents and energy consumption. The conference will identify transferrable principles and results and determine where additional research is needed. Dr. Martin Wachs is the conference chair.

The conference will be October 27-30, 1996 at the Williamsburg Hospitality House in Williamsburg, Virginia. For further information, contact Lynette Engelke at (817) 277- 5503, fax: (817) 277-5439, e-mail: l-engelke@tamu.edu.

Table of Contents

TMIP Review Panel Changes

The TMIP Review Panel was established to offer comments on the design and implementation of activities in the program. The review panel is chaired by Dr. Martin Wachs and is composed of leading professionals with experience in transportation planning, environmental protection, and land development (see the accompanying table for the list of current panel members). The panel has provided many important suggestions and direction to the project sponsors.

In the interest of not overburdening panel members, and to add new perspective, a process of refreshing the panel with new members has been instituted. Two original panel members, Anne Geraghty, of the California Air Resources Board, and Bob Harvey, of Seattle Metro, have stepped down from the panel. Anne and Bob represented air quality agencies and the transit community, respectively. Their efforts were much appreciated. The two new panel members are Annette Liebe and Keith Killough.

Annette Liebe is a project coordinator with the Oregon Department of Environmental Quality. She is responsible for researching and developing rules implementing the air quality criteria applicable to transportation planning under the 1990 Clean Air Act Amendments (CAAA). Ms. Liebe also provides assistance to the state in implementing the transportation conformity requirements of the CAAA and the Intermodal Surface Transportation Efficiency Act. Before joining the Department, she was the Air and Water Program Director of the Oregon Environmental Council. Annette is a member of the State Bar Association in Colorado and Oregon.

Keith Killough is the Deputy Executive Officer–County Wide Planning for the Los Angeles County Metropolitan Transportation Authority (LACMTA). He is responsible for strategic planning for county-wide transportation issues and represents the LACMTA in regional and state forums for the development and analysis of transportation policies and programs. He is also one of the principal staff coordinators for the agency's Long Range Transportation Plan. Previously he was Planning Manager at the Southern California Rapid Transit District, where he oversaw service analysis, travel simulation and forecasting, and the guideway and facilities planning functions.

TMIP Panel Members

Table of Contents

FHWA Launches Advanced Urban Travel Demand Forecasting Course

FHWA piloted a new course on urban travel demand forecasting in Dallas, Texas, May 14-16, 1996. This was the first course hosted by the TMIP Regional Training Center in Arlington, Texas, in cooperation with the National Highway Institute (NHI). The course was offered to members of the travel modeling community working for federal, state, MPO, city or county agencies, and consulting firms. Courses offered through regional training centers will emphasize this mix of professionals. The Texas Transportation Institute sent a targeted mailing to organizations or individuals with at least three years of hands-on experience with computerized four-step models.

FHWA has regularly presented an introductory level training course on Urban Travel Demand Forecasting through NHI (Course No. 15254). The purpose of this new course, Advanced Urban Travel Demand Forecasting (NHI course no. 15260), is to build upon the introductory travel demand forecasting course, emphasizing advanced practices for travel demand modeling at the system level. The course consolidates the best procedures and methodologies for estimating the travel impacts of multi-modal transportation and land use alternatives and demonstrates applications of these procedures.

Topics covered include advanced practices for land use allocation, network description, trip generation, trip distribution, mode choice, peaking analysis, traffic assignment, feedback analysis and speed post- processing. Model development, validation and application issues are also discussed. The course is oriented towards current modeling practitioners, i.e., those who have a thorough understanding of the traditional four-step travel demand forecasting process. Upon completion of the course, participants should be able to understand and select appropriate procedures to improve the capability of their four-step models to estimate demand impacts of a broad range of multi-modal infrastructure investment and transportation/land use policy options.

The course will be offered in additional locations across the country. Agencies wishing to host the course should contact Lynn Cadarr, of NHI, at (703) 235-0528 or Gordon Shunk, of TTI, at (817) 277-5503. For more information on technical content of the course, contact Patrick DeCorla-Souza of FHWA at (202) 366-4076.

Table of Contents

Upcoming Conferences Related to TMIP

Table of Contents

TMIP Program Update

As part of TMIP's on-going program, new commitments have recently been made to support activities in Tracks A (outreach), B (near-term model improvements), D (data collection), and Track E (land use). The projects described below all have committed funding for the 1995/1996 fiscal year. Additional projects may be added if additional funding becomes available.

Track A

Outreach, training, and technical assistance will continue with staff support of the Texas Transportation Institute. A specific commitment funds the October 1996 Urban Design, Telecommuting, and Travel Behavior conference described on page 2. Money has also been set aside to partially fund an additional regional training center.

Track B

Innovative Applications — Addressing challenging issues with traditional models in recent years has resulted in some valuable work in the areas of time-of-day, auto ownership, speed and post- processing. Under this project, each of the subjects listed above, and other innovative applications, will be documented.

Guidance for the Estimation of Logit Models — Much work has taken place since the document "A Self-Instructional Text For Estimating Mode Choice Models" was produced. This document will be updated including lessons from estimation efforts around the country, and information on nested logit models.

Destination and Mode Choice of Zero Car Households — It has long been assumed that the travel patterns of households without vehicles available are different than those having autos, yet zero car households are usually treated the same in trip distribution and mode choice. Using a survey with an enriched sample of households without a vehicle, this project will study the travel behavior of such households and develop techniques to better address that issue in the forecasting process.

Activity-Based Models — Two projects are being initiated in this area. The first provides technical support to the efforts of Portland's Tri-Met in the development of new models incorporating activity-based forecasting concepts. The process will be documented for subsequent distribution. The second project will develop additional techniques for using activity diary data, given the increasing availability of such data sets. This effort will examine the data for activity frequencies as well as for the effects of life cycle, gender, work status, and resource measures (income, number of vehicles, number of adults and number of workers).

Track D

Survey Non-Response — Given the problem of inevitable "non-responding" populations (e.g., low income, non-English speaking, renters, low education), two projects have been initiated to review options for reducing the impact of non- response and adjusting travel survey estimates accordingly. The first will develop guidelines on or for the application of non-response adjustment methods and illustrate those methods with concrete examples of real data set estimates. The second project will support an add-on to the Denver area household survey, with special emphasis on non-response issues.

Guidelines for Data Collection for Small and Medium-Size Areas — Over half the travel surveys conducted this decade have been in areas over 750,000 in population; experience in smaller areas is limited, as are the resources to conduct surveys. This project will develop guidelines for these smaller areas to acquire the data needed to support planning and analysis activities.

Track E

Examples of Land Use Forecasting — At a recent meeting of the Urban and Regional Information Systems Association (URISA) and at the 1995 TMIP Land Use Conference, the need for land use forecasting examples was discussed. This project will assemble a set of case studies illustrating the use of different land use forecasting approaches. Analytical methods, such as DRAM and EMPAL, the use of an expert panel and combinations of analytical methods and expert panels will be covered. Examples of the data preparation necessary to support land use forecasting will also be included.

Table of Contents

TMIP Provides Technical Assistance

The Technical Assistance service of TMIP is operational. Guidance is available for working with existing models, for new information to upgrade existing models and describing the emerging new TRANSIMS models. To date, requests from three agencies have been handled by TTI. Any transportation planning agency may contact TTI for technical assistance or for information about training. TTI staff are responsible for all activities in Track A—Outreach, including technical assistance and training among others. All requests for technical training and assistance will be handled initially by TTI. Some requests may be passed on to be addressed by consultants, including Cambridge Systematics. More information about training and technical assistance available through TMIP is on the TMIP home page (http://tmip.tamu.edu/) from which e-mail messages can be sent or information can be downloaded. Interactive response to questions and comments is available through the TMIP computer bulletin board (817-277-7674). Live technical assistance is available by calling TTI at 817-277-5503 or by telefax at 817-277- 5439. Give us a call!

Table of Contents

TRANSIMS: An MPO Perspective

By Ken Cervenka, North Central Texas Council of Governments

The North Central Texas Council of Governments (NCTCOG) is the Metropolitan Planning Organization (MPO) for the 5,000 square-mile, four million-person Dallas-Fort Worth region. With planning funds provided by the Federal Highway Administration and the Texas Department of Transportation, we are assisting Los Alamos National Laboratory (LANL) in a case study application of the TRansportation ANalysis and SIMulation System (TRANSIMS). Our role on this project is:

Traffic and Network Data

The traffic data provided to LANL includes 800-zone vehicle trip tables, for home based work, home based non- work, non-home based, and for other purposes, in production attraction and origin-destination format. Preparation of the network data consists of three steps:

Database Management Issues

While there are legitimate concerns about the extensive network data requirements of TRANSIMS, we do not see this as an insurmountable obstacle to a successful real-world application. If the computational requirements for a regional microsimulation are reasonable, the only significant network data issue is the level of accuracy that needs to be coded. Here is one strategy:

In other words, the man-hours to be allocated towards intersection coding and other "network cleaning" activities can be tailored to the specific needs of a transportation study. Most intersections in a region could perhaps be initially coded with default configurations, then edited, over time, in a graphical environment. The results of the sensitivity tests will help us answer this question.

Final Comments

One reason we at NCTCOG are excited about TRANSIMS is that it represents the first serious, well-funded attempt to fully integrate detailed traffic operations with a regional travel demand model. As additional IOCs and case studies are advanced, we hope that other MPOs and transportation planning agencies will have an opportunity to work closely with LANL on further TRANSIMS development.

Table of Contents

Electronic Reports – Making Your Life Easier?

Most of us have read articles or heard reports on the advantages of the Internet and electronic or digital information distribution. The TMIP project is adding to this electronic push by requesting documents to add to our World Wide Web site. TMIP has prepared documentation that provides a general introduction to electronic document formats, a basic recommendation on the files you should produce or request to be produced, and some specific guidelines for handling text, graphics, and tables. This article is a summary of the recommendation, and more detail is available on the TMIP web site (http://tmip.tamu.edu/) or from Kim Fisher at (202) 366-4054. This document is not intended as a guide to scanning existing hard copy reports, but instead is a guide for producing electronic files along with the traditional hard copy.

Why Now?

There are some very real reasons that electronic publishing has exploded in the last couple of years. The new ability to view graphics and tables of all kinds means that more documents can be distributed in their entirety.

Limitations on printing funds is another major impetus to move to electronic distribution.

Basic Recommendations

We recommend that each final report be available in hard copy and two different electronic formats. The first electronic format should be the original word processing format. The second electronic format should be a HTML document. Both of these electronic versions should include all text, tables, and graphics as delivered in hard copy format.

The cover graphic and basic document information should be in the first part of the document. The basic document information should include the title, originating agency or department, the author, publication date, a one-two page abstract, key words, and point of contact for more information on the document.

The size of the file, size of graphics, and presentation of graphics should all be considered very carefully. Graphics which take more than a minute to load or which require more memory than is available on the average user's computer should be avoided. The actual size of the graphic, when viewed by a browser, should generally be limited to an average screen size.

Table of Contents

Listing of Available Reports

Addresses for report contacts:

Kim Fisher
c/o Federal Highway Administration
400 7th St SW, HEP-22, Room 3232
Washington, DC 20590
Phone: (202) 366-4054, Fax: (202) 366-3713

Lynette Engelke
Texas Transportation Institute
201 E. Abram Street, Suite 600
Arlington, TX 76010
Phone: (817) 277-5503, Fax: (817) 277-5439

Norman Paulhus
Technology Sharing Program
U.S. DOT
400 7th Street, SW, M-453
Washington, D.C. 20590

Table of Contents

TRANSIMS Travelogue — May 1996

TRANSIMS TRAVELOGUE describes current activities within the TRANSIMS project.

What is TRANSIMS?

The TRansportation ANalysis and SIMulation System (TRANSIMS) is one part of the multi-track Travel Model Improvement Program sponsored by the U.S. Department of Transportation and U.S. the Environmental Protection Agency. Los Alamos National Laboratory (LANL) is leading this major effort to develop new, integrated transportation and air quality forecasting procedures necessary to satisfy the Intermodal Surface Transportation Efficiency Act and the Clean Air Act and its amendments.

TRANSIMS is a set of integrated analytical and simulation models and supporting data bases. The TRANSIMS methods deal with individual behavioral units and proceed through several steps to estimate travel. TRANSIMS predicts trips for individual households, residents and vehicles rather than for zonal aggregations of households. TRANSIMS also predicts the movement of individual freight loads. A regional microsimulation executes the generated trips on the transportation network, modeling the individual vehicle interactions and predicting the transportation system performance. Motor vehicle emissions are estimated using traffic information produced by TRANSIMS. TRANSIMS' major advantage for air quality analysis is the detail it provides regarding motor vehicle operation.

Project Approach

LANL is are developing an interim operational capability (IOC) for each major TRANSIMS component: Household and Commercial Activity Disaggregation, Intermodal Route Planner, Transportation Microsimulation, and Environment (primarily air quality). As each IOC is ready and with the collaboration of a selected MPO, we will complete a specific case study to confirm the IOC features, applicability, and readiness. This approach should provide timely interaction and feedback from the TRANSIMS user community and interim products, capabilities, and applications.

The Traffic Microsimulation is emphasized in the first IOC, with the goal of having it ready for testing in mid-1996. We are working with the selected MPO, the North Central Texas Council of Governments (NCTCOG) (Dallas-Fort Worth), on the case study that the IOC should support. (See related story on pages 6-7.)

Revised Case Study

In the November 1995 TRANSIMS Travelogue, we described a proposed case study that emphasized examination of several freeway alternatives for reducing traffic congestion. Since that time we have revised the case study to highlight unique TRANSIMS features and to maintain the focus on the traffic microsimulation — the emphasis of the first IOC.

The revised case study will continue to examine the transportation system performance within a 16-square-mile region of interest (ROI) along the Lyndon B. Johnson Freeway (I-635) in the Dallas area. NCTCOG is updating the case study network to include local streets for a 25-square-mile ROI to allow study of both the boundary effects on the ROI and the necessity for modeling local streets. As before, but with a slightly different view, infrastructure alternatives to traffic congestion reduction will be examined.

The new case study will illustrate TRANSIMS' ability to partition the benefits and costs of a transportation infrastructure change among subpopulations of travelers. For demonstration purposes, the study will focus on a major shopping/business center, the Galleria area, and the travelers to or from that area. A local system alternative will involve a non-freeway system change that would be expected to benefit the Galleria travelers, and possibly other travelers to some extent. A global (non-local) alternative, for example, an additional freeway lane in both directions through the ROI, also is intended to benefit the Galleria travelers, but would be expected to benefit all travelers to the same or comparable extent.

The proposed study matrix is shown in the figure below. Case 1 represents a microsimulation of the existing transportation system. Although it was not in initial plans, it is believed that micro-simulation results, such as travel times, can be fed into the route planner and derive new traveler routes that account for the observed micro-simulation dynamics. There is some uncertainty in the ability to do this iteration at this point in the project. Currently a single iteration (planner » microsimulation » planner » microsimulation) is intended, and the second microsimulation results will be used for comparisons between cases.

  With Galleria Trips Without Galleria Trips
 
Base Infrastructure		 Case 1
Replan?		  
Case 4
 
Local Alternative		 Case 2
Replan?		  
Case 5
 
Global Alternative		 Case 3
Replan?		  
Case 6

Case 2 represents the local alternative. For this case the planner-microsimulation iteration will account for the system change, and the microsimulation will use those revised trip plans. Similarly, Case 3 represents the global alternative.

Relevant measures of effectiveness for the comparisons between cases are being defined. In addition to the traditional measures such as VMT, VHT, etc., these variances will be included in these quantities. These Measures of Effectiveness (MOE) will quantify the benefits obtained from the alternatives.

In doing such comparisons, the MOEs must be carefully chosen. The comparisons will distinguish between the benefits to the total subpopulation and the benefits normalized on a per person basis or per vehicle basis. Thus, there will be a comparison of the subpopulation total improvement in travel time and the average improvement in travel time. The effort will distinguish between local and global MOEs. A person may benefit locally from an improvement, but it may be only a small portion of his overall trip. Similarly, the MOEs are time sensitive. Recognizing these caveats, MOEs will be carefully chosen before analyzing and comparing the subpopulation benefits as discussed in the following paragraphs.

For the cases represented by the right-side boxes in the preceding figure, travelers' trip plans that originate or terminate in the Galleria area will be removed from the trip plan set of the respective left box. Trips will not be replanned for the non-Galleria travelers when excluding the Galleria travelers. Replanning would introduce additional travelers' trips into the ROI and the system being studied, disrupting the sought partitioning. The beauty of TRANSIMS is that it permits such mathematical manipulations and abstractions to partition and understand both the demand and the supply sides of the transportation system.

Thus, though the non-Galleria travelers' trip plans will account for the Galleria travelers, in the traffic microsimulation execution, the non-Galleria vehicles will not interact with the Galleria vehicles (because they won't be on the network). Comparing the results of the right cases with the left cases using measures such as time of travel, travel time variability, and trip plan satisfaction should yield the impact of the Galleria travelers on the other travelers.

However, such direct comparisons are not useful in themselves because the non-Galleria travelers impact the Galleria travelers in a comparable way, probably even worse, as there are many more non-Galleria travelers. The interesting results occur when comparing the subpopulation benefits resulting from the transportation system alternatives. Because TRANSIMS tracks individual travelers through-out the simulation (planner and micro-simulation), how each alternative benefits subpopulations can be measured. For example, from Case 1 to Case 2 how the local alternative benefits each subpopulation can be measured: Galleria and non-Galleria travelers. In addition, from Case 4 to Case 5 whether the non-Galleria subpopulation benefits even if the Galleria subpopulation is not present when the non-Galleria travelers execute their travel plans can be determined.

To illustrate the possible outcomes of the case study, suppose the benefits for the non-Galleria and Galleria subpopulations from Case 1 to Case 2 are denoted by N2 and G2 respectively, and for the non-Galleria subpopulation from Case 4 to Case 5, by N5. If all three benefits are significant and comparable, then the local alternative affects the subpopulations equally, and each should expect responsibility for an equal partition of the alternative's cost.

If N5 is significant, but N2 is not, then the benefit to the non-Galleria travelers arises just by reducing the demand, which could be a reduction by any subpopulation. If G2 is significant and N2 and N5 are not, then the Galleria subpopulation benefits whereas the non-Galleria subpopulation doesn't regardless of the Galleria subpopulation presence. In this instance, the Galleria subpopulation should expect to finance the alternative's costs. If N2 and N5 are significant, but G2 is not, then the benefits are incurred primarily by the non-Galleria subpopulation, and the Galleria subpopulation should not expect to contribute to the alternative's cost.

Other possible combinations can be examined and benefits and financing equity to the subpopulations inferred. The point is that the TRANSIMS approach of following travelers throughout the microsimulation allows partitioning of benefits and liabilities among subpopulations. Furthermore, TRANSIMS permits additional MOEs, e.g., variance in time of travel and trip plan satisfaction by subpopulation, not present in current planning models.

A Practitioner's Perspective

Most of the work to date in the TRANSIMS Initiative has focused upon the development of new capabilities, and has of necessity concentrated on software design and development. While interesting in academic terms, practitioners have had a difficult time assessing the applicability of these developments to their practice. The case study offers transportation planners and engineers the opportunity to examine the potential of large scale transportation systems simulation within the context of a real–world transportation planning problem. By focusing on the microsimulation—the most computationally demanding element of TRANSIMS–the case study will give potential users a frame of reference with which to gauge the capabilities and scalability of the system.

One very noteworthy capability unveiled during the case study will be the simultaneous collection of statistics for both links and the travelers traversing them. While currently available traffic simulation and travel forecasting models collect some of these data, none can handle both (or the interactions between them) in a unified manner. Moreover, data on the variability of various measures of effectiveness will be collected as well. Information on variability is often as influential as absolute measures in travel decision- making, especially in goods movement and public transport planning.

The case study has been deliberately designed to illustrate the ability to conduct equity analyses — the so-called "winners and losers" identification. By removing certain classes of travelers (those destined to and from the Galleria in the case study) from the simulation without removing the perception of the their effect on the network by the remaining users, finally both efficiency and equity issues can be examined that are foremost in the minds of public policy makers. While focusing on a single user class, it will be possible to subdivide the simulation elements (both networks and users) into multiple groups, allowing the visualization and analysis of joint distributions which cannot be analyzed using current tools, or even observed in the real world. The ability to assess whether targeted user groups are affected in the intended way will be an important analytical contribution. By combining measures of network response and changes in user behavior, a truly holistic analyses of complex transport problems and issues will be conducted.

Interim Route Planner

Goals have been achieved to provide an interim route planner capability that supports the microsimulation IOC. Those goals are:

  1. Provide a representative set of NCTCOG trip plans so the microsimulation IOC can be demonstrated.
  2. Establish data structures between the HCAD and Planner, and between the Planner and Micro- simulation, consistent with long-term design.
  3. Implement basic versions of the Trip Plan Generation and Goal Measurement phases.

For the first goal, 10.3 million trip plans for individual travelers across the entire NCTCOG region were generated. Primary inputs included the Dallas-Fort Worth network augmented with mid-link addresses (labeled "parking accessories," which represent the travelers' origins and destinations), individual traveler demographics, and individual traveler activities (type, location, and start, end, and duration time intervals). One-half million trip plans were identified as passing through the 25 square-mile region for the traffic microsimulation (vicinity of the LBJ Freeway and Dallas North Tollway). These plans were stored in the TRANSIMS Oracle database for subsequent use by the microsimulation.

The "quality" of these plans was discussed with NCTCOG. Although a thorough calibration and validation process cannot be performed at this early point in the Planner's development, it is useful to compare these trip plans with known data and NCTCOG experience. This information is needed for the upcoming Case Study. Future Planner calibration and validation will occur after the preference adjustment and superposition phases, and established the planner-micro- simulation feedback loops are established.

Several actual link traffic counts and Planner link volumes were compared over a 24-hour period. [Note: It is more appropriate to comparemicrosimulation link volumes with actual counts. Trip plans represent travelers' intentions; microsimulation output represents "actual" execution of trip plans. We expect, though, that trip plan link volumes will have magnitudes and link type distributions similar to actual counts.]

In general, these trip plans capture the 24-hour volume trends (e.g., AM, PM, and noon peak hours). However, the total link volume may be too high, and volume by roadway classification (freeway, arterial, collector, etc.) may not be distributed correctly. An interim superposition algorithm is believed to be necessary before any substantive conclusions are drawn. In the meantime, these plans will allow demonstration of the microsimulation capabilities and to performance of the upcoming case study.

Second, the Planner's basic data structures are in place. Traveler and activity data from the Household and Commercial Activity Disaggregation are being transferred to the Planner and are placing trip plans into the trip plan database.

Third, both the basic trip plan generation and goal measurement phases are implemented. The generation phase uses stochastic and optimal routing algorithms. Consistent with the underlying premise that many travelers do not use optimal routes, the stochastic router is relied upon primarily. Individual traveler link preferences, or biases are imbedded within this router. These include a bias to head toward the destination, avoid turning, and select and remain on a freeway when far from the destination.

The goal measurement phase determines the adequacy of each candidate plan (i.e., the overall "goodness" of the specific set of sequential links in the plan) according to three travel goal thresholds: time, distance, and cost. A default process to select a good plan is used. Each traveler has his/her own set of goal thresholds, and these are generated using personal demographic attributes and expected travel times and distances between the origin and destination. For those travelers who cannot "find" a goal satisfying route/trip plan, the plan is selected (among the set of candidate plans) that minimizes deviations from the traveler's goals.

For further information about the TRANSIMS program, please contact:

Dr. LaRon L. Smith
Los Alamos National Laboratory
Mail Stop F606
PO Box 1663
Los Alamos, New Mexico 87545
Phone: 505-665-1286
Fax: 505-665-0879
E-mail: llsmith@lanl.gov

Table of Contents

To subscribe to this free newsletter send an e-mail to TMIP@tamu.edu or contact Gary Thomas at (ph.) (979) 458-3263, (fax) (979) 845-6001, (mail) Gilchrist, Room 112, Texas Transportation Institute, Texas A&M University System, 3135 TAMU, College Station, TX 77843-3135