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

PEER REVIEW PROGRAM

Application Package

Peer Reviews: Status and Reports

Peer Review Panel Functions and Organizations

Peer Reviews: Status and Reports

Status of TMIP Peer Reviews — Denver Regional Council of Governments (DRCOG)

Location: Atlanta, Georgia
Date: February 3-5, 2004
Peer Review Host Agency Contact: Guy Rousseau, Atlanta Regional Commission, Atlanta, GA
Peer Review Panelists: Gordon Garry, Sacramento Area Council of Governments, Sacramento, California; Keith Lawton, Portland Metro Planning Department, Portland, Oregon; Jeff May, Denver Regional Council of Governments, Denver, Colorado; Michael Morris, North Central Texas Council of Governments, Arlington, Texas; Rita Walton, Maricopa Association of Governments, Phoenix, Arizona

Report in PDF PDF files require the use of a PDF reader. You may download a free copy of Adobe Acrobat Reader to view these files. If you already have the Adobe Acrobat Reader installed on your computer it should automatically open to view the files when you click on the link to the document.

Executive Summary
Background
Presentation and Discussion
Summary of Panel Recommendations
Next Steps and Summary
Appendix: Agenda, Participants, Presentations

Executive Summary

The following report summarizes the results of a Peer Review Panel held through the Travel Model Improvement Program (TMIP), which is sponsored by the Federal Highway Administration (FHWA). The Atlanta Regional Commission (ARC) hosted the 2.5-day peer review composed of Metropolitan Planning Organizations (MPO), private consultants, and Federal and local transportation and resource agencies. The primary aim of the peer review was to provide ARC with an independent assessment of its travel demand modeling system and to compare the model with industry standards.

ARC’s current travel model, which is used to develop travel forecasts, was implemented using input received in a previous peer review on modeling in June 2000.

As a result of the 2000 review, ARC’s modeling structure evolved, taking advantage of previous panel comments. The model enhancements led to analyses and results included in the regional transportation plan. Through the Peer Review, the progress and performance of ARC’s most current travel model update were assessed. Areas in which ARC might improve its modeling procedures were also identified. Specific subjects considered and evaluated include:

Initially, ARC staff and the consultants under contract to ARC, Parsons Brinkerhoff, gave presentations on the existing state of the ARC modeling practice, including descriptions of collected data, assumptions, and objectives for modeling procedures. ARC’s planning partners, a group that included local transportation and resource agencies, also provided succinct lists of recent ARC modeling successes and areas for improvement.

Participants in the peer review included transportation and land use model experts from the Denver Regional Council of Governments (DRCOG), Sacramento Area Council of Governments (SACOG), Portland Metro Planning Department, North Central Texas Council of Governments, Maricopa Association of Governments (MAG), Parsons Brinkerhoff, PB Consult, Georgia Department of Transportation (GDOT), Georgia Regional Transportation Authority (GRTA), Metropolitan Atlanta Rapid Transit Authority (MARTA), Georgia Department of Natural Resources Environmental Protection Division (EPD), Environmental Protection Agency (EPA), FHWA Atlanta Resource Center, FHWA GA Division, and the Volpe Center. The peer review was held February 3 through February 5, 2004 at the ARC offices in Atlanta, Georgia.

Background

The Atlanta Regional Commission (ARC) functions as the Metropolitan Planning Organization (MPO) for 10 Georgia counties: Cherokee, Clayton, Cobb, DeKalb, Douglas, Fayette, Fulton, Gwinnett, Henry, and Rockdale. Each of these counties pays dues to be an ARC member. ARC’s transportation planning and modeling area encompasses 13 counties, an area likely to be expanded to 18 counties in the future.

The region’s prevalent transportation planning challenge is extensive growth. Currently, over 3,500,000 people live in an area of more than 2,981 square miles, and since the Atlanta region is not located at a seaport, the city relies heavily on its rail and highway networks. The area supports 16,500 roadway miles, 300 rail cars on 46 one-way rail miles, 850 buses and vans serving 5,600 miles. These networks are expected to continue experiencing both occupancy and mileage growth.

The number of total workers in the Atlanta region is also expected to continue expanding. Over the last 10 years, the Atlanta region workforce increased by 33.6%, representing an increase nearly three times higher than the national average (11.5%). Average travel time to work in the region increased 20.2% during this same time period, from 26 minutes in 1990 to 31.2 minutes in 2000, the fourth highest percent change in the nation. More recently, in 2002, people averaged 31.8 daily vehicle miles traveled (VMT), while transit ridership was over 560,000 people per weekday.

The regional growth and extensive roadway network has contributed to the designation of the 13 Atlanta region modeling counties as non-attainment areas under the 1990 Clean Air Act Amendment’s (CAAA) National Ambient Air Quality Standards (NAAQS). Twenty counties have been proposed for the 8-Hour Ozone Standards non-attainment.

ARC’s transportation demand model, which predicts the impact of travel growth, also helps ARC and its partners assess how potential transportation improvements might affect air quality conformity. The model also supports the federally required long-range Regional Transportation Plan (RTP), the basis upon which an annual short-range Transportation Improvement Plan (TIP) is developed. The current ARC travel simulation model follows the traditional four-step modeling structure: generation, distribution, mode choice, and network assignment.

In previous peer reviews, modeling experts suggested that there were several areas ARC might investigate for model improvements. ARC acknowledged these recommendations and has made efforts to significantly refine and enhance its travel simulation and forecasting capabilities. Improvements to the model as identified by ARC include:

An evaluation of both these current model improvement efforts and long-term goals, such as the development of a tour-based approach to trip generation and trip distribution, were the subjects of the February 3-5, 2004 Peer Review Panel.

Presentation and Discussion

Introduction and Four Step Modeling Process
Jane Hayse, Guy Rousseau

A peer review of ARC’s travel demand modeling process was held in June 2000. Based on recommendations from the panel of modeling experts, ARC made many improvements to its travel demand model. New data, assumptions, and model structure were used to create a mechanism by which more accurate and reliable transportation forecasts can be made.

The first model enhancement consisted of converting from a TRANPLAN platform to a TP+ platform. The TP+ platform is based on the established four-step transportation model – trip generation, trip distribution, mode split, and traffic assignment. These steps are combined to allow modelers to estimate vehicle trip counts and distribution based on household level information.

In order to determine at the household level vital information for generating estimates of home-based and non-home-based trips, ARC used a Household Travel Survey (HTS). In the survey, ARC asked all members of a household five years old and older to document all of their travel information for a 48-hour period. This level of information constituted "full participation." Once data were collected, it was determined that 8,069 households had fully participated. In total, the demographic data collected reflected 20,472 persons, 14,371 vehicles, and 131,323 trips made during the two-day travel periods.

ARC’s model also takes advantage of recent data updates. Census datasets and college/university enrollment numbers from 2000 are now included in model runs. Data improvements have necessitated revisions to the parameters utilized in the Disaggregated Residential Allocation Model and Employment Allocation Model (DRAM/EMPAL). Here, ARC disaggregates 589 census tracts into 1683 internal traffic analysis zones (TAZ). Before the June 2000 peer review, ARC used 948 TAZs. In order to accommodate the internal TAZ geometric configuration, ARC revisited how it defined and treated centroid connectors. ARC maintained the same number (57)of external stations.

Other areas of model improvement are as follows:

These improvements were integral to the creation of the 2030 Aspirations Plan, a plan that is the first phase in the process to create the 2030 RTP ("Mobility 2030"). The 2030 Aspirations Plan is intended to provide planners with the information necessary to understand the range of transportation solutions required to reduce congestion and improve mobility.

Population and Employment Forecasts
Bart Lewis

To appraise future transportation requirements based on economic and population (births and immigration) growth, ARC uses the DRAM/EMPAL model. The model combines demographic and econometric data for forecasts. DRAM/EMPAL produces forecasts at the census-tract level that must be disaggregated to the TAZ level. The disaggregated information feeds into the TP+ platform, which operates at the TAZ level. The model provides a method by which localized forecasts can be made.

These small-area forecasts are produced using a two-step procedure. First, control forecasts for the entire region are made using the Interactive Population and Econometric Forecast (IPEF Model). This cohort component model, which separates the population into 18, 5-year age groups, econometrically forecasts future jobs, then estimates the population required to fill those jobs. In ARC’s view, this method is appropriate because the job market created by an expanding economy is the driving force attracting new residents to the Atlanta region. For the second step, the IPEF created control forecasts are disaggregated into more localized estimates using 5-year iterative cycles of DRAM/EMPAL.

The DRAM component of the model forecasts household location based on prior employment location, prior population location, land use, household-income distribution, and travel cost. Similarly, the EMPAL component forecasts employment location based on prior employment location, prior population location, employment attractiveness, and travel cost.

As new data become available, ARC updates the DRAM/EMPAL model. The Atlanta region’s Livable Centers Initiative, a program that funds 42 towns and activity centers to collect demographic and economic data, has been a productive source of information helping to give ARC a level of community understanding it did not have previously. ARC attempts to receive input on these and other new data as well as on new DRAM/EMPAL assumptions and results by sharing the information with partnering agencies. Opportunities for active involvement, such as local meetings, are held, and reports requesting local government comments are drafted. By providing intermediate model data and results, ownership among local governments is fostered. Local governments are able to compare more easily their own local forecasts to ARC’s regional forecasts. The challenge for ARC is that local governments have different ambitions and ideas about future growth.

Mobile Source Emissions Modeling Process
Tracy Clymer

ARC seeks to continue providing consistency and credibility to various reviewers within the context of air quality and conformity determination in Atlanta. To do so, ARC utilizes a traditional link-based procedure to estimate mobile source emissions. This process satisfies Federal transportation conformity regulations that direct regional emissions analyses. It is also consistent with methodology used to develop emissions inventories needed to establish the Motor Vehicle Emissions Budget (MVEB) as part of the SIP.

ARC’s mobile source emissions model calculates levels of NOx and VOC emissions (precursors to formation of the criteria pollutant ozone) that the regional highway and transit networks produce. The emissions model has undergone significant enhancement since the June 2000 Peer Review, including updates to reflect the recalibration of the travel demand model after the release of the 2000 Census data and expansion to four time-of-day periods. In the model, daily vehicle trips are broken down into four time periods – the morning peak from 6:00 a.m. to 10:00 a.m., the mid-day peak from 10:00 a.m. to 3:00 p.m., the evening peak from 3:00 p.m. to 7:00 p.m., and the night hours from 7:00 p.m. to 6:00 a.m. This extension is important because output received from the time-of-day periods can be used to more accurately determine temporal emissions levels and impacts.

The mobile source emissions model, like the overall transportation demand model, has been converted to the TP+ platform. Within ARC’s TP+ emissions interface, two primary variables affect mobile source emissions estimates: vehicle miles traveled (VMT) and speed estimates. The model interprets loaded networks for each time-of-day period then post-processes speeds and VMT. Emissions model output are listed below:

Environmental Protection Agency (EPA) guidance recommends HPMS based forecasts for conformity analyses. ARC currently compares average daily, summer-adjusted HPMS VMT estimates for 2000 to average daily, summer-adjusted travel model VMT for 2000. The adjustment factors are applied at the link level using HPMS functional class codes.

HPMS Adj. Factori = 2000 HMPS VMTi/2000 Model VMTi
where, i = HPMS functional class

The adjustment factors generated by the model ranged from 0.69 for "rural interstates" to 2.76 for the "urban other freeways" classification. The panel commented that this range (and potential error) might have resulted from inconsistencies between ARC and HPMS classification systems and roadway type definitions.

Along with HPMS adjusted VMT, speed affects mobile source emissions estimates. Currently, ARC applies speed post-processors based on volume-delay functions to congested-flow speed estimates. The volume-delay functions are applied for each time of day period for individual road classes. The post-processed congested speed is calculated by multiplying the free flow speed on a link by an appropriate speed function, a function of facility type and the volume/capacity ratio as determined by a built-in look-up table. The panel recommended that ARC revisit the reasoning behind including these speed post-processors (See Summary of Panel Recommendations).

ARC’s emissions model has also now been converted to integrate MOBILE6.2 emissions factors model. ARC uses the MOBILE6.2 model to predict gram per mile emissions of volatile organic compounds (VOC) and nitrous oxides (NOx) under various conditions by drive cycle. The Georgia Environmental Protection Division (EPD) collaborates with ARC to generate MOBILE6.2 model input data. Emissions factors are produced for 2.5 mph and 3 mph to 65 mph in one mph increments.

ARC plans to continue to update its current emissions model and process, which is summarized in Chart 1 below. In the future, ARC anticipates incorporating PM2.5 emission factors and making model modifications as necessary based on the 8-HR Ozone and NAAQS requirements. ARC’s air quality modelers also look forward to updating the emissions model according to the peer panel’s recommendations.

Chart 1: Summary of ARC’s Emissions Modeling Process

Emissions Modeling Process Summary

Multi-year Activity-Tour Based Model
Bob Donnelly, Peter Vovsha, Mark Bradley, John Bowman

ARC consultants, Parsons Brinkerhoff, gave presentations on the status of the development of a tour-based transportation demand model for the Atlanta region. Tour-based models differ from traditional models in that they predict "tours" rather than "trips," while simultaneously calculating the main components of a person’s travel during a day. Tour-based models are based on actual data and can capture the interdependence of different activities in a trip chain and provide better understanding of non-home-based travel and travelers’ responses to transportation policies. These models can also explicitly model intra-household interactions, providing enhanced temporal resolution to transportation demand. Through log-sum calculations, this makes clear estimations of the time structure of transportation use possible. Each of these advantages of a tour-based model would allow ARC to develop transportation demand analyses, which include many added variables, more quickly, without losing detail or increasing computational complexity.

In 2003, the Atlanta region’s tour-based model design was completed. Development of the model structure and subsequent refinements were fashioned from lessons learned by other metropolitan areas studying tour-based models. As a part of this progress, a Population Synthesizer component was created, programmed, and put through initial testing. The Population Synthesizer creates synthetic populations of households and their members, two characteristics that affect travel activity and choices. The Synthesizer creates household populations through two basic steps, the Balancer step and the Drawer/Validator step. The Balancer determines initial estimates of the number of households in each zone then adjusts these estimates by iterative proportional fitting, a mathematical procedure that combines information from two or more datasets. The Drawer/Validator uses the Balancer’s simulated household population data to back-cast population estimates for comparison to actual data.

Other aspects of tour-based model development that have been completed include model estimation of household car ownership, household daily activity patterns (DAP) by eight person types, tour mode choice, tour time of day/duration, household joint tour generation, household allocated maintenance tour generation, and individual discretionary tour generation. Consultants have also incorporated grid cell data into the model, better allowing for the modeling of pedestrian activity.

The Atlanta region tour-based model has produced some initial empirical results. These results, some summarize in Table 1 below, are similar to results generated from other tour-based models.

Table 1. Summary of Initial results of Atlanta Region tour-based model.

Occurrence Trend
Duration Effect Longer travel times extend the duration of a tour and shift work tours earlier in the day (constrained arrival time).
Congestion Higher travel impedance in peak period shifts outbound and return
Effect trips to other times of day.

ARC consultants intend to finalize the first tour-based model for the Atlanta region during FY04. In order to do this, the following model refinements are scheduled:

Questions regarding the model’s ability to estimate air quality, for example, still remain. ARC is reluctant to be the first MPO in the country to try to attain air quality conformity determination with a tour-based model. To date, the tour-based model weaknesses are not well documented, nor has there been explicit enumeration of the model’s benefits. Over the next two to three years as ways to illustrate how a tour-based model adds value to a project — perhaps through exploration of "What If" scenarios, ARC hopes also to observe the benefits of parallel model development.

Stakeholder Comments

ARC and the panel provided ARC’s partners the opportunity to comment on ARC’s model enhancements. Major challenges to future model improvements were also addressed. The Georgia Department of Transportation, Georgia Regional Transportation Authority, and Metropolitan Atlanta Rapid Transit Authority each provided remarks about the current state of ARC’s transportation model. Stakeholder feedback is summarized below.

Georgia Department of Transportation

The Georgia Department of Transportation (GDOT) and ARC have generally maintained an amicable relationship. Each agency has regarded open and clear communication as an asset towards the improvement of transportation planning strategies. GDOT has been particularly satisfied by ARC’s conversion from TRANPLAN to TP+, its refinement of the TAZ structure, and its transportation surveys. It was noted that these surveys and surveying techniques might be exportable to other parts of the state.

GDOT also commented on the challenges that, in its view, ARC faces. Concern was expressed as to whether ARC could continue increasing the size of its modeling area for air quality attainment. In order to increase the model area and carry along the detail required for reliable planning, ARC may have to determine how the boundaries of an urban model that continues to reach into rural areas are drawn. GDOT acknowledged that sharing the interstate system plan might help ARC in the determination of external stations. GDOT also expressed how difficult finding freight movement data and incorporating it into the process might continue to be.

Georgia Regional Transportation Authority

The Georgia Regional Transportation Authority (GRTA), an agency that works to improve Georgia’s mobility, air quality, and land use practices, is responsible for advocating and implementing a transportation system that is multi-modal and accessible to all citizens. Representatives from GRTA outlined what GRTA viewed as ARC’s modeling successes and challenges. Noteworthy achievements included:

GRTA noted that the challenge of incorporating heavy-duty truck data, including inventory and volume information, would be difficult. With limited data on freight movement, accurate calculation and calibration of emissions shares in the air chemistry model may be compromised. The mode choice model may also be affected by incomplete heavy truck data, as the public’s desire to use roads may be affected by the amount of truck congestion at a given time. In order to avoid these areas of potential model error, GRTA suggested that ARC continue to explore and develop methods to survey and appraise freight figures for the region.

GRTA also recommended that ARC consider bringing all partnering agencies "up to speed" together during the creation of 4-5 year plans. By doing so, more useful input concerning model assumptions, modifications, and results might be garnered. Continued stakeholder participation could potentially strengthen support for TIP funding requests, while helping ARC to determine meaningful measures of model reliability and cost-effectiveness. ARC might also consider incorporating signal timing into the regional model.

Metropolitan Atlanta Rapid Transit Authority

The Metropolitan Atlanta Rapid Transit Authority (MARTA) agreed with GDOT and GRTA on ARC’s notable enrichment of their transportation demand model. MARTA noted the following three specific successes of the model:

These successes do not come without accompanying challenges. MARTA questions the transit friendliness of TP+. With the new platform, it has been difficult to determine station access, time of day for transit use, and the overall dependability of station-to-station analyses. Trust and confidence in the new platform must be rebuilt through development and presentation of performance measures.

MARTA requested that ARC provide more explanation of and training in how the model treats transit and the associated assumptions within model runs. New technologies in the future should be implemented at times when partnering agency staff has time to learn alongside ARC staff, ensuring that the entire process and quality control elements are understood. The greater the involvement of collaborating agencies throughout the model development process, the easier buy-in becomes later on.

Summary of Panel Recommendations

The peer review panel provided an overview of model improvement accomplishments that ARC had made since the previous peer reviews. Each modeling expert on the panel also presented to ARC a list of successes and recommendations on how the various models currently used might be improved further. To conclude, the panelists offered a summary of "10 Next Steps" that ARC might consider during future model development and implementation. All of the comments and recommendations presented were results of panel consensus.

Overview of Accomplishments
Michael Morris

The panel commented that over the last eight years, ARC had made numerous positive steps and, in general, remarkable progress in the improvement of the Atlanta region’s transportation demand model. The 10 most significant successes include:

Demographic Forecast Models
Rita Walton

Demographic Forecast Model Successes:

Demographic Forecast Model Recommendations:

Travel Demand Models – Trip Generation, Trip Distribution, Mode Choice
Gordon Garry

Trip Generation Successes:

Trip Generation Recommendations:

Trip Distribution Successes:

Trip Distribution Recommendations:

Mode Choice Successes:

Mode Choice Recommendations:

Travel Demand Models Part 2 – Mode Assignments, Air Quality Interface, HPMS Adjustments
Jeff May

Mode Assignments Successes:

Mode Assignments Recommendations:

Air Quality Interface Recommendations:

HPMS Adjustment Recommendations:

Activity/Tour-Based Models
Keith Lawton

Tour-based Model Successes:

Tour-based Model Recommendations:

Next Steps and Summary

Michael Morris

This TMIP Peer Review was intended to provide feedback from a panel of experts to ARC and its consultants on enhancements made to the ARC travel demand model, which were based, in part, on prior peer review comments. ARC also sought review and recommendations from the panel regarding the development of an activity/tour-based model.

The panel presented ARC a summary of its 10 chief recommendations steps to take in the future. These next steps are listed below:

  1. Determine whether the work plan will continue to include development of an activity/tour-based model.
  2. The model’s current boundary should be increased. ARC would benefit from collecting more clear data on external people, because these people’s trips into the region are included in the model.
  3. Performance measures should be emphasized, especially reliability and accessibility measures.
  4. Continue to place attention on freight modeling.
  5. Continue to work with transit agencies to clarify and correct any data discrepancies.
  6. Develop an interim approach to 8-Hr Ozone Standards. Begin planning on a critical path on how to reach conformity within the allotted time frame. The panel suggests a Three Agenda Interagency Meeting to discuss 8-Hr Ozone Standards, the activity-based model, and HPMS Standards.
  7. Consider a three-tiered approach for demographic models.
  8. Increase zone detail. The panel recommends investigating how other metropolitan areas determine people per zone.
  9. Interface Population Synthesizer with DRAM/EMPAL as a way to begin integrating activity-based features into current modeling strategies.
  10. Maintain a strong technical foundation to continue becoming a leader in the field.

Appendix

Agenda

Tuesday, February 3, 2004
ARC Offices, C Level Amphitheater
Morning Session — Overview of Current Practice
8:30 - 8:45 Welcome and Introductions - Jane Hayse
8:45 - 9:00 Overview/Background - Jane Hayse
Synopsis of 2000 Peer Review - Jane Hayse
9:00 - 10:00 Current Modeling Practice
Travel Demand Model - Guy Rousseau
10:00 - 10:15 Break
10:15 - 11:15 Current Modeling Practice, Part 2
Emissions Model - Tracy Clymer
Travel Demand Model - Guy Rousseau
11:15 - 11:45 Questions from Panel
11:45 - 1:00 Lunch
Afternoon Session — Continuation of Presentations on Current Practice
1:00 - 2:00 Land Use Model
DRAM/EMPAL and IPEF - Bart Lewis
2:00 - 2:30 Questions from Panel
2:30 - 2:45 Break
2:45 - 4:00 Tour-Based and Activity-Based Model in Development Consultants
4:00 - 4:45 Questions / Answers Session - Working Session
Wednesday, February 4, 2004
ARC Offices, Harry West Room
8:30 - 9:30 Presentations by Other Government Officials
GRTA, GDOT, MARTA, EPA, EPD, FTA, USDOT
Land Use Modeling in Phoenix and Sacramento
9:30 - 10:00 Questions/Answers Session
10:00 - 10:15 Break
10:15 - 12:15 Panel convenes in closed session to prepare recommendations
12:15 - 1:30 Lunch
1:30 - 4:45 Interactive Dialogue and Questions/Answers Session
Conversations between the Panel and ARC
Thursday, February 5, 2004
ARC Offices, Harry West Room
9:00 - 10:30 Panelists present findings/recommendations
10:30 - 10:45 Break
10:45 - 12:00 Questions / Answers Session

Partial List of Participants

Presentations

Introduction and Four Step Modeling Process (HTML or PDF PDF files require the use of a PDF reader. You may download a free copy of Adobe Acrobat Reader to view these files. If you already have the Adobe Acrobat Reader installed on your computer it should automatically open to view the files when you click on the link to the document. )

Enhancements to ARC Traffic Model, 2000-2004 (HTML or PDF PDF files require the use of a PDF reader. You may download a free copy of Adobe Acrobat Reader to view these files. If you already have the Adobe Acrobat Reader installed on your computer it should automatically open to view the files when you click on the link to the document. )

Enhancement to the Model System & Statistical Analysis (HTML or PDF PDF files require the use of a PDF reader. You may download a free copy of Adobe Acrobat Reader to view these files. If you already have the Adobe Acrobat Reader installed on your computer it should automatically open to view the files when you click on the link to the document. )

Mobile Source Emissions Modeling Process (HTML or PDF PDF files require the use of a PDF reader. You may download a free copy of Adobe Acrobat Reader to view these files. If you already have the Adobe Acrobat Reader installed on your computer it should automatically open to view the files when you click on the link to the document. )

Population and Employment Forecasts (HTML or PDF PDF files require the use of a PDF reader. You may download a free copy of Adobe Acrobat Reader to view these files. If you already have the Adobe Acrobat Reader installed on your computer it should automatically open to view the files when you click on the link to the document. )