The Travel Demand Model used for Project Traffic Development should be evaluated to determine its accuracy at both the regional and project levels. In many cases, additional validation work will be needed within the AOI. The validation process should include a review of all available land use, socioeconomic, and transportation network data to be used in the model. The District Planning Office should approve all data inputs used in the validation process, and the validation effort must be completely documented and approved prior to its use. This section discusses the general approach which should be followed to properly validate a subarea of the model for a project (site specific) analysis.

Evaluation of Base Year Conditions

The selected model should be run using base year data to evaluate its ability to accurately replicate base year ground conditions, both regionwide and within the study area. Both the socioeconomic data and transportation network, as well as the traffic counts, should be checked for accuracy and timeliness.

4.6.1.1 Base Year Land Use

The Base Year Land Use Data should be evaluated within the project AOI for its accuracy and consistency with local comprehensive plans. Local planning agencies and MPO/TPO/TPAs should be contacted to verify the land use within the project. All existing TAZs should be analyzed in terms of their size and the number of trips or activities generated. In some cases, it may be necessary to refine the existing TAZ structure to achieve a better trip assignment. Special care must be taken when coding new centroid connectors to properly represent realistic loading locations.

4.6.1.2 Base Year Network Data

The Base Year Model Network within the project AOI should be checked for connectivity, directionality, and turn penalties to make sure all vehicle movements are properly represented. Additional roadways may need to be coded into the network to provide better loading points for newly created TAZs, and to allow for an improved path building process. The roadway attributes should be checked regarding area type, facility type, number of lanes, and free flow speeds.

4.6.1.3 Base Year Traffic Counts

An analysis should be conducted to identify whether sufficient coverage counts are available within the project AOI. If critical links are missing counts, then additional counts should be obtained. If any roadways have been added to the network, the availability of counts should be checked for these added roadways. An analysis should be conducted to add screenlines, which might require additional counts, within the project AOI, to create the ability to quickly analyze the accuracy of the distribution patterns. These additional counts would have to be adjusted to the base year of the study, as well as to the units the model uses (AADT or PSWADT). Note that this may be a costly endeavor, and not always feasible or desirable.

Base Year Model Refinement

The commonly used model refinements include the following:

Once all refinements have been completed, the entire model should be rerun. An analysis should first be conducted on the entire model to ensure that the refinements in the project AOI did not negatively impact the overall model validation. When it has been established that the entire model operates on the same level of accuracy, or perhaps at an improved level, the project AOI should be analyzed on its accuracy (see Table 4-1 to Table 4-3 for standards) and its size. If significant changes occur outside the preliminary project AOI, determine whether changes to the project AOI are required. Based on this analysis, it should be determined if the project should be expanded to include the affected facilities and if other development mitigation infrastructure improvements are required.

Expansion of the project AOI may also require re-examination of the base year model volumes with the base year counts throughout the expanded project. If the project model evaluation is not acceptable through the entire expanded project AOI, it may be required to make further base year model refinements to achieve acceptable volumes and repeat Travel Demand Forecasting. Close coordination should take place with the District Planning Office to reach an acceptable level of accuracy. Expanding the project AOI will have impact on schedule and budget.

Evaluation of Future Year Conditions

After the Base Year Model Validation is approved, and appropriate validation refinements and future land use data revisions have been incorporated into the forecast year model(s), the model is ready to determine future year traffic forecasts for resurfacing projects. If the model is used for corridor or project analysis, additional validation procedures may need to be executed.

To develop project traffic for a given year, appropriate future year data inputs are required. For each of the future analysis years, the following model inputs should be summarized:

Chapter 4 Forecasting with Travel Demand Models

Each of these data items should be updated to reflect the approved elements of the MPO/TPO/TPA cost feasible Long Range Transportation Plan, master plans, and planned development mitigation infrastructure improvements anticipated to be in place in each analysis year.

Consistency with the Adopted LRTPs and LGCPs

There are three (3) steps that need to be performed to verify the project consistency with the MPO/TPO/TPAs’ Long Range Transportation Plan (LRTP) or a Local Government’s Comprehensive Plan (LGCP): Consistency with the Plan(s), Plan Amendment/Alternative, and Inconsistency Documentation/No Project.

4.6.4.1 Consistency with the Plan(s)

The number of lanes needed to accommodate future travel demands shall be compared with the existing MPO/TPO/TPA Long Range Transportation Plans in metropolitan areas and local government comprehensive plans and plan amendments. If the project is not consistent with the approved plans, go to the Plan Amendment/Alternative.

4.6.4.2 Plan Amendment/Alternative

If the Corridor Traffic Forecast results are inconsistent with the LRTP and/or LGCP, or a plan approved by FDOT, the proposed transportation alternatives (such as public transportation alternatives or parallel routes) need to be reexamined. If this analysis does not resolve the inconsistency issue, requests need to be made to the appropriate District Director or their designee(s) to modify either the existing FDOT plans (such as Action or Master Plans) or initiate the process to request the local government to amend the LGCP or the MPO/TPO/TPA to revise its LRTP. In any event, the party that requested the corridor study should be notified of the inconsistency and be involved in the decision to remedy it. If alternative transportation improvements are to be tested, redo the project traffic forecast process and perform calculations for the new alternative. If the local government and/or the MPO/TPO/TPA or the FDOT does amend or revise the applicable plans, prepare the necessary forecast. If the local government and/or the MPO/TPO/TPA or the FDOT does not amend or revise applicable plans, go through the steps as described in Section 4.6.4.3.

4.6.4.3 Inconsistency Documentation/No Project

If the appropriate District Director or his/her designee(s) approves the project due to extenuating circumstances, include a statement in the Corridor or adopted plan. State in the report the process that was used in Section 4.6.4.2 and the decisions made. Include in the document any written letters or agreements generated as part of the activities in Section 4.6.4.2. If the project is not viable, indicate in the conclusion of the report that the study resulted in a “No Project.”

Reasonableness Checks for Future Years

Future year traffic volumes cannot be validated against existing traffic counts. The model output must be checked and certified. The modeled volume changes for each year of analysis and for each alternative network should be evaluated against the expected changes. Although expected changes cannot be accurately quantified, approximate changes should be estimated. For example, if the region’s growth is expected to continue, freeway volumes should increase with some relationship to the trend. The average percent of change between years should be relatively constant unless some special factors affect the growth, such as roadway improvements along parallel facilities.

The model-generated volumes for the future years should be reviewed for logical traffic growth rates. The general growth trends prevalent in the area should be determined and compared with the modeled traffic volumes. The future year model volumes should be compared against the appropriate historical count data. If an unexplained growth rate exists, a thorough review of the base and future year land use, socioeconomic data, and network coding should be performed.

Logical reasons for any anomalies should be documented. A careful comparison is required, especially for urbanized areas where growth may be higher along undeveloped corridors, while on an area-wide basis it may be much lower.

Acceptable Model Refinements for Future Years

Models frequently provide insights into traffic route selection that may not be readily apparent. However, where model results do not appear to be reasonable, the deviations must either be explained or acceptable revisions to the network, land use, or socioeconomic data need to be made. If the model results are not reasonable and cannot be corrected, then use the historical traffic forecasting processes described in Chapter 5.