Association of Monterey Bay Area Governments (AMBAG) Travel Model Peer Review Report

5.0 Peer Review Panel Recommendations

Following the discussion of model enhancements, the peer review panel convened separately to discuss specific model development goals. Following this panel caucus, the panel presented a summary of their recommendations to AMBAG staff and other attendees at the peer review.

5.1 General Recommendations

The panel listed the many agencies who are current or potential users of the AMBAG regional travel demand model as well as model data (the MPO, 3 RTPAs, 2 transit districts, air quality district, 18 cities, 3 counties and Caltrans). The panel also listed the universities as another possible partner. AMBAG needs to leverage these partnerships realizing that all good information and analysis comes at a price. AMBAG already has budding relationships in place through the blueprint planning process and other planning efforts. The panel recommended that AMBAG continue to foster these relationships. The panel also suggested that the MPO could leverage existing grants, such as the grant to develop a bike model, to improve facets of the model. The panel emphasized continuing to explore additional grant funding sources. Even with these leveraged opportunities, the panel recommended that AMBAG needs to increase funding for model improvements in order to meet the aforementioned legislative and technical requirements for modeling in the region.

As part of building partnerships, the panel recommended developing a more open process for releasing the AMBAG regional travel model. To the extent that the full model goes out of office, user agreements documenting rules of governance of the use of the model can protect the agency and shift the liability to the user. Version control software could also be put in place to track official versions and to review and merge any improvements back to the main model development stream. The panel stated that an open process gives transparency and makes users accountable for their own work.

5.2 Short-Term Improvements

The panel recommended the following short-term improvements to the AMBAG regional travel model. Short-term was defined as items to be accomplished in preparation for the 2013 MTP. The panel acknowledged that this infers immediate action and that improvements would need to be done over the next year to be done in time for use in the plan. The panel also acknowledged that though the list looks long many of these items are small and would not require significant effort.

The short-term recommendations are ranked according to importance and are not in sequential order. Once AMBAG determines which improvements will be made, then they should proceed to implement the improvements in a logical sequential order.

1. Supplement 2011-2012 CHTS and oversample special populations and get a good demographic description of these households; do not be overly concerned of getting a 1% household sample, rather sample should be based on obtaining enough observations in a particular cross-classification of how the data will ultimately be used; have a focus to improve future mode choice and destination choice models realizing that mode choice will also need to be supported by a transit on-board survey; if possible try to use resources to get bike surveys or choice surveys
2. Fix free flow speeds on the highway network to ensure they represent true free flow conditions and not congested conditions (may also want to investigate using Highway Capacity Manual methods to calculate free flow speed)
3. Change to five time periods, recalculate time of day factors from survey and use period assignments to construct daily measures (do not use daily assignment)
4. Report production/attraction balance
5. Fix observed travel times by assigning survey origins/destinations to the network; use weighted average skim for trip distribution
6. Report screenline validation, in particular for Santa Clara County
7. Trip length frequency distribution needs to be revisited; report trip length frequency by county and don't prohibit K-factors
8. Check and report traffic assignment and increase convergence to 10E-6
9. For select link analysis use an assignment method that maintains proportionality
10. Test and report system convergence of distribution, mode choice, assignment feedback loop as it may take more than two iterations to reach convergence
11. Consider overall or multi-year trends when using validation data and do not be tied to a specific anomalous year
12. Report how trip purposes were combined for estimating time of day factors
13. Report at which level coefficients are applied in the mode choice nesting structure; report constants in terms of equivalent in vehicle travel time
14. Make the mode choice coefficients and transit skims weighting consistent
15. Recalibrate mode choice
16. Report and analyze screenline data
17. Smooth trip generation rates to provide a smoother transition between classes (optional)

5.3 Mid-Term Improvements

The panel recommended the following medium-term improvements to the AMBAG regional travel model. Medium-term was defined as items to be accomplished in preparation for the 2017 MTP. The improvements would cover approximately the next four years. As with the short-term recommendations, the mid-term recommendations are ranked according to importance and are not in sequential order. It is left to the MPO to determine which improvements will be made and then to put them in a sequential order.

1. Consider using statewide model for EE, IE, and EI travel; continue to maintain consistent inputs between the statewide and AMBAG regional models
2. Reduce bias constants in mode choice prior to use in new starts
3. Implement a hybrid activity/trip based model to better represent unique issues with full/part time workers and long distance travel (intra-community travel); could be implemented at a TAZ or parcel level:
   1. Use daily activity model instead of trip generation model
   2. Use primary destination choice model instead of trip distribution model
   3. implement an auto ownership model
   4. Leverage statewide model population synthesizer
   5. This will partially replace '5D' tool
4. If transferring models from another area make sure biases based on regional modeling techniques are not introduced
5. Replace current truck model with goods movement model
6. Develop mid-level land use forecasting model that recognizes seasonal residents and workers and that can take advantage of parcel data; give adequate time for testing before put results into production
7. Develop parcel data set through establishment of regional data consortium
8. Incorporate bike model
9. Work with transit providers to conduct on-board transit survey
10. Collect boarding data at same time
11. Transit agency collect data and AMBAG support process
12. Data collected with travel model in mind
13. Seek grant funding from FHWA, FTA, regional partner agencies or Caltrans
14. Continue to get regular updates of visitor data from tourist industry
15. Incorporate land use planning visualization tools to support continuing blue print planning efforts and SB 375
16. Estimate new mode choice model with new socioeconomic and demographics and that incorporates urban design and level of service characteristics (broaden explanatory variables to make more robust) for use in the '5D' analyses

5.4 Long-Term Improvements

The panel recommended the following long-term improvements to the AMBAG regional travel model. Long-term was defined as items to be accomplished after the 2017 MTP, or as resources became available. These were seen as lower priorities to the other improvements recommended.

1. Implement full activity based model.
2. Implement region wide dynamic traffic assignment (DTA)