In 2011, a team composed of the City and County of Denver (CCD), the Regional Transportation District (RTD), and the Denver Regional Council of Governments (DRCOG) was awarded an Alternatives Analysis grant by the Federal Transit Administration (FTA), whose purpose was two-fold: to evaluate transportation alternatives in a ten mile corridor of Colfax Avenue (US-40) in Denver and Aurora, using DRCOG’s new activity-based travel model (called Focus); and to review/adjust/update Focus with the goal of rendering it acceptable to FTA for use in New Starts applications. Work on the Focus model update is underway, and this paper describes work already accomplished, and work underway and scheduled for completion in early 2013.

Modifications to Focus to be implemented under the FTA grant may be grouped into three categories:

• Model basics – these include improvements common also to trip-based models, such as transit path-building parameters, trip tables, highway speeds, etc.

• Issues unique to disaggregate models – examples include the larger number of tour/trip purposes, which makes parameter consistency more challenging between path-building and mode choice, and detailed time of day structure, which increases the ability/requirement to use more trip/tour characteristics in mode and time of day choice. A key goal of this project also is to move the modeling community toward a firmer conclusion on inputs to FTA's Summit software from activity/tour-based models when calculating user benefits, as well as approaches to holding trip or tour tables constant between scenarios.

• Further enhancements that may be applicable to both trip and tour/activity-based models – examples include better depiction of wait time for long headway transit service.

Re-estimation of mode choice models also will be explored, as new or re-formed variables may be identified during the project, or skims may be changed compared to those used for estimation of Focus. Current progress on this project includes a simplified approach to transit origin-destination assignment, modification of transit path-building to enhance consistency with mode choice and assignment, and preliminary conclusion (subject to further testing) that Summit inputs should be taken from the tour mode choice model. Further results will be available by Spring, 2013.