In the mid-1990s, NYMTC and its member agencies undertook the Transportation Models and Data Initiative (TMDI).
As part of this TMDI process, NYMTC and its member agencies implemented a preliminary model system in a very short timeframe in order to meet the basic requirements of ISTEA. As a result, the Interim Analysis Method (IAM) was first developed, and was used for two rounds of Air Quality SIP/TIP Conformity Analysis.
IAM was a relatively primitive planning tool. It met the needs at the time, and integrated the highway databases and analysis tools developed by NYMTC for the downstate New York region with the transit forecasting procedures developed by the Metropolitan Transportation Authority (MTA) in its Long Range Forecasting (LRF) project. Additionally, transportation networks and forecasts from northern New Jersey and Connecticut were incorporated to provide for the full representation of travel in the metropolitan region.
In a parallel effort, a detailed modeling requirements and data assessment effort was undertaken to establish the needs and specific modeling approach for the best practice models to be developed in the second phase of the project. This effort was the pre-cursor to NYBPM.
As the final phase of the NYMTC TMDI project, the NYBPM was developed and first previewed in front of a national audience on January 2001 during the TRB Conferences in Washington, and immediately was in wide implementation throughout the region on various projects of different sizes including:
The NYBPM is an activity-based model, which estimates the detailed travel patterns of a diverse population using numerous travel modes. It does this by introducing some innovative approaches to the traditional travel demand models including the concept of a journey or tour as the unit of travel, a micro-simulation approach to demand modeling, and the inclusion of non-motorized modes as an alternative to motorized travel.
The development of the NYBPM, and the input data, largely occurred in the 1990's. Data and networks were developed to represent a 1996 base year condition. At the time of completion and implementation in 2002, the data needed to be updated using the new socio-economic data based on the 2000 Census. In addition, adjustments needed to be made to the employment and labor force data to reflect the impact of the September 11th disaster.
Since the NYBPM was first developed and released in early 2002, NYMTC has updated its model every three to five years to better reflect the regional transportation system and have more accurate travel demand forecasts of future years.
The first update occurred in 2004-2005 and the NYBPM base year was changed from 1996 to 2002. In addition to updating all of the input data and recalibrating the model to the new base year, several NYBPM procedures were improved. Improvements included a major improvement in the external auto travel model, the inclusion of bus vehicle pre-loads in the highway assignment process, revisions to the multi-class assignment method, and the elimination of the adaptive highway assignment methods (i.e. adjustment of trip table to traffic counts). Improvements were also made to the post-processing procedures used with the NYBPM for air quality analysis and the model was implemented within the PPSuite platform.
The second major update occurred in 2008-2009 when the base year was changed from 2002 to 2005. In addition to updating all of the input data and recalibrating the model to the new base year, several NYBPM procedures were improved. Improvements included further refinements in the external auto travel model, correcting the taxi tour/trip generation, updates to bus vehicle pre-loads process for highway assignment, accounting for one-way tolls in tour level choices, and revisions to multi-class highway assignment. NYBPM was also improved by linking to a regional TIP project database, a new GUI was designed, and the production of a series of model-related output reports and maps was automated. Additionally, the feedback and convergence processes were improved and formalized. The model was improved and optimized to utilize distributed processing, which resulted in a substantial improvement to run-time (from 173 hours to 12 hours).
The NYBPM is used extensively by many agencies and consultants and there is an active model users group. The model is used for multi-modal long-range planning and programming, air quality analysis, and project level forecasting at a corridor scale.
NYMTC requested a TMIP peer review to continue to build its institutional capacity to develop and deliver travel-related information to support transportation and planning decisions. NYMTC has a keen interest in developing and improving travel modeling techniques that respond to the needs of the planning and environmental decision-making process; and developing mechanisms to ensure the quality of travel modeling results used to support decision-making, and to meet local, state, and Federal requirements.
The charge to the peer review panel was to identify modeling capabilities and functionalities required to meet the region's transportation analysis needs. NYMTC engaged the panel to help gain an understanding of what is needed to maintain its modeling systems and NYBPM's status as a state-of-practice model over the next 10-12 years.
