Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
NCHRP, a consortium of metropolitan planning organi- zations and state departments of transportation, or pos- sibly software firms take the lead in helping bring a central focus to model development, testing, and deploy- ment. Practitioners, academics, and consultants all need to be involved in the process. I think the lack of stochastic models is also an obsta- cle, as is the need for techniques to conduct risk analyses. At the Puget Sound Regional Council, we are just begin- ning a 1-year testing phase of the UrbanSim land use model. We have an enhanced four- step model that is being used to test pricing and high- occupancy toll lane alternatives. It appears to be working well. We are also conducting a new 2-day travel diary survey. We may try microsimulation tour- based modeling in 2 or 3 years. Along with the 2-day travel survey, we have a major problem of converting all our data from the standard industrial classification system to the North American Industrial Classification System. We are compiling a 2006 base year for UrbanSim and we will recalibrate our travel model to 2006. We would also like to add a regionwide dynamic traffic assignment model or a sketch accessibility calculator, if these tools are developed. Ronald Milone Thank you for inviting me to offer comments on whatit will take to bring advanced travel modeling meth- ods into practice. This is certainly an important and timely conference, and TRB is to be commended for bringing us all together. Both the practitioners and the research com- munity have a great deal to gain by working more closely together in the coming years. Earlier, Frank Koppelman made the interesting comment that many of the groups involved in the planning process are increasingly talking past each other and not really listening to each other. I think the same phenomenon has been occurring between practitioners and researchers. This conference has helped to close the gap between research and practice, and that is needed for progress in the travel forecasting field. The Metropolitan Washington Council of Govern- ments Transportation Planning Board (TPB) employs a conventional four- step travel demand model that is typi- cal of such models applied at most metropolitan plan- ning organizations (MPOs) around the country. We are working hard to address a wide variety of complex trans- portation issues in the capital region, and we are increas- ingly pushing the travel model to go beyond what it was originally designed to do. In addition, our modeling process has received a great deal of technical scrutiny, particularly by environmental groups. As a result, much time has been spent examining and refining our methods used in application. The motivation of this conference is to facilitate the implementation of advanced methods into practice. Yet, there have been many indications from conference speakers that existing practices are inadequate. Bill Woodfordâs session on FTA- sponsored research touched on problems associated with existing traffic assignment procedures. Martin Wachsâs presentation on the results of a recent state- of- the- practice survey indicated that many MPOs are behind the curve with respect to rea- sonable modeling practice. How can we realistically bring advanced methods into practice when we are not doing an adequate job with existing methods? The TPB adheres to a multitrack model development process. We have an applications track, which focuses on short- term improvements to existing methods, and we also have a methods development track, which focuses on the development of more advanced models and tech- niques. The methods track activities proceed in parallel with the applications track so that advanced methods may be phased in when deemed appropriate. If resources are available, I would strongly encourage MPOs to adopt a similar approach. We feel the âdual trackâ approach is reasonable for incrementally phasing improvements into production when the improvements are ready. We also believe in the development of empiri- cally based models and the use of locally collected data. One obstacle to moving advanced methods into prac- tice is the grueling production schedule faced by many MPOs, which includes periodic plan updates, air quality state implementation plan and conformity work, and project planning studies. Given this busy schedule, I feel that there must be a firm understanding about the amount of time and resources that will be required to develop new methods. In order to secure time and resources, our staff and board also need to understand how the advanced methods are superior to existing methods. Another obstacle to moving forward is the lack of observed data. Several speakers highlighted both the lack of data and the limitations of existing data used in model development work. It is important that the data needed to calibrate and validate advanced methods are clearly understood. I would submit that understanding observed data and input data is just as important as understanding the model itself. There was a great deal of discussion at the conference about what makes one model âbetterâ than another model. I do not think that is a useful dis- cussion. I would submit that the âbestâ model is the model that is calibrated with quality data and is driven by quality inputs: accurate land use forecasts, accurate network coding, and accurate policy assumptions. Finally, the advanced methods need to be fully trans- parent to practitioners before they are used in applica- tion. Practitioners presently understand the four- step model very well. They know how to develop four- step models and how to interpret results. They understand the limitations of the four- step approach. As advanced 64 INNOVATIONS IN TRAVEL DEMAND MODELING, VOLUME 1
