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.
124 Traffic Forecasting in a Visioning Workshop Setting Don Hubbard, Fehr & Peers Associates Visioning workshops have become a vital tool inregional planning. Unfortunately, traffic forecastinghas played only a small role in these workshops even though traffic congestion is often viewed as a critical long- term issue. This creates the danger of a consensus forming in a work- shop around a vision that traffic modelers later declare is unworkable from a traffic standpoint. A disconnect of this kind can lead to one of two undesirable outcomes: 1) the agency abandons the consensus vision, in which case the workshop participants rightly wonder whether their views are being taken seriously, or 2) the agency is stuck trying to implement the unworkable. One way to ensure that the consensus forms around a workable vision is to perform traffic forecasts during the workshop and give participants immediate feedback as to the likely consequences of their plans, allowing them to adjust their plans accordingly. This approach was successful in visioning workshops sponsored by the Council of Governments for San Luis Obispo County, California (SLOCOG), and the Sacra- mento Area Council of Governments (SACOG). These are believed to be the first public workshops to forecast traffic in real time. The experiments used different mod- els and approaches, both of which provide important lessons for agencies that may want to play a role in visioning exercises. GENERAL REQUIREMENTS FOR WORKSHOP MODELS Travel demand models are typically designed to be used in a private, unhurried setting with ample opportunity to scrutinize inputs, analyze outputs, and, if necessary, per- form additional model runs. Models are usually designed to accommodate detailed changes to networks or model- ing parameters and to provide a rich assortment of potential outputs. In other words, their normal operat- ing environment is completely unlike a public workshop. A workshop model must produce sensible results within 15 min of receiving inputs from the participantsâ anything longer will make for unreasonably long work- shops and/or loss of interest by participants. Included in that 15 min is whatever processing is needed to compute key indicators and report the results, which may take the form of printed reports or figures projected on a screen, plus time the operator needs to analyze and interpret the results. There is not enough time to rerun the model if something goes wrong, so the inputs must be prepared correctly the first time. Moreover, the model must be robust enough to produce logical results for a wide range of input values, because it is difficult to predict what sort of proposals will arise during a public workshop. Fortunately, the outputs needed from a workshop model are much simpler than in a traditional model application. Public participants have neither the time nor the training to sort through long tables of subtle indica- tors; they prefer results expressed in a few easily under- stood numbers or figures. This greatly simplifies the modeling task because it allows a modeler to pick a few key indicators and then eliminate any model components that do not contribute to those outputs. For example, a workshop model might report the regional mode split but is unlikely to report patronage on individual transit