We describe our experience teaching an overview course on urban modeling for graduate students at the University of California, Davis. It was decided at the outset that we would let the students run actual calibrated models in this class. This may be the first course where students were exposed to GIS models, as well as to spatial competition (economic) models. We conclude that the active learning format was successful in getting the students to understand the purposes of the various types of models and their limitations.
Effective urban planning and infrastructure investment rely on our ability to assess future needs today. Models are one instrument for obtaining projections of what future conditions will be. While models can be useful tools, there are barriers that discourage their use more broadly. One of the earliest criticisms of urban models was that they are too complicated for many people to readily understand (Lee, 1973). This produces the “black box” effect in which decision makers and citizens are left to trust the model outputs without sufficient explanation of how those outputs were generated.
While the use of large-scale urban models has increased in recent years (TMIP, 1998; Wegener, 1994), the primary user of any urban model is most often its creator. This situation may be changing. Evidence of the growing interest in these models is demonstrated by the number of model applications in the United States within the last few years (see Table 1).
These models can suffer from a number of problems, ranging from poor documentation, to complicated or non-existent user interfaces, to closed-code licensing.