A NITC research project from Portland State University introduces a method of cleaning up land use data, for use in improved transportation models.
Transportation and land use are closely interdependent. Considerable work is underway, in Oregon and elsewhere, to develop models that integrate the two.
Planners creating these models often spend the bulk of their time preparing data on the various land uses. Many times the data, gathered from diverse sources, is incomplete and requires the planner to find missing information to fill in the gaps.
In fields outside of transportation, there have been considerable advances in techniques to do this. Data-mining and machine-learning techniques have been developed, for example, to systematically detect fraud in credit data, reconcile medical records and clean up information on the web.
In the transportation modeling community, by contrast, most efforts to tackle the problem are tied to a specific model system and a chosen study area. Few have produced reusable tools for processing land use data.