Contenu téléchargeable
Télécharger le fichier PDFMarshall, Norman L. with Stephan J. C. Lawe, "A Comparison of Regional Models Using Different Levels of Geographic Aggregations", 1994
- 1994
- Processing documentation available at: https://wiki.albany.edu/display/SCA/Processing+Ingested+Digital+Files
- Contact Us for questions or additional rights.
We have developed several models linking land use changes and transportation for different regions with populations ranging from 100,000 to 2,000,000. The models are uses to long-term planning for land use, transportation, and air quality. The level of detail desired by clients requires that the models have a high degree of spatial disaggregation, typically several hundred zones. The spatial detail is a strength of the models, but it is achieved at cost. Model development is time-consuming and expensive. Even on the fastest microcomputers, model run time is measured in hours. Testing new models structures and making structural enhancement is slow and difficult. The complexity of the models inhibits experimentation and learning by clients. Simpler models are needed to aid in prototype development and in communication and training. We compared results of policy simulations from regional models with three levels of spatial disaggregation. The most detailed models are as described above. For a highly-aggregated example, we developed a system dynamics model with two concentric rings using STELLA software. Finally we evaluate two simplified grid models with intermediate levels of spatial disaggregation models. The STELLA model is able to demonstrate some of the policy results, and is an excellent tool for model development. In some ways the grid model combine the worst of both worlds. They are neither simple nor fully detailed. The detailed models exhibit the most rich behaviors. We are developing techniques to move from STELLA structures directly to the more disaggregated model
This is the whole item.
ITEM CONTEXT
- This collection is part of the University Archives