Improvements to the projective land use model for producing small area forecasts.

In forecasting regional growth in the San Diego area the Projective Land Use Model (PLUM) is one of 3 subregional models employed by the San Diego Association of Governments. Incorporating information on the future location of basic employment from the Basic Employment Allocation Model PLUM allocates population housing units and local serving employment to subareas based upon the location of basic employment availability of usable land and transportation accessibility. PLUMs current zonal system includes 161 subareas combinations of census tracts overlaid by incorporated city city sphere of influence and community plan area boundaries. The models logic and structure are described and recent modifications to PLUM which have improved its forecasting accuracy for population housing and employment in subcounty geographic areas are discussed. In general PLUM is a derivative of the Lowry gravity model which allocates a regional growth increment for the time period between a base year and a target year. The model functions in 3 main allocation stages which are described. To correct inaccurate forecasts in some subareas along with other minor problems PLUMs algorithms were modified and its software completely rewritten. Changes were made in the areas of approved residential and commercial projects vacant land uses infill and redevelopment densities and error and warning messages. The revised PLUM better replicates land use and density policies more accurately forecasts housing unit structure type mix and better assesses the distribution of future growth between vacant and already developed land.