DRAM/EMPAL projects the interactions and distribution of employment and housing in a specified geographic area. DRAM/EMPAL combines two spatial interaction models: the Disaggregated Residential Allocation Model (DRAM) and the Employment Allocation Model (EMPAL) to quantify the interactions between the metropolitan patterns of employment and population location and the networks of transportation facilities that connect them. DRAM/EMPAL provides a tool that relates future estimates of the location and type of employment in an area to their prior distributions, regional growth or decline, and the region's transportation system.
DRAM/EMPAL formed the two major components of an integrated set of computer models known as the Integrated Transportation and Land Use Package (ITLUP). Output from DRAM/EMPAL (i.e.,employment and household location and land use, trips generated for home-to-work, home-to-shop, and work-to-shop) were used with the third component of ITLUP to perform standard travel demand modeling (including submodels to estimate trip distribution, modal choice, and traffic assignment). DRAM/EMPAL currently is the most widely used employment, population, and land use forecasting application; it has been used internationally in more than 4 dozen metropolitan areas.
DRAM/EMPAL has been incorporated into a new system called METROPILUS, which combines employment and residence location and land consumption into a single, comprehensive package operating within an ArcView GIS environment.
General Info
Developers:
S.H. Putman and Associates, Inc.
Website:
dolphin.upenn.edu/~yongmin/intro.html
Strengths:
• Has been (and continues to be) used by numerous metropolitan areas and is a robust model.
• Has ability to introduce constraints or other influences, particularly to account for local knowledge.
• Input requirements use generally available data sources.
• Calibration of model is relatively easy.
Resources
User Input Requirements:
This model requires the following user input:
• Regional level for EMPAL: Target year values of total employment by economic sector.
• Regional level for DRAM: Total population, total person trips by purpose, percent unemployment by sector, employees per household by household type, matrix of households by income per employee by sector, jobs per employee, and net regional rate of employee commuting.
• Analysis zone level for EMPAL (base year): Household by type, employment by sector, total land area, land area occupied by basic and commercial employment, zone-to-zone travel times and/or costs.
Analysis zone level for DRAM (base year): Households by type, total population, total employed residents, group quarters population, land area use, land area occupied by basic and commercial employment, employment by sector, zone-to-zone travel times and/or costs.
Note: Data for DRAM/EMPAL are required both for the region, or overall modeling domain, and for each analysis zone. Input/output can be a number of different formats, including Excel or dbf.
Equipment Needs:
DRAM/EMPAL requires a MS Windows 95/98/NT operating system, a Pentium PC, color monitor, and printer. Spreadsheet, database, statistical, and ArcView geographic information system software are necessary for data management.
Limitations:
• DRAM/EMPAL modeling process is statistical, or aggregate theory based, rather than disaggregate or micro theory based; a reduced form of logit is used for location.
• Little or no scope to introduce planning policies other than land zoning, except by specific constraints or attractiveness functions.
• The absence of any mechanism for simulating the land market cleaning process underlying multi-year infrastructure change.
• The impact of zoning policies cannot be well represented in DRAM/EMPAL. Monetary and non-monetary incentives to guide land-use development cannot be represented in DRAM/EMPAL.
• Limited number of independent variables used to make projections may lead to underestimates of the full impact of some infrastructure improvements.
• The spatial resolution of the zones in DRAM/EMPAL is limited by a number of factors, the principal factor being availability of data.
• Sensitivity analyses are not possible.
• In order to achieve relative ease of use the model focuses on aggregate choice behavior rather than on individual choice behavior.
• The model requires training and experience to run correctly and efficiently. It is not an off-the-shelf product. It requires initial consultant involvement.
Staff Requirements:
Requires about one senior modeler with junior support.
Software Cost:
$30,000 to $60,000. Purchase cost includes consultant services, maintenance, and training costs.
