Skip to main content
Decision Modeling for Housing and Community Development: A Methodology for Evidence-Based Urban and Regional Planning
56th Annual Conference of the Association of Collegiate Schools of Planning (2016)
  • Michael P Johnson, Jr.
Urban community development corporations and other local institutions routinely face challenging problems in housing and economic development that require substantial expertise in data analytics and decision modeling. While CDC employees have significant experience in diverse application areas, they often face limitations in acquiring, analyzing and sharing data, and using these data to solve decision problems whose solutions can generate novel strategies to address localized needs. Recent research, inspired by local responses to
the housing foreclosure crisis, and developed in cooperation with Boston‐area CDCs, has resulted in a collection of analytic methods and applications that can assist CDCs and similar organizations to design local interventions and initiatives for acquisition and redevelopment of housing in various stages of foreclosure. These interventions are intended to make best use of human and financial resources and to reflect diverse stakeholder values, quantify important success metrics and generate policy and planning insights for local action. In this
talk I will highlight a collection of these modeling applications and explain how they may be useful to local organizations addressing a range of contemporary issues in housing and community development.

The first of these applications is values and objective design, i.e. the process of identifying decision opportunities, as distinct from problems that an organization may feel constrained to solve, or which may be presented to it. Values design can also help identify performance metrics, or attributes, which allow an organization to assess progress towards meeting goals.

The second of these applications is data analytics, or the process of identifying, collecting and analyzing data to generate management, policy or planning insight. In this case, data analytics can help CDCs identify areas of interest for interventions, and select among a range of potential activities based on local area characteristics. Data analytics can also help CDCs quantify important but ambiguous notions that are closely tied to local development success, such as ‘strategic value’ or ‘property value impact”.

The third of these applications is decision modeling, or the process of formulating and solving problems that capture important aspects of operations and strategy design in order to generate insights regarding specific courses of action to take that represent improvements over the status quo. Decision modeling applications we will discuss include selection of candidate foreclosed housing units for acquisition and redevelopment, designing bidding strategies for units to acquire in a competitive local environment, and managing a portfolio of properties for development over multiple time periods and neighborhoods.

These applications of data analytics and decision modeling are not limited to foreclosed housing, however. I will demonstrate how empirical problem‐solving, incorporating elements of management science and analytics, and inspired by the ‘big data’ and ‘smart city’ movements, have the potential to generate valuable insights in multiple areas of housing and community development: vacant land management, gentrification and displacement,
acquiring and redeveloping blighted structures and affordable housing development.
  • decision modeling,
  • foreclosed housing,
  • community development,
  • data analytics,
  • smart cities,
  • big data
Publication Date
November 3, 2016
Portland, OR
Citation Information
Michael P Johnson. "Decision Modeling for Housing and Community Development: A Methodology for Evidence-Based Urban and Regional Planning" 56th Annual Conference of the Association of Collegiate Schools of Planning (2016)
Available at: