An Empirical Analysis of DoD Construction Task Order Performance

Abstract

Cost and schedule overrun plague over 50 percent of all construction projects, engendering diminished available funding that leads to deferred maintenance and impaired award ability for needed projects. Though existing research attempts to identify overruns sources, the results are inconclusive and frequently differ. Accordingly, this research reviews DoD construction contract data from the past ten years to identify the contract attributes of 79,894 projects that correlate with superior performance for use in future project execution. This research starts with creating a database that houses the largest single source of construction contract information. The research then evaluates the data to determine if differences in project performance exist when comparing contracting agents, funding agents, and award months. Next, the research utilizes stepwise logistic regression to determine the significant contract attributes and predict future projects overrun likelihoods. Model accuracy for predicting the likelihood of cost and schedule overrun is 65 percent and 75 percent, respectively. Finally, this research concludes by providing insights into efforts that could improve modeling accuracies, thereby informing better risk management practices. This research is expected to support public and private sector planners in their ongoing efforts to execute construction projects more cost-effectively and better utilize requested funds.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 25, 2021
Accession Number
AD1149112

Entities

People

  • Adam B. Teston
  • Tyler S. Stout

Organizations

  • Air Force Institute of Technology

Tags

Communities of Interest

  • Energy and Power Technologies
  • Human Systems

DTIC Thesaurus Topics

  • Air Force
  • Business Administration
  • Civil Engineering
  • Computational Science
  • Congress
  • Construction
  • Contracts
  • Data Mining
  • Databases
  • Department Of Defense
  • Engineering
  • Engineers
  • Governments
  • Information Science
  • Law
  • Machine Learning
  • Management Personnel
  • National Governments
  • Neural Networks
  • Organizational Structure
  • Statistical Analysis
  • United States
  • United States Government

Readers

  • Government Contracting/Procurement.
  • Life Cycle Cost Analysis
  • Regression Analysis.