Back to search

University of South Florida

Algorithmic Design in Recommender Systems: Expanding DBE participation in the U.S. Construction Industry

Abstract

dc:description

The multifaceted U.S. construction environment, characterized by vast projects and intricate agreements, is anchored in trust, exemplified by bonding. For Disadvantaged Business Enterprises (DBEs), the bonding process can be challenging. This dissertation provides an in-depth evaluation of the "System for facilitating a project between contractors and owners," a patented methodology innovatively developed by my father, Dr. Dick Davis, Sr. and later approved through my research efforts, as indicated by the patent number US8346582B1. The system adeptly gathers and processes data from owners, projects, and contractors. Despite its proficiency in managing extensive contractor and project data, the system faces challenges in coherently rating contractors for specific projects without substantial human intervention. This is a critical function as it informs the recommendations given to project owners, who have the final say in contract awards. Ensuring accuracy and impartiality in these recommendations is vital to support an equitable landscape for Disadvantaged Business Enterprises (DBEs) without an increase in cost and better performance outcomes. Using Elaborated Action Design Research (Mullarkey & Hevner, 2019), the study sought to answer: How can data-driven algorithmic methods refine the patented system to enhance DBE policies and expand the contractor pool in bondable public-sector construction projects? A scenario and sensitivity analysis was employed, focusing on the diverse attributes and profiles of contractors. Initial findings highlighted ambiguities in the patented system, leading to the development of diagnostic instruments for clarity. The subsequent cycle identified myriad attributes impacting DBE participation, emphasizing the need for a dynamic, adaptable approach. The design cycle involved rigorous iteration, culminating in a modified Match Score Algorithm to expand DBE participation based on contractor profiles, attributes, and interventions. This algorithm was tested across various simulations, revealing its adaptability and effectiveness in ensuring an equitable contractor-project alignment process. The final cycles presented and evaluated the results, with visual tools and expert evaluations underscoring the algorithm's significance. Two hypotheses were central: The following two hypotheses were made in this study. Hypothesis 1: Merging two algorithms into an optimized algorithmic method will increase the size of the contractor pool in public-sector construction projects.Hypothesis 2: The optimized algorithm method will result in a more equitable distribution of bondable opportunities among a pool of DBE contractors in public-sector construction projects. The research outcome envisions a construction sector where DBEs, supported by unbiased algorithmic evaluations and forward-thinking policies, navigate the bonding process seamlessly, fostering a diverse, technologically advanced, and ethically sound industry.

Degree

thesis:*
Level thesis:degree_level
Dissertation
Discipline thesis:degree_discipline
Business Administration
Grantor
University of South Florida
Year dc:date
2023

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Davis, Dick C.E., II
Contributors dc:contributor
  • Matthew Mullarkey,, Ph.D.
  • Hemant Merchant, Ph.D.
  • Richard Tarpey, D.B.A.

Subjects

dc:subject × 5

Identifiers

dc:identifier.*
Repository record dc:identifier
https://digitalcommons.usf.edu/etd/10712

Chain of custody

source
Harvested from
University of South Florida
Base URL
digitalcommons.usf.edu/do/oai/
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Davis, Dick C.E., II. Algorithmic Design in Recommender Systems: Expanding DBE participation in the U.S. Construction Industry. Dissertation thesis, University of South Florida, 2023. https://digitalcommons.usf.edu/etd/10712