Since the creation of the Department of Homeland Security (DHS) in 2002, its components’ financial systems have been operating under legacy policies and disparate business processes because of outdated technology. As a result, these systems have been mostly non-integrated or non-interoperable with one another, and many components still rely on manual processes, which have led to inconsistent data and reporting. The United States Coast Guard (USCG) has been the financial service provider for three DHS components: USCG, the Transportation Security Administration (TSA), and the Countering Weapons of Mass Destruction Office (CWMD). These three components—DHS TRIO—exemplify the challenges posed by obsolete financial systems. A new financial management system is therefore needed to fully support Coast Guard financial and acquisition needs and comply with DHS and security requirements. Specifically, the Core Accounting System (CAS) had significant problems such as internal control weaknesses and an inability to produce accurate, reliable, and timely financial information and reports. Likewise, CAS lacked integration with feeder systems such as the Treasury Department, the Office of Management and Budget (OMB), and other USCG inventory and acquisition systems. It thus failed to meet some but not all requirements of the acquisition and financial management community of the USCG. Under these circumstances, the USCG needed to modernize its CAS. This thesis investigates how the USCG can achieve this goal.
The research for this thesis sought to answer the following questions. What can the United States Coast Guard do to meet the cost and schedule, but more importantly, to increase the likelihood of procuring a system that will meet its financial and operational needs or acquire the value that the agency has paid for? What are the program management processes that USCG needs to develop to make sure the transition to the new financial system can go smoothly?
This thesis first includes a comparative analysis of DHS and the US Coast Guard’s Financial System failure to determine whether this project failure is unique or shares characteristics and challenges with other large government IT projects. Toward this end, the first part of this thesis reviews the lessons learned from DHS TRIO components—USCG, TSA, and CWMD—with the Department of the Interior-Interior Business Center (DOI-IBC) and compares them with other government failures in IT acquisition projects such as HealthCare.gov; the USCG Integrated Health Information System (IHiS); and the Federal Bureau of Investigation’s (FBI) virtual case file (VCF). These IT projects share common challenges with the U.S. Coast Guard financial system for their relatively large government acquisition programs. Such projects bear inherent risks in terms of their size, interoperability, and integration with different government systems’ requirements. Research for this thesis reviewed and identified the similarities that contributed to these large IT acquisition failures. Second, based on the findings of the comparative analysis, this thesis provides recommendations related to both project management and business processes that DHS should apply to successfully transition the USCG and TSA to the new financial management system and the rest of their financial management modernization.
Research for this thesis examined the scholarly debates on the failure of government IT projects. Specifically, it analyzed the scholarly views on causes, risks, and ways to mitigate IT failures. To this end, it reviewed a variety of academic peer-reviewed papers from the Institute of Electrical and Electronics Engineers (IEEE), Journal of Scientific Research, Science Direct, studies from the Standish Group, Government Accountability Office (GAO) reports, government agencies acquisition lifecycle processes and procedures, and articles from Computerworld and Govtech.
This research found that large government IT projects are often categorized as high-risk and likely to fail. They are complex and take longer than two years to develop. Government IT projects are usually built to connect to multiple agencies. These government IT projects affect a wide range of agencies’ operational missions and the different type of services that these agencies provided to the public. The four government IT projects this thesis analyzed—DHS TRIO, USCG IHiS, HealthCare.gov, and FBI VCF—are categorized as large projects. The challenges revealed in the case studies in this thesis are not unique. The initial failures and challenges of the case studies are not isolated cases, and these issues have affected other government projects as well. The case studies and lessons learned in this thesis may contribute to more effective practices that can be used in future big government IT projects.
The four case studies all experienced schedule delays and cost increases. The DHS TRIO and HealthCare.gov projects required significant rework while IHiS and FBI VCF were cancelled and restarted as new projects. The DHS TRIO project costs increased by 54 percent from the original estimate and delayed the delivery schedule by more than two years. The USCG IHiS spent over $56 million; after five years it was cancelled and restarted as a completely new project named DOD MHS GENESIS. The HealthCare.gov cost increased from the original estimate of $292 million dollars to $2.1billion dollars. The FBI had wasted $105 million by the time the VCF project was cancelled and restarted as a new project in 2005.
The four large government IT projects, DHS TRIO, USCG IHiS, the HealthCare.gov, and FBI VCF, shared similar results. They all experienced significant schedule delays and cost increases, which inevitably led them to fail. The major factors that contributed to the four projects’ failures were 1) not defining the project outcomes at the beginning of the Acquisition Lifecycle, 2) lacking the right expertise, and 3) having weaknesses in leadership throughout the business process change and the internal control procedures. The projects’ challenges and shortcomings led to lessons learned in these four areas for future government and private sector large IT projects.
The four case studies that this thesis reviewed revealed a few important elements that contribute to successful projects, especially government IT projects: defining the project outcomes at the beginning, having the right expertise, leading the organization through the business process change, and fostering internal control procedures.
 CAS is the current financial management system for USCG, as well as for the Transportation Security Administration (TSA) and Domestic Nuclear Detection Office which became the Countering Weapons of Mass Destruction office (CWMD) in 2017. It is a highly customized version of Oracle Federal Financials Release ® 11.5.10.
 The Standish Group, CHAOS Report: 21st Anniversary Edition (West Yarmouth, MA: The Standish Group International, Inc., 2014), https://www.standishgroup.com/sample_research_files/CHAOSReport2014.pdf. According to the 2014 Standish report, large IT projects have budgets over $10 million and take over two years to develop, which was the case of all four case studies.
 Asif A. Khan, DHS Financial Management: Improved Use of Best Practices Could Help Manage System Modernization Project Risks, GAO-17-803T (Washington, DC: Government Accountability Office, 2017), https://www.gao.gov/assets/690/687359.pdf.
 David A. Powner, Coast Guard Health Records: Timely Acquisition of New System Is Critical to Overcoming Challenges with Paper Process, GAO-18-59 (Washington, DC: Government Accountability Office, 2018), https://www.gao.gov/assets/690/689565.pdf; “MHS GENESIS,” Military Health System, accessed April 9, 2021, https://www.health.mil/Military-Health-Topics/Technology/Federal-Electronic-Health-Record-Modernization/MHS-GENESIS.
 Alex Wayne, “Obamacare Website Costs Exceed $2 Billion, Study Finds,” Bloomberg, September 24, 2014, https://www.bloomberg.com/news/articles/2014-09-24/obamacare-website-costs-exceed-2-billion-study-finds.
 Harry Goldstein, “Who Killed the Virtual Case File? [Case Management Software],” IEEE Spectrum 42, no. 9 (September 2005): 24–35, https://doi.org/10.1109/MSPEC.2005.1502526.