Projects Need More Disciplined Oversight and Management to Address Key Challenges
GAO-09-436T: Published: Mar 5, 2009. Publicly Released: Mar 5, 2009.
This testimony discusses the National Aeronautics and Space Administration's (NASA) oversight and management of its major projects. As you know, in 1990, GAO designated NASA's contract management as high risk in view of persistent cost growth and schedule slippage in the majority of its major projects. Since that time, GAO's high-risk work has focused on identifying a number of causal factors, including antiquated financial management systems, poor cost estimating, and undefinitized contracts. Because cost growth and schedule delays persist, this area - now titled acquisition management because of the scope of issues that need to be resolved - remains high risk. To its credit, NASA has recently made a concerted effort to improve its acquisition management. In 2007, NASA developed a comprehensive plan to address systemic weaknesses related to how it manages its acquisitions. The plan specifically seeks to strengthen program/project management, increase accuracy in cost estimating, facilitate monitoring of contractor cost performance, improve agency wide business processes, and improve financial management. While we applaud these efforts our recent work has shown that NASA needs to pay more attention to effective project management. It needs to adopt best practices that focus on closing gaps in knowledge about requirements, technologies, funding, time and other resources before it makes commitments to large-scale programs. For instance, the Mars Science Laboratory, which was already over budget, recently announced a 2-year launch delay. Current estimates suggest that the price of this delay may be $400 million--which drives the current project life-cycle cost estimate to $2.3 billion; up from its initial confirmation estimate of $1.6 billion. Also, in just one year, the development costs of NASA's Glory mission increased by 54 percent, or almost $100 million, because of problems NASA's contractor is having developing a key sensor. Total project costs for another project, Kepler have increased almost another $100 million within 2 fiscal years because of similar issues. Taken together, these and other unanticipated cost increases hamper NASA's ability to fund new projects, continue existing ones, and pave the way to a post-shuttle space exploration environment. Given the constrained fiscal environment and pressure on discretionary spending it is critical that NASA get the most out of its investment dollars for its space systems. The agency is increasingly being asked to expand its portfolio to support important scientific missions including the study of climate change. Therefore, it is exceedingly important that these resources be managed as effectively and efficiently as possible for success. The recent launch failure of the Orbiting Carbon Observatory is an all-too-grim reminder of how much time, hard work, and resources can be for naught when a space project cannot execute its mission.
We assessed 18 projects in NASA's current portfolio. Four were in the "formulation" phase, a time when system concepts and technologies are still being explored and 14 were in the "implementation" phase, where system design is completed, scientific instruments are integrated, and a spacecraft is fabricated. When implementation begins, it is expected that project officials know enough about a project's requirements and what resources are necessary to meet those requirements that they can reliably predict the cost and schedule necessary to achieve its goals. Reaching this point requires investment. In some cases, projects that we reviewed spent 2 to 5 years and up to $100 million or more before being able to formally set cost and schedule estimates. Ten of the projects in our assessment for which we received data and that had entered the implementation phase experienced significant cost and/or schedule growth from their project baselines.3 Based on our analysis, development costs for projects in our review increased by an average of almost 13 percent from their baseline cost estimates--all in just two or three years--including one that went up more than 50 percent. It should be noted that a number of these projects had experienced considerably more cost growth before a baseline was established in response to statutory reporting requirements. Our analysis also shows that projects in our review had an average delay of 11 months to their launch dates. We found challenges in five areas that occurred throughout the various projects we reviewed that can contribute to project cost and schedule growth. These are not necessarily unique to NASA projects and many have been identified in other weapon and space systems that we have reviewed and have been prevalent in the agency for decades. (1) Technology maturity. Four of the 13 projects in our assessment for which we received data and that had entered the implementation phase did so without first maturing all critical technologies, that is they did not know that technologies central to the project's success could work as intended before beginning the process of fabricating the spacecraft. (2) Design stability. The majority of the projects in our assessment that held a critical design review did so without first achieving a stable design. If design stability is not achieved, but product development continues, costly re-designs to address changes to project requirements and unforeseen challenges can occur. (3) Complexity of heritage technology. More than half the projects in the implementation phase--eight of them--encountered challenges in integrating or modifying heritage technologies. (4) Contractor performance. Six of the seven projects that cited contractor performance as a challenge also experienced significant cost and/or schedule growth. (5) Development partner performance. Five of the thirteen projects we reviewed encountered challenges with a development partner. In these cases, the development partner could not meet its commitments to the project within planned timeframes. This may have been a result of problems within the specific development partner organization or as a result of problems faced by a contractor to that development partner.