The Gravity Probe B Launch Decision – Case Study
Gravity Probe B (GP-B) was a satellite based mission funded by NASA. Efforts were led by the Stanford University physics department and Lockheed Martin, Stanford University being the primary contractor and Lockheed Martin as the subcontractor. Rex Geveden, the Deputy Center Director at Marshall Space Flight Center, was the Program Manager at NASA and the primary leader of the GP-B launch mission. The mission objective was to test two of the predictions of Albert Einstein’s general theory of relativity: the geodetic effect and frame dragging. The GP-B experiment was one of the longest-running projects of NASA, with the funding received in 1964 to its final launch on April 20, 2004. Political pressures, poor risk management and hasty decision making were the primary reasons for the numerous conflicts in the GP-B launch. Fear of cancellation and financial losses influenced the teams at NASA, Stanford University and Lockheed Martin to oversee the intricate technical problems. Each of these organizations consisted of cross-functional teams that made independent decisions under different biases, and no co-ordination. In addition, the task group also experienced several technical anomalies with the Experimental Control Unit (ECU), a box on the spacecraft that housed a number of electronic components. All these reasons led to a build-up of issues till the very end. Proper management and collaborative thinking could have led to timely decisions, avoiding many of the problems that GP-B faced. Pressures and biases that influenced the stakeholders
The three major stakeholders in this case were NASA, and its two contractor teams from Stanford University and Lockheed Martin. Stanford University physics department, under the leadership of
Program Manager Gaylord Green, was assigned the primary contractor for the spacecraft by NASA. Lockheed Martin, under the leadership of Program Manager Bill Reeve, was the subcontractor that supplied the spacecraft and some components of the payload, and reported directly to Stanford. During the Flight Readiness Review at Vandenberg in July 2003, engineers identified a problem with the ECU. The ECU created significant signal interference in the Superconducting Quantum Interference Device (SQUID). ECU housed important electronic instruments and gauges, and also monitored the dewar, a critical piece of hardware that formed the main structure of the space vehicle. Fixing the ECU required a collaborative management decision involving all the major stakeholders mentioned above. Stanford’s Program Manager Gaylord Green was confident that the ECU would work on orbit for the necessary length of time and would not pose a risk to the mission. He based his statement on the extensive test program he implemented to check the spacecraft. However, he clearly portrays two significant biases. He bases his theory on his initial experiment results and fails to consider subsequent information that could influence his judgment, thus demonstrating an ‘anchoring’ bias. Also, he chose to disregard the concerns brought up by Stanford’s electronics manager Bill Bencze, which shows Green’s ‘confirmation’ bias to the situation. In some situations, Green also exhibits ‘overconfidence’ and ‘escalation to commitment’ biases, by suggesting to proceed with the launch with supreme confidence in its success. There was a primary reason that influenced Green. Fixing the ECU would require significant investment in terms of time, money and resources. Funding for this program had been cancelled numerous times, and was restored by the lobbying efforts of Dr. Francis Everitt, the Principal Director at Stanford University. Thus, there
was a need for somebody with Green’s power and position to come up with a solution to the problem quickly. This greatly influenced Green to suggest launching GP-B without fixing the ECU. On the other hand, Bill Reeve, Lockheed’s Program Manager for...
References: . http://www.hse.gov.uk/risk/fivesteps.htm
. John Stephens, 100812 Functionality in Teams [PowerPoint Slides]
Please join StudyMode to read the full document