By David Bancroft, SDM ’02
Project Manager, MK54 Torpedo Program
Naval Undersea Warfare Center
May 22, 2002
Before I even graduated from MIT SDM in June 2002, my employer, the Naval Undersea Warfare Center (NUWC) had already received ROI on its investment in my education. This was due to findings from my thesis research, which focused on the loss of two developmental torpedoes during testing in the Bahamas. Each was worth over $1.5 million. More importantly, the accident nearly caused the cancellation of the $100 million program.
Before I go on, I’d like to take a step back and give an overview of my background. I graduated from University of Massachusetts, Dartmouth in 1982 with a BS in electrical engineering. Initially I worked for Sperry Systems in Oakland, California, but came back home in late 1983 to work at what was then called the Naval Underwater System Center (NUSC) in Newport, Rhode Island. For the past 18 ½ years, I’ve been working in Newport for the Navy, primarily on torpedo development. During the last four years, I’ve been project manager of the MK54 torpedo, which is the U.S. Navy’s response to the changes in the threat (post 1991 Soviet Union) and the acquisition process.
The MK54 combines existing hardware and software with commercial off-the-shelf (COTS) processors. The introduction of COTS into weapon systems provides the processing power needed to take advantage of new algorithms being developed by the science and technology community. The combination of reuse and COTS is allowing the MK54 to be developed for one-tenth of the budget of previous torpedo programs.
So if I had all of this experience and responsibility, why did I opt to take the SDM program? There were two reasons: First, the combination of engineering and management classes was the right fit for my work requirements. Although technically I am a project manager, I had never received any formal training. Second, you can’t beat MIT’s world-class reputation, faculty, and facilities.
The breadth of the SDM program is quite remarkable. I’ve been exposed to system thinking and architecture, marketing and management, and organizational and leadership disciplines. I’ve learned to look at things with a holistic view rather than the tunnel vision that typically becomes the norm, especially in government employees.
Several courses were especially useful, among them: Systems Architecture, which referenced the principles of holistic thinking, complexity, and pseudo-regulatory influences on the upstream process; and System and Project Management and Organizational Processes, two courses that explored the concepts of organizational structure and knowledge sharing/transfer.
As mentioned earlier, my thesis focused on a test failure that occurred last August when we lost two torpedoes off the coast of Andros Island in the Bahamas. Initial failure analysis resulted in a finding of “no fault found,” mostly because there was no physical evidence to examine. It’s been compared to trying to conduct an autopsy without the body!
A shortcoming in the engine-start sequence software was found that could potentially explain how the incident occurred, but extensive testing could not replicate the failure mode. We were left with several questions:
What caused the accident?
What could be done to prevent it from happening again?
Was the accident a result of a compressed schedule and a minimal budget – or poor management and communications?
My research explored why such a failure occurred in light of in-place system engineering and safety processes and used the concept of hierarchical causality to identify root causes. Hierarchical causality helped me examine the accident in a holistic manner, expanding the investigative space beyond the mechanism of failure and into the technical, human, organizational, and regulatory components of the U.S. Navy’s torpedo enterprise.
I found that prior program delays resulted in a development timetable with no slack and because of this, several “first-time” events occurred at launch time. Most significant was the first use of U.S. Navy helicopters with associated fire control systems and the fact that this was the first time the tests were conducted in waters deeper than the outer hull crush limit. The former may have caused the failure, while the latter disposed of any evidence.
The SDM curriculum helped me address all of these components and the tools I have learned enabled me to view the situation holistically and use my own hierarchical causality analysis of the test failure. I found that a number of root causes existed. These included:
Organizational differences in the priority given toward safety.
Independent and subsequent communication failures between system engineering and safety engineering. In fact, in our matrixed organization, the safety department was purposely kept separate and was located three buildings over in another organization!
The use of software to overcome a known hardware deficiency.
This last issue was the most damaging. The decision had been made to alter the software because it was easy to accomplish, so this approach became a short-term solution for potentially long-term problems.
Since completing my research, we’ve corrected the hardware deficiency that was “fixed” by software and are making inroads into changing the cultural bias towards safety and communications. The estimated ROI to NUWC is on the order of $15 million for this year and, as a result, is keeping the program moving forward.
One unexpected thing I’ve taken away from the SDM program is the number of new friendships I’ve made. The SDM program puts you in a team environment right from the start and you quickly learn to work with people you’ve never met before. Now, after completing the program, it seems as if I’ve known my classmates my whole life! I know I’ll keep in touch with many of them in the future.