NATIONAL AERONAUTICS AND SPACE ADMINISTRATION
Washington, D.C. 20546
REPLY TO ATTN OF: MT September 22, 1975
In your letter of September 15, 1975, you invited my thoughts as to what were several of the most important managerial or technical lessons learned from Skylab about space flight development and operations programs. My thoughts on this follow: For the most part, these are not new lessons, but lessons learned in different circumstances and surroundings. The fact that they are not new is of course a lesson in itself - we must continually strive to benefit from past experiences and structure our management so that past related experience can be brought to bear on current problems.
First, I would point to a lesson from our meteoroid shield failure:
That lesson relates to the importance of interdisciplinary communication among designers. The shield failure was traced to an "aerodynamic sneak circuit" wherein the structure design and fabrication inadvertently provided a path for air from a high pressure region on the vehicle to travel to a critical load area. The aerodynamicists, load engineers, and structural designers had not adequately cross-communicated in their reviews of the integrated system during design and during reviews of the completed hardware after fabrication. The lesson is the continued necessity for integrated cross-discipline reviews of designs and hardware led by an overall system oriented "chief engineer."
On the subject of the meteoroid shield, another lesson comes to mind and that is the need for continued hard review of design requirements. In retrospect, the requirement that led to the provision of a meteoroid shield was questionable. The shield was required in order to meet the arbitrary numerical design goal with the limited environmental knowledge then existing. Certainly with the benefit of hindsight, however, the shield was not necessary.
A second lesson "relearned" relates to the importance of judicious design margins and provision for contingencies. The Skylab electrical power system had a substantial margin in power available for basic housekeeping and experiment requirements. As a result of this margin, it was possible to operate the Skylab, with judicious power management, at full effectiveness in spite of loss of one complete solar wing following the meteoroid shield failure. Had we not had these power margins in the basic system design, the loss of the one solar wing would have greatly reduced or completely negated our mission capabilities.
Conversely, it is important to plan for maximum utilization of contingent capability. This was illustrated by Skylab weight management. Early in Skylab, a weight design margin and contingency of about 30 percent was provided. As the system matured, it became evident that the experiment complement and expendables provisions could be substantially augmented within the limit weights. This was done and as a result, mission duration and content of the experiment program were increased substantially at no increase in program cost. Specifically, it was possible to increase the final mission duration by 50 percent.
The electrical power systems on the ATM and the OWS, although both solar cell-battery systems, were quite different in their implementation. Failures occurred in both systems, but the failures were complementary to each other so that total system capability was not reduced to the extent it would have been with identical systems of either design. conversely, the risks of identical redundancy were evident in the multiple failures of the control gyros for the attitude control system. Had it not been possible for the crew to repair the gyro problems in flight, the mission would have been prematurely terminated.
I hope these thoughts will be helpful to you; I believe your treatise will be a most valuable addition to the art and science of complex technical program management.
John H. Disher
Home - NASA Office of Logic Design
Last Revised: February 03, 2010
Digital Engineering Institute
Web Grunt: Richard Katz