Sheas Systems Approach

As the AGC hardware evolved, so did NASA's management of the project. In early 1963, Brainerd Holmes, who headed Apollo at NASA headquarters, left the agency after a disagreement with Webb. That July, Webb consolidated management at the agency and hired a new man to bring coherence and control to the program. George Mueller came to NASA from air force systems engineering contractor TRW, taking over Holmes's job as NASA deputy associate administrator for manned spaceflight. Mueller had classic systems-engineering credentials: a Ph.D. in physics, time at Bell Labs, and experience in adopting systems-management techniques while working on the Minuteman missile for TRW. ''One of the things that became apparent immediately,'' Mueller recalled about taking over the Apollo program, ''was that there wasn't any management system in existence.''

The statistical naysayers supported his position. Mueller received a pessimistic report that under the current program, the chances of landing on the moon were one in ten. The probability of failure provided the justification Mueller needed as he quickly moved to consolidate control of Apollo in NASA headquarters and to better coordinate among the NASA centers at Cape Canaveral, Houston, and Huntsville. He also created ASPO, which had dispersed sites but was part of the headquarters structure. ''The laboratories were going to have to become a support to the program offices or else we weren't going to get there from here,'' Mueller believed.29 One result of Mueller's changes was the famous ''all up'' testing approach, wherein the entire Apollo system was tested as a unit, rather than the more conservative, step-by-step approach favored by the NASA centers, Huntsville in particular. He also hired Bellcomm and General Electric to provide technical advice from outside the centers.

Mueller also brought in General Sam Philips, who had directed the U.S. Air Force's Minuteman ICBM project. Philips came with a number of his managers from the air force, and together they imposed their systematic management methods on a NASA organization used to the freewheeling style of research. The Space Task Group was already straining under the demands of Mercury and Gemini (which were both running late and over budget). Apollo, while still progressing quietly in the background in the early 1960s, was beginning to show strains. Philips employed techniques like configuration control and interface control, formal methods of documenting the hardware and their connections.30 While these were appropriate for a large organization, the craft-oriented engineers sometimes interpreted them as simply layers of bureaucracy. ''I've been identified as a procedures and methods man,'' Philips lamented, ''with a management manual all written that I intend to force on the Apollo program and down the throats of the existing 'good people' base.''31

Putting these controversial methods to work on the ground was Joe Shea, a tough, focused man, not always diplomatic, but deeply committed to Apollo. Like Mueller, he represented the new kind of engineer that had emerged during the ballistic missile projects of the 1950s: the systems expert. He had bachelor's, master's, and Ph.D. degrees from the University of Michigan in engineering mechanics, and had spent much of the 1950s as a mathematical analyst and development engineer at Bell Labs. From there he went to work for AC Sparkplug on the Titan missile program, where he brought systems methods to bear on the troubled program and pushed it toward success. He then moved to the West Coast bastion of systems expertise, TRW, whence he was recruited to NASA headquarters as ''Deputy Director of Systems Engineering.'' Taken to quoting Milton and citing Greek mythology in engineering lectures, Shea took an unusually broad view of Apollo and its significance.

Shea's mission was to translate Kennedy's all-encompassing mission statement into precise engineering specifications. Along with George Low, Shea soon managed a 400-person organization to oversee NASA's projects. Here he wrote requirements for safety, reliability, and the probability of mission success, and studied how those measures manifested themselves in hardware specifications. Shea also managed Bellcomm as it was looking over the shoulder of the Apollo contractors, including the IL. Shea and his reliability analyses played a key role in the LOR decision, although he came to distrust a pure statistical approach. As he put it: ''So we finally concluded there was no way to assure statistical reliability. You had to really use engineering confidence and do a little wishing and lucking and hoping.''32 Shea's methods focused on ''interface control,'' formally defining in great detail every possible link among the spacecraft components, essentially creating a virtual model of the system in a mountain of paper.

Shea actually mixed the analytical methods of the systems men with an appreciation for focused, insightful, creative engineering. One understanding of systems methods calls for breaking a large system down into component parts defined as ''black boxes'' by specifying their input and output interfaces. Then the black boxes can be contracted out, and then recombined by a systems integrator. Shea indeed employed these methods, but he also brought an approach that adapted to the ultrareliability required for human spaceflight, what might be described as ''white box'' systems engineering. No box could be truly black, because every component was open to scrutiny (hence ''no random failures''). An assemblage of black boxes made a brittle system, a house of cards. By contrast, in a white box system, systems engineers always had the ability to peer into every subsystem, to examine every component. Nothing inside the Apollo system could be unknown or taken for granted. As an engineer, Shea saw how the system held together as a whole, but he could also see inside each piece and make tradeoffs between widely separated subsystems.

In October 1963 Mueller sent Shea to Houston, where he replaced Charlie Frick as manager of ASPO (''I was exhausted so much of the time I just couldn't hardly take it,'' Frick said of his Apollo tenure).33 Even though the official contract with North American for the CSMs (command and service modules) had barely been signed, unofficially the program was nearly a year late and plagued with problems.34

Shea's arrival in Houston marked a sea change in Apollo culture. Mercury and Gemini were largely being run by old Langley hands in Robert Gilruth's group who had made the move to Houston. These were aeronautical engineers, raised in research and test flying. Headquarters felt they were running Apollo too informally, like a research program, rather than a large-scale development and operations project.

Once in Houston Shea found himself in the middle of a classic NASA conflict between headquarters and the centers. Though he was Mueller's man, Shea tried to fit into the center. The Space Task Group engineers saw him as a spy;understandably so, as Shea, with his fancy specifications, insistence on documentation, and white-collar style was wresting control from them. ''I found a mess when I went to Houston,'' he recalled: deficient management of the contractors and poor control of the spacecraft and the interfaces between their various components.35 ''The spacecraft basically needed a major redesign.''36 At North American, it was even worse. He found a ''lack of configuration discipline,'' meaning that engineers and technicians were changing the design without complete documentation. This sloppiness, he confided to Mueller, was more than a matter of procedure: ''In the limit, we are really discussing the competence of their [North American's] people. I am convinced that we are not dealing with a first rate engineering organization.''37

Plenty of people resisted Mueller's centralization and Shea's style. ''I thought it was wrong and I still think it was wrong,'' complained Caldwell Johnson, one of the original Space Task Group.38 Powerful Max Faget, with nearly a thousand people in his design bureau, froze Shea out from gathering data. Chris Kraft viewed Shea as ''an outsider and an enigma . . . his close tie to George Mueller made him a stepchild among Mercury veterans.''39 Their objections were understandable: the close, hands-on engineering style inherited from Langley was slipping from their fingers, never to return. ''The growing mountain of Mueller's bureaucracy looked to us working troops as unnecessary, cumbersome, and expensive,'' Kraft wrote in his memoir. ''It turned out to be all three, but it never went away.''40 These tensions, between centralized control and decentralized management, between hierarchical systems and craft-based approaches to engineering, would remain throughout the Apollo program. They mirrored the human-machine tensions being played out in the hardware design: systemization versus intuition, professionalism versus informal organization, documentation versus personal trust. They would come to a head in the aftermath of the Apollo 1 fire.41 But that was still years away, and in 1963 the Apollo program had barely gotten started and already it was behind.

Telescopes Mastery

Telescopes Mastery

Through this ebook, you are going to learn what you will need to know all about the telescopes that can provide a fun and rewarding hobby for you and your family!

Get My Free Ebook


Post a comment