The American Institute of Physics Bulletin of Science Policy News

Number 81: July 11, 2000

“In September 1999, NASA flew the $125 million Mars Climate
Orbiter into the red planet, turning an advanced science platform
into so much junk burning up in the Martian atmosphere. In
December 1999, NASA repeated the feat and crashed the $185 Mars
Polar Lander as well.” – House Science Committee Chairman James
Sensenbrenner (R-WI)

NASA Administrator Daniel Goldin has appeared at congressional
witness tables numerous times this year to discuss NASA’s
failures and to defend his management approach of “faster,
better, cheaper.” On June 20, the House Science Committee heard
from Goldin yet again, and several other witnesses, on actions
NASA should be taking to avoid similar mishaps in the future.
Investigations have concluded that the Mars Climate Orbiter
crashed because a failure to convert English units to metric
affected navigation. Analysis of the loss of the Polar Lander is
less conclusive, but software problems and inadequate testing of
components have been suggested as possible causes. Several
recent reviews of these and other recent NASA problems pointed to
additional underlying factors, including a lack of experienced
personnel and open communication, failure to follow proper
engineering procedures, and willingness to tolerate risk to
achieve cost and schedule objectives.

The committee members in general were hesitant to be too critical
of Goldin and the space agency – Ranking Minority Member Ralph
Hall (D-TX) pointed out that “what NASA does really is ‘rocket
science'” – especially as Congress and the Administration have
slashed NASA’s budgets in the past decade and repeatedly called
on Goldin to do more with less. As he has in past hearings,
Goldin openly acknowledged that in trying to cut costs, he
“pushed too hard.” But he insisted on the validity of the
faster, better, cheaper concept, “if properly applied.”

Goldin said NASA has put together a team to address some common
themes from the reviews. While the team’s conclusions are not
yet finalized, Goldin commented on several areas. He discussed
the need to enhance employee and contractor training, coaching,
and mentoring, using the latest educational methods and
technologies. He noted the importance of “capitalizing on the
knowledge of others” to ensure that lessons learned are not
forgotten. He said all future programs should have a rigorous
formation phase with objective development of cost estimates and
budgetary reserves. He emphasized NASA’s increasing reliance on
software development, and said the agency needs tools that can
simulate missions end-to-end, rapidly explore options and
contingencies, and address safety hazards. He remarked that,
although in a few cases additional funds may have been needed, he
still believes “money is not the magic ingredient,” and he made
it clear that he has no intentions of asking the President for an
amended FY 2001 NASA budget request.

Two other witnesses took issue with some of the recommendations
presented in the reviews, warning that they might reverse NASA’s
progress. Alan Binder, formerly the principal investigator for
the successful Lunar Prospector mission, professed himself
“deeply concerned about” the main conclusions of several of the
reports, which he said call for more funding, management, and
oversight. If those recommendations are followed, Binder said,
“I firmly believe that we will have started down the slippery
path back to the old, expensive, slow way of exploring space.”
Both Binder and Pedro Rustan, who was mission manager for the
Clementine spacecraft, strongly endorsed the faster, better,
cheaper concept and opposed any moves toward greater oversight
and higher funding. They instead stressed the importance of
experienced staff and managers, with full responsibility placed
on a single, “seasoned” person empowered to make decisions,
continuity of the team throughout the mission from “cradle to
grave,” freezing requirements early in the mission, extensive
testing, and limited reviews.

While Goldin agreed with many of their comments, he objected that
NASA needs flexibility in mission development, and might miss
opportunities if it could not change requirements throughout a
program. Rustan countered that NASA cannot add new requirements
and still constrain program schedule and costs: “You can’t have
it all.”

Sensenbrenner chided Goldin for offering “more platitudes and
generalities than specifics” in his testimony. To suggestions
for continual learning in the workplace, high confidence in the
early phases of projects, and focusing oversight where it is most
needed, Sensenbrenner’s wry response in each case was, “no
kidding.” He said NASA has “learned and applied” the formula for
success, but then “apparently forgot.” For all his criticism,
though, the chairman remains a supporter of the agency and the
faster, better, cheaper approach. “I hope sincerely that [the
two Mars failures] were just a bump in the road,” he concluded.

Audrey T. Leath
Public Information Division
The American Institute of Physics
(301) 209-3094