Management Tips From the Real Rocket Scientists

By Jeffery Kluger

It's got so you can't tell the celebrations apart at the Jet Propulsion Laboratory (J.P.L.) in Pasadena, Calif. They back-slapped and high-fived when the Pathfinder lander bounced down on Mars in 1997 and when the Spirit and Opportunity rovers followed in 2004. They cheered when the Cassini probe went into orbit around Saturn last summer and when the Huygens lander reached the surface of the planet's moon Titan months later. If it's possible to grow tired of popping corks and raising glasses, the J.P.L. engineers may be getting close.

Such high spirits--and the successes behind them--don't always make headlines. The world has long had a weak spot for spaceships that carry astronauts, often overlooking the robot craft that fly out on their own. And yet in the long history of space travel, it's the robots that have gone the farthest and done the most. And it's J.P.L. that has sent the majority of them on their way.

In the four decades it has been affiliated with NASA, the lab has dispatched probes to seven of the planets and dozens of their moons and has tossed in the sun, our own moon and a comet for good measure. "We do interplanetary," says Pete Theisinger, a deputy director of Mars exploration. "We're not the only people who can do it, but we're the only ones who have done it for 40 years."

The question for both science types and business types is, Why? What is it about J.P.L. that has made it the little space engine that could? The lab is undeniably a science center first. But it's a business of sorts too, facing all the challenges that confront any going concern. Industry could learn something from six straightforward rules J.P.L. has always taken care to follow, rules that may have done more than anything else to make it such a thriving operation.

CULTURE OF CRITIQUE

Big-company R&D is a secretive thing, often taking place in isolated Skunk Works closed off from the rest of the company, to say nothing of the world. That's not the way things work at J.P.L. The lab is not owned by NASA but rather is a nonprofit, federally funded research center managed by the California Institute of Technology and does its work for NASA under contract. The academics who work there come from the world of peer review, in which even theoretical work isn't considered sound until a lot of objective eyes have had a chance to look at it. When smart people ask questions of other smart people, often as not they get smart answers. That has saved more than one J.P.L. mission.

The two rovers currently rolling across Mars may never have made it to the surface if someone who saw the plans had not begun fretting about the Martian winds the ships would encounter and argued for additional thrusters to counteract them. The thrusters were added, and the design change made all the difference. More recently, the lab was planning the less publicized Gravity Recovery and Climate Experiment (GRACE)--a twin-satellite pas de deux designed to measure Earth's gravitational field and its effect on ocean currents. A critical step was eliminating any wobble between the ships. J.P.L. staff members had been working with engineers at Stanford University on a thruster that could nudge a spacecraft with boiled-off helium. It was perfect for GRACE's needs. "At J.P.L.," says Rob Manning, Mars program manager and former chief engineer for the Pathfinder mission, "anybody can go to any meeting and criticize or comment."

PIE-IN-THE-SKY THINKING

Sometimes those comments can get downright loopy, and that's how J.P.L. often likes things. Openness to wild ideas goes back to the 1960s, when the lab established an office to dream up plans for future missions, and an engineer crunching numbers one day happened to notice that in 1977, Jupiter, Saturn, Uranus and Neptune would fall into a rare planetary conga line that they would not form again for 176 years. This insight set the stage for the spectacular four-planet Voyager flights of the 1970s and '80s. Today the business of blue-skying ideas has become more institutionalized with a 20-person group called the Advanced Projects Design Team--or Team X--which meets three times a week.

Scientists with a suggestion for a future mission book time with Team X and pay for it out of their own operations budget. At a lot of institutions, that would be a little like buying the stool that will be used at your own public dunking. But J.P.L.-ers don't see it that way. Team X even has an improbable name for the scientists who come before it: customers. "We hardly ever end up where the customer came in thinking we'd end up," says engineer and team member Tracy Leavens.

That's no surprise. Team X manager Robert Oberto describes the several-day process during which an idea is discussed as "organized chaos," with all the dimensions of the mission being considered at once. That speeds things up by letting the engineers working on, say, the propulsion system know what the engineers working on the camera system are thinking--a good thing, since cameras add weight, affecting thrust. Of the 70 or so missions that come before Team X in a year, only one or two are ever recommended to NASA. The group technique assures that those two missions are well thought through and, by this time, well loved. "I need them all to own the mission," says Oberto. "That's very important."

EACH ONE TEACH ONE

The greatest resource J.P.L. brings to the space game, of course, is not good ideas as much as the people who generate them. The lab has always benefited from having its farm team of apprentice scientists right next door at Caltech. The first thing young engineers who come to work at the lab must do is learn the ways of J.P.L. as an institution, something that's easier to do here than at most other places of business. As long ago as the 1960s, J.P.L. embraced a concept known as "each one teach one," under which senior members of any team were charged with the responsibility of bringing at least one junior member along. In recent years, that system has taken a bit of a hit.

Tight budgets in the '80s meant fewer missions and fewer new employees, and when the engineers with the institutional memory retired, no one took their place. "There were no longer chances for people to study under the masters and learn their trade," says Gentry Lee, J.P.L.'s chief engineer for solar-systems exploration. "The depth in the process of building the next generation to go where no one has gone before was a little bit broken." The money spigots opened up again in the '90s, but by then there were gaps in the lab's history chain.

J.P.L. is doing what it can to fix that problem, filling the mentoring hole by compiling painstaking lists of design principles--lessons learned when particular engineering techniques either did or didn't work. That system should serve in a pinch, but the lab is determined never to let the teaching system lapse again. "Mentoring is something we've been doing ad hoc for years," Manning says. "That's allowed us to collectively grow."

THE INCOMPRESSIBLE TEST

Of course, just passing on rules is not all there is to keeping things running. You have to enforce them too. Space deadlines are uncompromising things: miss a launch window when you're trying to send a probe to Mars, and you will have to wait two years until the planets line up properly again. The danger is that this kind of pressure can cloud the judgment of even the best project managers, leading them to rush as time grows short. For that reason, before a single bit of metal is cut on any spacecraft, J.P.L. bosses draw up what they call an Incompressible Test List--milestones that must be achieved before the spacecraft is certified to fly. "We do that early in the game when things are still calm," says J.P.L. director Charles Elachi. "We put it in a drawer, and I tell the team, 'I'm not going to let this thing launch until you verify that each item has been done.'" Sometimes a single rule can save a mission.

When Manning was helping design the two rovers now operating on Mars, he was faced with a problem of how to sever electrical cables so that the surface-bound vehicles could separate from their delivery craft. Rules require that power be shut down before the cables are cut in order to prevent a short, but there was insufficient space inside the ship for all the electrical relays to do that job. Manning's team designed a fix that he thought would sidestep the problem, but when he tried it out in tests, it failed--as the people who drafted the design principle had predicted. "People came back to me and said, 'Look, this really is a rule,'" Manning recalls. It's not one he ever broke again.

TOLERANCE FOR FLOPS

As smoothly as the J.P.L. systems run, the true test of the lab's business model comes when something goes wrong. Every time the lab pushes the launch button, billions of dollars, dozens of careers and decades of planning can be on the line. J.P.L. not only accepts the likelihood of the occasional costly flop but also expects it. Such a stomach for setbacks is a legacy of J.P.L.'s first director, William Pickering, a Caltech alumnus who learned his trade setting off rockets in the dry riverbed that is all J.P.L. once was. Dozens of those rockets sometimes blew themselves to bits before one finally flew, but each failure taught Pickering something. "The era of rocketry really was trial and error," says former J.P.L. director Ed Stone. Adds Elachi: "Almost every lab director has kept the same philosophy." That has helped the lab survive many rocky patches, such as in the 1960s, when Pickering's moon probes flopped six times before Ranger 7 succeeded; and in 1999, when the Mars Polar Lander and the Mars Climate Orbiter failed. The message from management remained the same. "We encourage [employees] to push the limits in a thoughtful way," says Elachi. "We keep telling them, 'Look, setbacks are going to happen. And when they do, we will stand by you.'"

OLD-FASHIONED MARKETING

There is one final--much larger--constituency that must stand by the lab: the public. Like any going enterprise, J.P.L. must sell its product, and one way it does that is with pictures. Out in the regions where J.P.L.'s ships fly, the sun is reduced to little more than a very bright candle. The color of planets and moons can be inferred by their chemistry, but even the most sharp-eyed cameras can't see the palette for real. J.P.L. has never been shy about photographing the bodies in grays and pastels and then adding a little colorizing--always taking care to disclose that fact. Critics have grumbled about that, but J.P.L. knows that if it's going to sell the steak of the science, it must rely on the sizzle of the pictures. "I would compare it to the people who painted the Rocky Mountain West in the 19th century," says Roger Launius, chief of the space-history division at the National Air and Space Museum. "They came back and presented the pictures to people, and everybody got excited about the prospect of going there."

Just where J.P.L.'s westward-ho push will take it next is not yet set. More missions to Mars are certain. Others to Pluto and Jupiter's moon Europa are possible. The still vague nature of these plans does not bother the veterans who know the lab best. "What we do here is try to figure out how impossible a task is and then we take it on," says Lee, a 30-year J.P.L. veteran. "The result is that we add to human knowledge. Now, can you have a better life?"

With reporting by with Dan Cray / Pasadena