SpaceX Loses One


I was listening to NPR on Saturday morning when a brief news item mentioning SpaceX’s planned Sunday launch said two previous launches to resupply the ISS had failed. I couldn’t help but wonder: what are the chances? Actually not that high—or low, depending on how you want to look at it. The SpaceX Falcon blew up 150 seconds into the launch phase and before staging.

It was the 19th launch of a SpaceX booster, but hardly the first vehicle loss. It took SpaceX three tries to get a successful launch, having started in the space biz in 2006, a record that’s hardly unusual in a high-risk, high-difficulty business like space launches. Given that the two previous ISS resupply launches also failed, I figured the chance of a hat trick to be remote. You can see why I’m not in the aerospace risk analysis field.

In due course, they’ll figure out what happened and move forward. NASA says the ISS crew isn’t in danger of going on short rations unless a fourth mission fails. Hey, how likely is that? (Never mind.) But I did see that one NASA official confidently said “there is no negligence here,” referring to Sunday’s SpaceX failure. Really? Maybe a bit soon for that sort of pronouncement for when ^%$& happens, it usually has someone’s or a group of someones’ fingerprints on it.

I’m not casting blame, mind you, but trying to reel in a segue to a book I’ve been reading, Stanley McChrystal’s new Team of Teams: New Rules of Engagement For a Complex World. This is an interesting piece of work that’s part management treatise, part sociology theory and part academic dissection of organizational dynamics. The theme of the book is McChrystal’s comparing his experience as a commanding general in Iraq with other organizations that have faced complex challenges in a chaotic world. In 2004, Iraq certainly was that. As have other military leaders, McChrystal says lavishly trained and equipped U.S. forces were in no way prepared to cope with a metastasizing insurgency whose outlines bore no resemblance to anything taught at West Point. He examines his experience through the lens of many historical organizational and leadership successes and failures and one of these was NASA’s Apollo program.

Apollo is commonly thought of as a towering technical success and it was, but without the supremely competent organizational method developed by George Mueller, man might still be trying to get to the moon. Mueller, you may recall, introduced the idea of all-up testing, replacing tests of individual systems and components with fewer tests of larger, integrated elements. Thus it was that only the second Apollo vehicle to fly a manned mission, Apollo 8, went into lunar orbit, testing many systems for the first time. More important, Mueller and NASA had discarded the traditional command-top-down management style in favor of a decentralized approach that knocked down at least some of the silos that are naturally erected inside large, complex organizations.

In his analysis, McChrystal compares NASA’s organization during the Apollo years to a foreign counterpart and not the Soviet Union, but ELDO. ELDO? What the hell is that? I consider myself a bit of a space nerd and I’d never heard of it, either. It’s the European Launcher Development Organization, a precursor to the European Space Agency that, in the early 1960s, was on a parallel track with NASA with the goal of developing a booster suitable for commercial space operations. ELDO was a consortium of six countries, including Australia.

As McChrystal tells it, ELDO suffered a string of failures not because it wasn’t technically competent, but because of shortfalls in organizational communication. The silos weren’t talking and ELDO never figured out a solution before dissolving the effort in 1974. Of course, it’s well understood that by the early 1970s, NASA was losing the Apollo bubble and even as early as 1967, following the Apollo 1 fire, management and organizational oversights were at fault more often than technical blinks were. This proved true of both Shuttle accidents, as well. And remember the Mars Climate Orbiter that burned up because one working group was using the metric system while another the imperial? More silos.

And that gets back to SpaceX’s launch failure on Sunday. I’m sure they’ll track down the precise cause, but I can’t help but wonder if they’ll also discover that it has an organizational element. Undertakings as demanding as hurling things into space are technical challenges to be sure. But they have proven to be far greater organizational challenges for it is the organization that provides the focus and marshals the resources for the rocket scientists to do their thing. And it’s the organization that will be—or should be—capable of detecting that fatal gap between what one group thought another was doing and what both actually weren’t doing. SpaceX has roots in the tech sector and that industry is no more immune to silo construction than government agencies are. People are people, no matter who they work for.

I wonder if Elon Musk will be chewing on that thought this week.