I came up with an interesting new idea based on the recent press conferences about the successful completion of NASA’s COTS program, Bigelow’s report on the study they did on commercial lunar development, and the SLS/Orion developers discussion of “Removing Barriers to Deep-Space Exploration“. I think the COTS model might actually work reasonably well for the development of SLS and Orion going forward. Admittedly, there aren’t competing suppliers, which is one big drawback, but the idea of using a firm, fixed-priced, milestone-based contracting method for SLS/Orion, combined with requiring skin-in-the-game is actually really intriguing for several reasons:
- One of the biggest challenges facing SLS and Orion is that with current NASA budgets, they can’t develop either system as quickly as they’d like to, and more importantly, a lot of the hardware that needs to be developed for non-stunt deep space missions (BLEO-optimized upper stages, hab modules, L2 waypoints, depots, landers, etc) can’t be developed in parallel with SLS/Orion. So you get a situation where you spend about 8 years developing SLS and Orion, and then you have about another 4-8 years of not being able to use them very effectively because they lack the other hardware needed. If SLS and Orion were developed using a FFP milestone-based approach with a 33-50% cost-share requirement (ie similar to what was used for COTS and what is used by the DoD for Other Transactions awards), that would free up $1-1.5B/yr over the next 8 years. That would be plenty to develop a large upper stage, hab module, and maybe even landers. Especially if at least some of those also used a COTS-like approach. Which would mean that for the same NASA budget, we could have the tools to do actual space missions in the early 2020s, instead of nothing better than stunts until 2025-2030.
- From what they say in all of their press briefings, it’s clear that Boeing in particular sees SLS as being on-time, on-budget, low-risk, and something that has wide market-potential both for NASA, DoD, and commercial launches. If they’re going to be competing against EELVs and the COTS launchers for government and commercial payloads, they should have to put skin-in-the-game just like the EELV, Falcon 9, and Antares developers had to. The good news is that if they did, they could make a strong case for being allowed to offer SLS launches on a commercial basis, reaping all the profits from all of those 20-30m telescopes, 3000m^3 inflatable modules, 2-3 launch ISS clones, and deep-space missions! Especially since it’s supposed to be so cheap, cheap, cheap!
- Especially if they drop the requirement for SLS to launch people, a COTS-type model should work really well–the propulsion systems exist, the requirements are pretty steady, and everything seems to line up for the kind of contract that should be FFP instead of cost-plus.
- By switching to a COTS-like insight vs. oversight approach for at least SLS, you’d be able to streamline the process, cut out a significant amount of red-tape, and speed its time to market.
- The best way for LM and Boeing to guarantee that SLS and Orion don’t get canceled is to get them to market quickly, and with enough of the other tools around to enable doing cool and useful missions. If the best they can do before 2025 is lunar flybys that could be done with Falcon Heavy and Dragonrider, their odds of getting canceled in budget squeezes is high. But if by the early 2020s, they had all the pieces in place to say support the Inspiration Mars mission, or launch a long-duration mission to an asteroid or Phobos/Deimos, or put an L2 waystation around the moon, then their odds of maintaining strong congressional support is higher.
- By pulling a lot of the Exploration hardware development from the 2020s to this decade, when SLS/Orion reach operational status, you might have enough money freed up to do higher-tempo, non-stunt deep-space missions in the early 2020s, even without huge increases in NASA’s top-line, and without having to splash ISS or do other drastic measures to free up money for HEOMD.
One important aspect of making this feasible would be to insist that none of the money saved by this approach would go towards the Commercial Crew program. The political supporters of SLS and Orion for the most part are deeply suspicious of Commercial Crew, and anything that looked like raiding their pot of money for a “competing” program would be bad. So my suggestion would be to require that say 80% of the saved money has to go towards developing the remaining key exploration systems (habs, upper stages, landers, and maybe depots), with the other 20% going towards funding serious exploration tech development and precursor missions. Wouldn’t it be cool if we had the budget to double down on Lunar and Mars ISRU demos, test out innovative aerobraking technologies, demonstrate cryogenic propellant transfer and storage on-orbit, and perform flight experiments to determine how much gravity humans actually need to survive/thrive, and then flight demonstrate the ability to do artificial spin gravity. $200-300M/yr could enable doing several of those over the next 5-10 years, if done competitively.
Now, there are some potential objections I can think of:
- There aren’t any competitors to SLS or Orion so you lose the competitive element of COTS: I think this one is the easiest to deal with. The reality is that right now there’s no competitors, and the contracts are cost-plus. So going to a COTS-based approach should be no worse than the status quo as far as “incentive to keep costs down due to competition” is concerned. In fact, since the contracts would be FFP, milestone-based, the contractors would have strong incentives to minimize costs, limit scope creep, and make sure these systems actually make it into commercial operations.
- Didn’t NASA want FAR-based contracts to provide the level of oversight needed for human rating? This is probably a more serious issue, but there may be a simple way of solving the problem. If SLS/Orion get rebaselined to not launch with a crew on-board, most of the man-rating requirements for SLS disappear entirely, and a lot of the hardest-to-meet man-rating requirements for Orion also go away. This may be enough to enable going to a COTS-like “insight” vs. traditional oversight approach.
- What about the loss of momentum due to a contract switch? If done correctly, you could let them continue on their existing contracts while the COTS-like program details are negotiated and finalized, and do a transition to the new contracting method down the road. The lowered bureaucracy going-forward should at least partially counteract the losses due to program transition.
- Would Congress go for something like this that decreases the federal dollars going to SLS/Orion contractors? From a jobs perspective around NASA centers, this should be neutral or even slightly positive. The jobs lost to doing things more efficiently are made up for by having more programs going on and leveraging commercial as well as NASA money. If anything this should be a net win for MSFC, KSC, JSC, and GRC.
- Won’t requiring LM and Boeing to put skin in the game just result in NASA paying more money in the operations phase? I know some knowledgeable people who’ve pointed out that if a company has to put its own skin in the game, that money comes with a cost, and they’ll factor that into operations, so the life-cycle costs actually go up. This may actually be the case in this situation, but when you compare those life-cycle cost increases to the waste of having SLS and Orion twiddling their thumbs for 5-10 years before the pieces needed to really use them come online, my guess is it’s a wash. Also, because LM and Boeing would be allowed to seek other non-HEOMD customers for their hardware going forward, it may lower the life-cycle costs to HEOMD (so long as they require Boeing and LM to charge customers a fair-share of the infrastructure).
- You hate SLS, why are you trying to save it? What’s your ulterior motive? I’ll be honest and admit that I’ve never been in love with NASA-run HLV projects. The main reason has been that it sucks the air out of the room for all the technology and other hardware you need to actually do interesting things. The other reason is that HLVs are mostly propellant launchers, which I think would be a better market for small RLVs and propellant depots. This solution doesn’t solve the latter problem, but may solve the first problem by freeing up money for trying out some of those other technologies and developing those other pieces of hardware sooner. Anything that aligns the incentives of Congress closer to funding development of the technology and hardware we really need is potentially a net win.
Anyhow, that’s my modest proposal. If LM and Boeing are serious about how low-risk, on-budget, and on-schedule their systems are, and how many awesome missions they’ll fly in the future, I expect them to want to jump in and support this approach wholeheartedly. Sure it’ll require an investment of their resources, but they think this is such a good investment for the government, I’m sure they’d agree that what’s a good investment for NASA should be a good investment for them too!