Trends in ISS Anomalies
International Space Station (ISS) Anomalies Trending Study, NTRS
“The NASA Engineering and Safety Center (NESC) set out to utilize data mining and trending techniques to review the anomaly history of the International Space Station (ISS) and provide tools for discipline experts not involved with the ISS Program to search anomaly data to aid in identification of areas that may warrant further investigation. Additionally, the assessment team aimed to develop an approach and skillset for integrating data sets, with the intent of providing an enriched data set for discipline experts to investigate that is easier to navigate, particularly in light of ISS aging and the plan to extend its life into the late 2020s. This report contains the outcome of the NESC Assessment.”
I found the paper interesting because the contents did not seem to match the title. What the team did was to create interactive tools to allow users to look for trends in anomalies. That said, this looks to be a very useful tool for engineers (and even managers) to search for trends related to their areas of responsibilities.
I wouldn’t get too excited. I’m familiar with the ISS problem reporting databases firsthand. They remind me of the older Shuttle databases, developed almost exclusively as interlocks for very narrow work processes.
That is, the data tracking’s first purpose is to identify a condition needing attention, then to document and convey the interlock as resolved, that the condition no longer needs attention. Calling these ISS problem reporting systems, like MADS and PART, “databases” is a like calling my PC from 1987 a Hal 9000. A bit ridiculous. These systems are designed and meant to track conditions narrowly with little to zero regard for how that condition relates to any bigger picture, trend or insight.
This is why you get studies like these trying to explore anomaly trending, and being very kind about the potential of these systems, as if to say – it’s all we have and maybe we can make some lemonade out of these lemons.
I feel the paper was focussed very heavily on the process and software, which was mostly descriptions of proprietary database tools, minimally on a few very high-level trends, and to a very limited degree on procedural controls. There was no identification of actual failure modes at a hardware level or recommended engineering controls (hardware modifications) to eliminate failure modes, and it isn’t even clear that the data is sufficiently detailed to identify such countermeasures.
Simply indexing the anomaly reports with Google or a similar text indexing system would probably provide a reasonable degree of access to people investigating reliability problems, however one of the issues that isn’t addressed is the question of who, exactly, has access to the reports. Another difficulty, as numbers_guy points out, is that recognition of an occurrence as an anomaly may change with the experience of the crew.
It sounds like a response to review. Someone got an action item to improve trending and closed it by paying someone to develop trending software. Now the software is written and tested with a few examples. If that is what happened, and the action item is closed, I’m not sure anyone will check on whether that software is actually used.
By the way, Google’s indexing system wasn’t developed as class A software. Anything about anomalies on a human spaceflight mission probably requires that.
Rolling on the floor laughing…. Of course you are right! Why didn’t I think of that?
Hopefully they have a better handle on it than this reports shows. ISS has now been in orbit longer than Mir, which lasted 15 years and which had a lot of anomalies in its last years. ISS probably was designed to be more maintainable, though without Shuttle the capability to trade out major spares is pretty limited. I think recent problems with the EMUs, the sabatier reactor, ammonia and hydrogen leaks, and other technical issues are symptomatic of things yet to come. And from what I can tell the mindset is not too different than before Challenger or Columbia-the problems have not caused fires, or explosion, or decompression, or killed anyone yet so lets keep going…
Folks, as a space cadet this is really important. We need to learn what kinds of systems work best in the LEO environment. Hopefully most (but not all) of those systems will also function at the higher radiation environment of cislunar free space. It should be easier to create systems to work on the lunar surface, with abundant shielding and partial gravity.
But the knowledge needs to be captured.
As a policy wonk I’m horrified that we have spent $70+ billion building and operating ISS and don’t really have a handle on how to best capture the data on how operable the system is. The answer can’t always be “ask Boeing”. We need to develop space goods and services with an eye towards affordable/sustainable operation to deliver the public goods (research time on orbit, astronaut sortie time on the lunar surface, etc…) that are promised, not just winning development funding.
I hope AES’ NextSteps program is paying attention to this.
This is really what one of the primary reasons was for ISS (or the Space Operations Center, or Space Station Freedom) was at the outset. I am not quite sure why in the last few years ISS is mainly about world class scientific research-sure we hope that cancer and other major diseases will be resolved as a result of research on board, but it might be a bit too much to expect; it may never happen. But there were a lot of other reasons for a LEO station, including first and foremost a classroom for learning to live and work in space. Yet, these short sighted, non-strategic thinking space operators seem to now want to ditch the thing and go onto something more exciting. They ought to maximize returns and learn to use what they have.
It seems to me that learning to live and work in the space environment is the major thing ISS should help us to do, and what we need before venturing farther and longer into deep space. Of course it needs a centrifuge, so we can learn more about the human response to various levels of gravity, for various durations.
I think this is the long-tent pole we need to address before we can confidently send a crew off to Mars or anywhere much beyond Sun-Earth/L2. Every year we delay this is likely to be a year lost in the program
We need a good countermeasures regimen for weightlessness, so it should also be adequate for partial-G. There’s no practical way I’m aware of for people to live in a centrifuge in space, so it becomes just another exercise device and probably not as efficient as a reasonably well designed resistance/cardio machine, though I’m not sure we have one yet.
I suppose the livability of a “centrifuge” is a function of size. At some point while centrifugal force is involved the idea of living in a centrifuge becomes unhandy.