Defense News website has an article about Navy ship design plans that almost … not quite ... nibbles around the edge of a good idea and then, like everything the Navy tries, collapses into the trash heap. Let’s take a look at what the idea is, why and how the Navy is going to screw it up, and how it could, with just a bit of tweaking, be a good idea.
What It Is
The gist of the idea is that the Navy wants to emphasize maintainability in future ship designs. Hey, that could be a good thing, right? I mean, how could it not? The ability to perform maintenance by a ship’s crew can only lead to greater availability and readiness of the ship. The close brother of maintenance, repairability – allows a crew to conduct repairs in the heat of battle and will enable a ship to stand and fight longer and to recover quicker. Modifying routine maintenance to make it easier and more efficient would also be a worthy goal. Reducing the amount of routine maintenance by designing less maintenance-intensive equipment would be a time saver. These are all worthy goals. We have the potential for a good idea, here.
What enormous benefits we could reap by designing a ship, from the very start, to have easier and less maintenance, more maintenance involvement from the crew, and expanded and easier repairability!
So, exactly, is the Navy saying about enhanced maintenance? Here it is,
The U.S. Navy is willing to pay more for new ships upfront if it means saving on maintenance and personnel costs throughout the life of the program, the head of amphibious, auxiliary and sealift programs told industry this week. (1)
‘… if it means saving on maintenance and personnel costs’. Oh, oh. I’m getting a bad feeling about this. Please don’t screw up a good idea, Navy.
How The Navy Will Screw It Up
What does the Navy mean by this? Here’s an example,
“Having reliable equipment that does not require a lot of operator action to maintain will enable LAW [Light Amphibious Warfare ship] to handle reduced manning, which, that’s the goal. It may require a larger upfront investment in higher-quality equipment, but we’re willing to do that to offset the sailor cost in the future.” (1)
You caught it, right? : ‘reduced manning, which, that’s the goal’
Well, there goes a good idea. We’re starting to see how this idea is falling apart. The issue and the emphasis isn’t actually maintainability … it’s reduced manning – the Navy’s long-time Holy Grail. This was the key characteristic of the LCS (not combat) and a major driver for Zumwalt, and Ford. All those ships have been abject failures in terms of manning and maintenance. The Navy has learned nothing and now they’re going to repeat the actions that led to failure
Seemingly unrelated side note: The definition of insanity is performing the same set of actions and expecting a different result. But, I digress …
The Navy goes on to explain,
… LAW will include sensors and monitors enabling not only conditions-based maintenance, through which maintainers can see how well a system is running and conduct maintenance guided by performance data, but also remote condition-based maintenance. The small crew may not be the recipient of the data, but maintainers in regional hubs could receive the data remotely and help schedule maintenance periods into the ship’s operational plans. (1)
Uh, déjà vu! This is almost word for word the description of the LCS maintenance model. The LCS was fitted with hundreds of remote sensors that were supposed to continuously send their equipment monitoring data back to regional shore maintenance facilities (EMCON, anyone? But, I digress …) that would use the data to schedule condition-based, predictive, preventative maintenance as well as routine maintenance. The concept failed miserably for a variety of reasons.
And now the Navy is going to do it again with the LAW. What was that definition of insanity?
So, let’s review.
An emphasis on maintainability, crew involvement, and
repairability with maintenance enhancements and considerations designed in
could have been a good great idea, as we noted.
The Navy immediately screwed it up by focusing, yet again, on reduced manning rather than actual maintenance as it relates to combat, damage control, and repairability.
What Could Be
Properly and robustly implemented, enhanced maintainability could give us:
- Ships with greatly enhanced combat resilience thanks to easier and faster repairability.
- Ships with longer service lives thanks to enhanced maintenance.
- Ships with greater endurance at sea since they wouldn’t have to put into port for every minor maintenance or repair issue.
- Ships with enhanced combat readiness since equipment that failed would be more likely to be repairable at sea and equipment that is easier to maintain and repair means that the overall ship’s readiness would be consistently greater.
You realize what we’re talking about, here, right? Yes, we’re talking about designing and building ships with greater combat readiness and resilience. There’s part of your long-sought force multiplier, right there: just the ability to stay at a higher level of combat readiness!
Why, oh why, Navy, can’t you just see and do the right things? Why do you have to ruin every good idea? Stop trying to link everything to reduced manning. That’s already been proven to be a failed concept. I know it hurts to think but just read this blog. I’ll do the thinking for you. I promise, your naval world will be so much better and you’ll look like geniuses.
______________________________________
(1)Defense News website, “Navy willing to pay more for more maintainable ships”, Megan Eckstein, 5-Aug-2021,
Could you please elaborate on the failure of predictive maintenance ? I searched on the internet for information about the LCS predictive maintenance and found next to nothing, apart from a small article that said they were using it (https://militaryembedded.com/ai/big-data/u-s-navy-lcs-adds-predictive-maintenance-analytics-after-installation), to be honest it read more like advertising than anything else. I also found a GAO report talking mostly about failures due to the small crew size but nothing about predictive maintenance.
ReplyDeleteI am curious to know why you think it is not working.
D614-D623
As you know, the LCS was designed to be maintained by shore-based personnel and facilities with no maintenance actions by the ship's crew. One of the foundations of that effort was automated shipboard data collection on all the equipment. The hundreds (thousands?) of remote monitor devices would broadcast their data back to the shore facilities so that repair personnel could use the data to observe trends in equipment and predict when equipment failures were imminent when or routine maintenance was required.
DeleteThe concept failed from the start when it was discovered that the LCS lacked the communications bandwidth and frequency to transmit the enormous amount of operating data back to shore. Thus, the land-based maintainers were unable to predict anything.
Empirically, the constant stream of mechanical breakdowns proved the failure of the concept to predict maintenance requirements.
This was described in detail in some of the GAO reports on the early deployment attempts, among other sources.
Thank you for your reply, I only found a very recent GAO report that didn't mention much about this.
DeleteI fail to understand the way they wanted this to work.
I get the failure because of the lack of bandwidth but if they were planning to use the deep learning side of AI it seems badly designed : to make deep learning work you need to have some knowledge of the desired situation, in this case to be able to characterize a failed or soon failed piece of equipment. At the beginning of the operational life of an LCS you do not have any of this knowledge because you do not have any failure (I hope) ! There are only two ways to get round this limitation :
- Wait until you can build the knowledge (the sensible approach in my view) but this would mean that the predictive maintenance system isn't operational initially, with probably some months if not years to have it ready (the time needed for the failures to arrive and for the time history to be analysed).
- Have some simulation data that will tell you how a piece of equipment changes its behaviour when it is wearing out. This seems quite far fetched but you never know ...
It looks to me like somebody oversold the concept big time to a decision maker that didn't understand anything. This is a shame because I believe that it can be made to be a valuable tool, not perfect obviously but useful, provided you don't expect miracles.
D614-D623
Predictive monitoring assumes that component failure modes are linear and that stress/workload are predicable. Linear in this case means "X component will has YY% chance of failure after ZZZ cycles or conditions".
DeleteElectronic component failure modes are rarely linear.
Reliability is measured in Mean Time Between Failures (MTBF) which is statistically extrapolated by the manufacturer during component design and testing.
Electronic failure tends to be due to thermal or power cycle stress. This is why computers tend to fail when they are restarted or powered on. There is no methodology I am aware of that can adequately predict electronic component failures since reliability has to be extrapolated over the component's service life since we're not going to wait years to complete an actual testing cycle.
Mechanical components are more linear, but stress and workload for military implementations are not predictable. In other words, we can't predict how much time a ship will spend at full speed over a given timeframe, especially in wartime.
These systems are also complex and tightly coupled. This means that a failure of one component (such as a cooling system) would immediately impact other components, and those interactions may not have been fully modeled and tested.
Finally, you have to have enough lead time to move a ship to where it can be maintained. If a ship is on the far side of the Pacific, the predictive alert has to allow enough time to return the ship to regional hub.
Regional hubs would therefore be a key interdiction target for the enemy as they are unlikely to be heavily defended and hardened. Even "take cover" warnings from a single incoming missile at a large base would disrupt service and delay ships from returning to combat.
"Predictive monitoring assumes that component failure modes are linear and that stress/workload are predicable."
Delete"I fail to understand the way they wanted this to work."
I suspect that you're both jumping way too deep into this concept. I suspect that the LCS concept wanted to operate on a simpler level. For example, monitoring a lube oil temperature … if the temp began climbing over time that might suggest a worn bearing that needs replacement. In other words, I suspect they were looking for indicators of impending failure rather than a pure, AI type, truly predictive failure capability.
Another example might be monitoring the current draw (amperage) of a motor. If it begins to increase over time, the motor may be failing.
And so on.
All of that is, however, pure speculation on my part as I have not seen any description of the detailed workings of the concept.
On a related note, I would also point out that continually broadcasting maintenance data during war is a good way to reveal one's position so the concept falls apart during war.
I haven't heard anything about the predictive maintenance program in quite some time. The Navy has been trying to transfer some of the maintenance responsibilities back to the ship/crew of the LCS after the failure of the shore-based maintenance concept, in general. I don't know whether the predictive maintenance has been abandoned, as so much of the LCS concepts have (module swapping, minimal manning, high speed, all the module concepts, etc.) or whether the predictive maintenance is still being attempted in a less comprehensive manner.
DeleteIf you want to investigate further, here's a quick link to get you started: LCS Maintenance
DeleteComNavOps what can you say about this idea to make Mark VI boats suitable against enemy mosquito boat Navy? https://www.navalnews.com/naval-news/2021/08/u-s-navy-advises-on-the-fate-of-the-mark-vi-patrol-boats/?fbclid=IwAR2uG7QzUOUd6c35c5FPJ2MmYnTcMc0pkZ36sggypVJrLhc5IoHdyCIOUfw
Delete"make Mark VI boats suitable against enemy mosquito boat Navy?"
DeleteThat's not what the article was suggesting. That aside, small patrol boats like the Mk VI are peacetime boats - useful for patrol duties where they can be forward maintained at nearby friendly bases. During war, no such bases will exist so the short ranged boats simply cannot operate. That said, they can be used in war for harbor patrol and similar rear area duties but they have no front line combat role. We've discussed this in multiple posts. Check the archives.
You could also look to the failure of predictive maintenance with the F-35. ALIS was a failure according to everyone from the GAO to finally USAF itself. The Air Force even has a replacement system authoriszed...but in the same procurement idiocy that gave us the LCS and F35, the contract to replace Lockheed's ALIS went to...Lockheed.
DeleteI don't understand the Leadership obsession with reduced manning. I rather have a 200 ship fleet that could be at the highest readiness, maintained by the sailors who will take them into harm's way than a 600 ship navy with most of the fleet waiting for contractors who don't have their asses on the line.
Besides, getting educated on repairing sophisticated equipment is why many join the Navy to begin with. Turning it over to contractors reduces one of the ways to recruit people for a short handed force.
The Navy apparently forgot in battle, one will likely suffer human casualties, i.e., a ship will lose maintenance crew members to injury and death, preventing her from receiving maintenance and repairs. I believe historical warships carried larger crews than they actually needed for day-to-day operations, "just in case" some crew members became casualties.
ReplyDeleteAnd now the Navy wants to omit the extra crew members its ships used to carry "just in case"? Is it actually preparing to fight wars against those who can fight back? Obviously not.
You need enough crew to
ReplyDelete1) sail the ship,
2) maintain the ship, and
3) fight the ship.
If you don't put enough crew on to maintain the ship, you can have no chance to have enough to fight the ship.
I read this a while back, but it didnt really fall into the current topic. You've dissected it perfectly. It's another example of 1) minimal manning and 2) tech for tech's sake.
ReplyDeleteWe all know minimal manning is a horrible idea- but continuing with 'predictive maintenance' is idiocy. When I was aboard ship, I quickly learned that everything aboard ship, EVERYTHING, had a dedicated book and PMS plan. From the radar, to the flight deck nets, to the VERTREP slings, to the hull and bulkheads. Every inch of the ship had a schedule for inspection, lubing, painting, testing, replacing, etc. And sure, some of it was overkill. Often the maintenance revolved around undoing/redoing the preservation (paint) work. But the fact was, the crew did everything. They became intimately familiar with all the systems if for no other reason than they were always fiddling with them. But now we want to run these systems until another system tells us work is needed. Im sure most of us can appreciate the difference between an actual gauge, and a warning light on the dash of our cars, right? What about oil changes in our car? Does anybody just keep using that oil until its so old that its going to cause problems, or do we just change it regularly?? Now to be fair, those predictive monitors could have thresholds set that trigger work before any damage happens, but why rely on that kind of system at all?? What's wrong with the mountain of manuals and scheduled maintenance thats served the Navy so well for a century?? And what if the predictive system fails?? Never mind that, even today, things just fail, with no warning, maintenanced regularly or not. So either way this idea has cost us X amount of dollars to help insure, not prevent failures, that may endanger or even lose the ship!!
This is another example of "just because we can, doesnt mean we should", and anyone with critical thinking skills can see this is one bad idea, driven by another bad idea...
"You've dissected it perfectly."
DeleteHey, I like that. The scalpel of naval analysis. Yeah!
"... When I was aboard ship, I quickly learned that everything aboard ship, EVERYTHING, had a dedicated book and PMS plan. From the radar, to the flight deck nets, to the VERTREP slings, to the hull and bulkheads. Every inch of the ship had a schedule for inspection, lubing, painting, testing, replacing, etc. And sure, some of it was overkill. Often the maintenance revolved around undoing/redoing the preservation (paint) work. But the fact was, the crew did everything. They became intimately familiar with all the systems if for no other reason than they were always fiddling with them...."
DeleteThis! This is one of the best comments in a long long time!!! This is a silent KILLER RIGHT THERE, even more than the crappy reliability or low manning! You can survive BOTH if you have a working knowledge and good habit/routine of fixing the broken systems BUT USN in it's infinite wisdom (sarc) is removing the sailor from the equation, take him away and have some contractor fix it, when sh#t hits the fan and contractor isn't there because there's a war going on, something breaks, everybody going to be looking at each other...discovering how to fix a gun in the middle of a firefight isn't healthy, it's a sure fire way to get shoot first before you even have time to diagnose the problem. Then when Congress will do an inquiry because USA loves those, we'll discover that some $10 part failed, they didn't have it, it was only the contractors that could touch it and sailors couldn't find the procedure in the manual to find the part or the-go around solution because they never practiced how to fix it or do it for real!!!
Another problem when you remove so much maintenance from the soldier/sailor and somewhat related to above, does the trooper/sailor HAVE COMPLETE CONFIDENCE that it works or how to operate it?!?! All us ex-military and I bet a few civilians here, we had to operate some heavy machinery, some new WHATEVER high tech gadget, first time, it's like like "crap, this is hard" then you open it up, do some preventive maintenance, maybe help the mechanic change some stuff in there, take it again a few times, you are now somewhat proficient or at least not as afraid as the first time, you have CONFIDENCE and KNOWLEDGE about the entire system, when you remove that maintenance part, you lose the knowledge part and a lot of the CONFIDENCE!!! "Is it going to work? Am I doing it right? Did I just break it? Is it working? Crap, now what?" bad time to ask those questions in a shooting war....
USN is BASICALLY degrading it's own fighting capability to save a few measly bucks!!!
I admit these issues were above my paygrade during my time in the Fleet but I ate my share of the terrible sandwich called "optimal manning".
ReplyDeleteI thought fatigue and a lack of sleep was a problem throughout the Fleet, as there are not enough sailors available for watch. Sure, I can be called "soft" but if sailors are fatigued during peacetime operations, what more less a wartime.
I don't know how the Navy is going to resolve fatigue/chronic lack of sleep against reducing manning.