OR text on DPICM

EDIT: Due to an embarrassing, but in my opinion not terminal, mistake I had to rework the figures after a few days. The emphasis was not on the originally incorrect figures, but on showing the approach to the problem, any way.

OR = operational research; the analytical approach on military effects
DPICM = dual purpose improved conventional munitions; "bomblets"

I've encountered a pattern of thought about artillery that's distinctly optimistic, almost Clancy-style. There's no OR study on this in my library (though there is one disclosed, for sure), so I'll improvise a quick&dirty OR mini-study on 227 mm DPICM vs. 155 mm HE to balance this optimism a bit:

A common claim is that MLRS (multiple rocket launcher system, MARS) batteries take out entire large areas in one fire mission. One claim I found repeatedly  was about the effects of a concentrated 24 MLRS launcher fire mission on a large area. The areas varied, but IIRC the smallest area claimed I ever encountered was 4 square kilometres (for such a quantity of launchers).

24 MLRS, each 12 M26 rockets, each 644 M77 DPICM HEAT-frag bomblets.
That's 185,472 bomblets. I looked up the dud rate quickly and it appears to vary from 2% to 23%, supposedly it is 5%. Let's subtract 5% thus:
176,198 bomblets.

Let's use a BMP vehicle as a notional target; length 7.14 m, width 3.2 m. The relevant target area is surely a bit smaller, more as 6.5 x 2.8 m: 18.2 square metres.

4 square kilometres equal 4 million square metres. 219,780 target areas would fit into this.

That's almost exactly 80% as many non-dud bomblets as target areas. Indirect hits don't matter much, for the small fragments are too weak to penetrate the bulletproof BMP (the frag effect is smaller than with defensive hand grenades).

Neither bomblets nor targets will be distributed evenly and regularly, though. Well-trained and disciplined mechanised troops would keep 50 metres spacing, while the bomblets would tend to bunch up and impact in a round-ish pattern. These patterns would overlap, of course. A BMP may remain unscathed even inside the pattern (if release wasn't too low for this), while others might be hit multiple times. Overall, we could expect that much less than 80% of notional targets would be hit. The percentage would drop accordingly if larger areas, longer-ranged rocket types or fewer launchers are claimed.

Let's assume in 4 sq km a third of the BMPs were hit; what would this mean? Some would have their driver knocked out, others their commander, others one or two infantrymen, others their engine, others their battery or other electrical components and a couple would probably burn out. Some might even explode because of missile or RPG munitions carried inside. A large quantity of externally mounted anti-tank guided missiles would become useless, as these are unprotected on a BMP (except with some upgrade packages).
The effects would largely be from the tiny, tiny shaped charge in the bomblet. These shaped charges are rated for about 70-100 mm penetration only and have a hollow cone in which during flight part of another bomblet is stored. This volume-saving design has been exploited for a simple countermeasure that ruins DPICM penetration; hedgehog-like rubber mats.* It's thus a very, very fragile mode of anti-tank attack.

M77 bomblet models
Now the comparison to HE shells. Obviously, a 155 mm HE shell direct hit is major trouble to any AFV, but even near misses are. See here once again for details. All those light AFVs rated as protected against artillery fragments are really only protected within constraints. A typical description is "minimum 15 metres". So we should not only count the target area for HE effect, but also an area around it. My guess is that the area relevant here is more like 300 sq metres.

A tactical equivalent of a 24 MLRS fire mission could be a four-round MRSI** fire mission of 24 self-propelled howitzers. That's 96 shells.

96 shells times 300 sq metres = 28,800 sq metres. Roughly a sixth of the MLRS mega-salvo.

Alternatively, they could keep firing for a minute, which would increase the fire mission of 24 modern self-propelled howitzers to 240 shells times 300 sq metres = 72,000 sq metres.

The problem for "team MLRS": The self-propelled howitzers may easily store enough ammunition for several such fire missions, the big Panzerhaubitze 2000 can even store 60 rounds. The MLRS batteries would scoot and reload, while the self-propelled howitzers could scoot, shoot, scoot, shoot, scoot, shoot ... the area it can cover with fires within an hour (nominally about 180 shells/hour) is bigger than a MLRS'. And the ammunition price per sq metre affected is much lower with HE (there's no unguided MLRS HE munition).

- - - - -

This isn't the only problem for MLRS awesomeness power fantasies: A single M26 rocket weighs 306 kg + cage, while a 155 mm HE shell + modular propellant and packaging comes at little more than 50 kg. This equals twice the AT-affected area per kilogram ammunition weight***. And ammunition supply is the supreme logistical problem in full blown European style wars (albeit alternating with diesel supply, depending on what the forces do at the time).
But weight isn't even the limiter here; it's volume. Only two to four MLRS sixpacks (about two tons each) can be carried on a heavy truck (nominal capacity 14 tons). Rockets use propellant inefficiently in comparison to howitzers (as do all recoilless systems), so they need a lot of (relatively low density) propellant. HE shells and modular propellant charges can be transported much more efficiently. Specialised tracked ammunition resupply vehicles can hold about 60-100 155 mm shells + propellant, and truckloads are similar. Comparison is a bit difficult because propellant modules are packaged and transported separately and published info is very scarce on this.

The fragmentation effect against soft targets is a different story and favours DPICM more unless the enemies are in fox holes. It's better to do this comparison with 155 mm HE vs. 155 mm DPICM, and the outcome is where the great preference for DPICM in the 70's and 80's was really coming from.

Maybe you'll hear some boasts about how rocket artillery as area weapon can take out / neutralize / whatever large areas. Remember then that such claims can actually be fact-checked easily with basic OR methods (=basic school math). And quite often, the boasts are just that; boasts, omitting the down sides.
There's a reason why artillery ammunition resupply is such a pain in the ass: You need much to cover a large area or to sustain suppressive fires for long. There's no "I win" button in any artillery system, and this includes the MLRS.

One more anecdote I've heard about the reliability of arty:
Some artillery fires demonstration during the Cold War, M110 howitzers (203 mm, twice as heavy shells as 155 mm) fired into an area of woodland. The shells were so big they were even visible in the air and huge fountains of dirt and dust emerged from the woodland area.  The spectators were most impressed by the awesome destructiveness and concluded that nobody could have survived such a bombardment. Just at that moment a group of deer ran out of the woodland - equally impressed and equally alive.


*: See here 2nd page, photo captioned "protection" for this effective low tech protection. The rubber inside the hollow charge disrupts the hollow charge principle of operation.
**: MRSI = multiple rounds simultaneous impact, in which the first round is fired in a high arc, while the last one is fired with more propellant at a low angle - impacts are within less seconds than the firing sequence and exploit the initial surprise effect better than normal fires would. MRSI is not available close to maximum range, though. Then again, it's very much possible below the 10 km minimum firing range of the MLRS.
***: With my assumptions about target, dud rate and effective indirect hit area.


  1. "24 MLRS, each 6 M26 rockets, each 644 M77 DPICM HEAT-frag bomblets." should be "24 MLRS, each 12 M26 rockets, each 644 M77 DPICM HEAT-frag bomblets." unless you meant HIMARS instead of MLRS.

    "The fragmentation effect against soft targets is a different story and favours DPICM more unless the enemies are in fox holes" is quite correct: table "Lethal areas per expended mass (m2/kg)", pp 73, of http://www.ffi.no/no/Rapporter/07-02345.pdf shows clearly that MLRS (M26 w M77) and M483A1/155mm are quite effective against soft uncovered targets and DM662/155mm is even more so. All these sub-munition warheads are substantially more effective, per logistic mass of warhead, against soft targets in the open, than M107/175mm shells - even if the later are equipped with prox. fuses. However, against harder targets, I believe your analysis is valid.

  2. Your numbers don't seem to work out.

    96 rounds could beat up an area of less than 200x200 m (40,000 sq m) with good destructiveness.

    2km x 2km is 4,000,000 sq m. i.e. 1% of the total area.
    (pretty insignificant).

    1. "good destructiveness" against well-dug in forces, yes. The frag-covered area of 155 mm HE is much larger than '96 for 200x200 m'.

  3. FYI, your numbers are off somewhat.

    The HIMARS carries only 6 rockets. Each MLRS can carry 12 rockets. That doubles your numbers.

    One other thing to consider is that while the tubes are probably best in an area consideration, in an area X time situation, MLRS delivers significantly more destruction saturation, which has some serious benefits.

  4. Sven,

    I believe the M270 MLRS has 12 rockets so that would be 12 rockets x 24 launchers. Maybe that changes the outcome and your position.


    1. True, but I think there's enough safety in there to cushion the impact even of a factor 2 error. Much was about the supply issue anyway.

  5. The Crusader proponents may have been right about the weapon system being a missed opportunity.


  6. I've always understood MLRS to be a complement to tube artillery rather than a replacement. The nature of the ammo makes it suitable for reinforcing roles, conducting counter-battery missions, where many of the targets are either lightly armoured or (in the case of many of the supply vehicles) unarmoured. They also offer the option of firing ATACMS, which I'll grant are very specialised and would probably be used far less than either conventional rockets or shells, but at least don't require yet another launch system.

    From a different perspective, M270 MLRS are lighter than most modern 155m self propelled howitzers, so have more tactical mobility options when using existing local infrastructure, and need a marginally lighter logistical support trail (offset, though, by the bulk of their ammo, as you mention).

    Finally, and this is a little beyond the scope of your post, a problem with shoot, scoot, shoot, scoot, shoot, scoot etc is terrain management. Keeping routes open for both the firing units and their depth supply echelons, and keeping firing boxes free from other units is a staff challenge, and one which is best managed by quickly winning the counter-battery (and air superiority) fight. Which brings us back full circle to MLRS in the reinforcing role . . .

  7. I am shocked that there have been no comments - this was a very good post and the ramifications for artillery (and air power) are obvious.


    P.S. I think there are nine MLRS in a firing battery and they fire 12 missiles each (or six ATACMS)

    1. Actually, it was faulty and I was merely without a computer and didn't unlock comments for days.

  8. I fixed the figures and wrote a correction as intro.

    It was a bit of bad luck: I took some of the less excessive boasts as example and made assumptions which were rather favouring the other side. Then I proceeded with the most stupid mistake conceivable and this messed up part of my case. I could have picked other figures favouring my side more and would have ended up with a more comfortable safety margin.

    Upside: The moral of the story (italic part at the end) was probably conveyed nevertheless.
    There are Clancy-style boasts of laymen or of low-ranking members of a military branch and then there's often the possibility to crunch the numbers and see whether the boasts are well-deserved.

    Artillery resupply is a huge pain in the a## of manoeuvre commanders, and an inferior efficiency munition only worsens the problem. In the end, it's not important what an individual round can do but what the arty resupply of a week does.