|
Author
|
Topic: "Attrition: Forecasting..." Q
|
Mad Dog
Senior Member
|
posted 12-01-2009 10:39 PM
  
I recently picked up "Attrition, Forecasting Battle Casualties and Equipment Losses in Modern War" by T. Dupuy.Reading through it, I was surprised to note that while infantry losses are dependent on the P/P ratio, armor losses are not. This seems somewhat counter-intuitive, yet, the examples given in the book match historical rates fairly well. Does this still hold in TNDM (TDNM ?) ? Can anyone explain why this is ? Is this the inability of a large force to effectively focus on a much smaller enemy force ? thanks, Mad Dog
IP: Logged |
Chris Lawrence
Moderator
|
posted 01-06-2010 12:26 PM
quote:
Originally posted by Mad Dog:
Reading through it, I was surprised to note that while infantry losses are dependent on the P/P ratio, armor losses are not.
Armor losses are tied to personnel losses as modified by various factors. This is true for the QJM and the TNDM. The Attriton Book provides a very stripped down description of the model. As armor losses are tied to personnel losses, then as such, they are tied to P/P ratio. The original basis for this was a series of studies done in the 1970s that compared various equipment type losses to personnel losses. This study was later used by the people working on the Atlas model. In their case, their model produced armor losses (due to the Carmonette model) but no personnel losses. So they simply reversed the equations to produce personnel losses. I gather the "Atlas curves" are used in some current DOD models to this day.
IP: Logged |
Mad Dog
Senior Member
|
posted 01-09-2010 01:54 PM
Thanks Chris. I see I overlooked that. I was looking for the "op" factor in the armor loss formula.It struck me as odd that the armor losses were based off of personnel losses - I was hoping to find a way just to derive armor losses without having to find the personnel losses first. Im still scratching my head on how to do that, but I will see if I can track down Atlas or the Carmonette mode.l Its nice to know someone is still here. thanks, Tom
IP: Logged |
Chris Lawrence
Moderator
|
posted 01-11-2010 09:16 AM
Carmonette and Atlas are old models designed in the 1950s and 1960s that ran on mainframes. Don't know if they are currently available anywhere or if you could get, how easy it would be to get them to run.Some of the more current DOD models have evolved from them. We have always had some criticisms of these models. Atlas was supposedly validated to the France 1940 campaign. According to rumor, the French marched to the Rhine! Also according to rumor, the designers concluded back in the 1960s that "history has no relevence to modern combat". I don't know how true these rumors are and have never been able to find anything in print to confirm them.
IP: Logged |
Mad Dog
Senior Member
|
posted 01-30-2010 08:27 PM
Is there a way to predict armor losses without working through the personnel loss rates ?If I try and predict the loss rates using the system in "Attrition", for a single tank battalion attacking by itself (US WW2 tank destroyer, for example, with 671 men and 36 M-10), the loss rates are large. The system predicts that the unit will lose ~80 AFV within a day (this assumes most factors are = 1). Am I doing something wrong here ? Personnel loss:
= 0.04x(N)x(rc)x(hc)x(uc)x(tz)x(op)x(Su)x(so)
N=671 men (a single US M-10 TD Bn)
rc=hc=uc=op=Su=so=1 (for our test case)
tz = 7.61 (derived from line fitting "Attrition" data) = 204.28 (or a casualty loss rate of 0.304) Feed this to the armor loss rate: armor losses = (CR)x(CKT)x(NT)x(CEV)x(tz)x(Sui)
CR=0.304
CKT=5.4 (using the value from NPW instead of Attrition)
NT=36
CEV=Sui=1 (baseline assumptions)
tz = 1.35 (derived from line fitting "Attrition" data) daily loss = 79.99 tanks thanks, Mad Dog
IP: Logged | |
The Dupuy Institute Forum
UBBFriend: Email This Page to Someone! 
