Modeling a Miracle: MANA and the Battle at Xa Long Tan 18 AUGUST 1966.

Modeling a Miracle:MANA and the Battle at Xa Long Tan

18 AUGUST 1966

Xa Long Tan:The History

- Battle occurred as a result of Australian involvement in the Phuoc Tuy Province of South Vietnam (1965-1971)

- After receiving mortar fire, the Australian Task Force (ATF) Base sends out Patrols to find the staging site for the attacks

- Patrols continue into the next day and D Company is sent to relieve B Company on patrol

Xa Long Tan:The History- The 108 men of D Company under Major Smith find the Mortar Site and then advance into a rubber plantation near Long Tan to search for the enemy.

- 11 Platoon makes contact, moves quickly ahead to pursue the enemy, but begins to receive fire.

- 10 Platoon moves south to relieve 11 Platoon but is also bogged down.

Xa Long Tan: The History- As first shots are fired, the monsoon breaks and reduces visibility to 50m

- “It’s bigger than I thought it was. They’re going to attack us!” - 2nd LT Sharp CO 11 Platoon

-ATF Base organizes relief force, but D Company is forced to hold out for nearly four hours before it finally arrives.

Xa Long Tan: The Outcome-D Company had engaged a full strength North Vietnamese Main Force Regiment!

- i.e. 108 men vs. 2500 men

-D COY suffers 17 DEAD 19 WOUNDED

-Vietnamese officially suffer 245 DEAD and an estimated 300-500 WOUNDED

? WHY ?

MANA: A Means to an Answer

- Analyzing history using computer modeling

-Each Agent represents 10 Soldiers (APC= 7)

- Battlefield represents 1 km2 on a 1:2 Scale (1 unit = 2 m)

- 20 Shots required to kill an agent in “shot at” trigger state (2 shots per man)

- 10 Shots to kill an Agent in any other trigger state

- Each agent can engage 10 items per timestep

MANA: A Means to an Answer

- Base case Sensor Range (SR) at 25 units (50m)

- Effective Weapon Range at 60 units (120 m) and Firepower (Accuracy) generally at 50

- Australian Situational Awareness at 40 (COY HQ at 70)

- Scenario ends upon arrival of APC Squadron

MANA: A Means to an Answer

- Australians represented by Blue (18 total Agents, 1 APC Agent)

-Australians given high Stealth due to their defensive posture (Stealth ~ 70-82)

-North Vietnamese are represented by Red Agents (85 total Agents)

- Australians are the limiting reagent in this experiment, so the focus of this study revolves around them and limiting the Casualties of the smaller force

WITHOUT FURTHER ADO

THE BASE CASE SCENARIO:Xa Long Tan as it Happened

LongTanBase1.dat

Experimental History

Now that we have been able to capture history, what do we do with it?

EXPERIMENT!- HOW?

- By changing Parameters and “Farming” Data

What If ?- What if there was no monsoon on 18AUG1966? (SR 75 = 150m)

- What if it had rained harder? (SR 10 = 20m)

- What if the Australians had been more aware of what was happening? (Satellite Uplink / Net Centric Warfare) (AWR +50%)

- What if all of the radios had been shot out? (AWR -50%)

- What if the Australians had more/less weapon training and or more/less National Servicemen in their units (FP +/- 50%)

The Answers?- The Data Farming in this experiment was done by taking the aforementioned questions and making them “reality”

- Each scenario of experimental history was run a total of 10 times (time constraint prevented more experimentation) and the number of casualties was recorded for each run

- The variation per seed is based upon the starting position of the Agents on the map

DISCLAIMER- This Scenario was developed and tested under MANA v. 0.93

-Due to a memory leak in that version, any scenario utilizing a sensor range of 75 units was tested under MANA v. 0.97

-It has been observed that Agents in MANA v. 0.97 behave distinctly different from those in MANA v. 0.93 under identical conditions

-Therefore all Data involving SR 75 should be observed with some caution when compared to v. 0.93

Altering Awareness Against Sensor Range

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 10

SensorRange = 25

SensorRange = 75

Ca

su

alt

ies

Maximum

Mean

Minimum

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 10

SensorRange = 25

SensorRange = 75

Ca

su

alt

ies

Maximum

Mean

Minimum

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 10

SensorRange = 25

SensorRange = 75

Ca

su

alt

ies

Maximum

Mean

Minimum

Awareness –50% Awareness: Base Value Awareness +50%

- Casualties increase steadily as Threat Awareness increases

- Possibly a result of increased engagement as more Agents are visible

Altering Firepower Against Changing Sensor Range/ Base Awareness

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 10

SensorRange = 25

SensorRange = 75

Ca

su

alt

ies

Maximum

Mean

Minimum

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 10

SensorRange = 25

Ca

su

alt

ies

Maximum

Mean

Minimum

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 25

SensorRange = 75

Ca

su

alt

ies

Maximum

Mean

Minimum

Firepower: -50% Firepower: Base Value Firepower: +50%

-Casualties decrease with increasing firepower

-Inverse relationship based upon the notion that the more enemy that are killed, the less of them are there to kill you

Altering Firepower against Changing Sensor Range at –50% Awareness

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 10

SensorRange = 25

SensorRange = 75

Ca

su

alt

ies

Maximum

Mean

Minimum

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 10

SensorRange = 25

Ca

su

alt

ies

Maximum

Mean

Minimum

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 25

SensorRange = 75

Ca

su

alt

ies

Maximum

Mean

Minimum

Firepower: -50% Firepower: Base Value

Firepower: +50%

- Fewest Maximum/Average Casualties occur when the Awareness is lowest and firepower is highest.

Altering Firepower Against Sensor Range at +50% Awareness

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 10

SensorRange = 25

SensorRange = 75

Ca

su

alt

ies

Maximum

Mean

Minimum

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 10

SensorRange = 25

Ca

su

alt

ies

Maximum

Mean

Minimum

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 25

SensorRange = 75

Ca

su

alt

ies

Maximum

Mean

Minimum

Firepower: -50% Firepower: Base Value Firepower: +50%

- Greatest Maximum/ Average casualties occur when firepower is at a minimum and Awareness is greatest

- Is ignorance Bliss? Probably not . . .

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 10

SensorRange = 25

SensorRange = 75

Ca

su

alt

ies

Maximum

Mean

Minimum

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 10

SensorRange = 25

SensorRange = 75

Ca

su

alt

ies

Maximum

Mean

Minimum

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 10

SensorRange = 25

SensorRange = 75

Ca

su

alt

ies

Maximum

Mean

Minimum

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 10

SensorRange = 25

Ca

su

alt

ies

Maximum

Mean

Minimum

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 25

SensorRange = 75

Ca

su

alt

ies

Maximum

Mean

Minimum

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 10

SensorRange = 25

Cas

ual

ties

Maximum

Mean

Minimum

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 25

SensorRange = 75

Ca

su

alt

ies

Maximum

Mean

Minimum

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 10

SensorRange = 25

Ca

su

alt

ies

Maximum

Mean

Minimum

0.001.002.003.004.005.006.007.008.009.00

10.0011.0012.0013.0014.00

SensorRange = 25

SensorRange = 75

Ca

su

alt

ies

Maximum

Mean

MinimumFIR

EP

OW

ER

-50%

+50

%B

AS

E

AWARENESS-50% BASE +50%

Impact of Sensor Range- For Maximum casualties SR had little effect

- Average casualties dropped slightly on either side

- Greater vision would allow greater use of weapon range i.e. Fighting in defensive posture over a greater range.

- Poor vision allowed to enemy to get in closer, but left enemy vulnerable to be shot before gaining a defensive posture (higher kill probability)

Tactical Scenarios: More “What If ?”-What if Major Smith had heeded the warning of the dead Vietnamese Regular and proceeded cautiously in a tight “One Up” formation? (One Up)

-What if Major Smith had decided to cover a larger search front with a “two up” formation? (Two Up)

-What if the American tactics of using trails had been accepted by the Australian 1RAR during their time with U.S. Army’s 173rd Brigade? (Trail)

-What if the Australians were quick to recognize the capabilities of APC’s and had used them employed them with D COY? (APC Trail)

- What if Major Smith saw the coming monsoon and decided to keep his formation within visual range? (Tight)

Tactical Formations:From the Australian Soldier’s Handbook Circa 1965

“One Up”

D COY HQ

10 Platoon

11 Platoon 12 Platoon

“Two Up”

D COY HQ

10 Platoon 11 Platoon

12 Platoon

“Column”10 PlatoonD COY HQ11 Platoon12 Platoon

Tactical Demonstrations

- LTBase1UP.dat “One Up”

- LT2UP.dat “Two Up”

- LTBaseTrail.dat “Trail”

- LTBaseTrailAPC.dat “Trail APC”

- LTBaseTight.dat “Tight”

* Nothing else besides the Tactics have been changed from the Base Case Scenario except for “Trail” where Australian stealth is reduced due to reduced cover

Altering Tactics

0

2

4

6

8

10

12

14

16

BASE 1 UP 2 UP Trail APC Trail Tight

Cas

ual

ties

Tactical Summary-“One Up” and “Two Up” scenario averages were close to or slightly less than the Base Case

-Range Variation due to the fact that tighter grouping may result in either greater concentration of Blue firepower or in greater concentration of Red forces

- “Trail” Scenario left the Australians too vulnerable and open to Ambush/ Annihilation

Tactical Summary- Addition of APC allowed for a lower minimum death toll than the “trail” scenario

- Disparity between Minimum and Maximum is because casualties become highly dependent on the behavior and focus of the APC squadron

-Average remains high because of troop vulnerability

- “Tight” scenario provides greater coordination and cooperation in the battle

Observations/ Lessons Learned

- In order to produce the least casualties in battle, Major Smith should have kept his units within visual range.

- Allowing units to act more independently of the knowledge of the HQ unit and training troops more also could have produced fewer casualties.

- Rain may not have been the deciding factor in the battle

Observations/ Lessons Learned

- Stealth protects the Australians the most

- Was stealth dependant on training, terrain, camouflage, or rain?

- APC’s effective as a rescue force once the Vietnamese were weakened, but if APC’s were ambushed in the initial assault, it is likely they would have been destroyed too

-Tactics were important in casualty levels

MANA Critique

- MANA v. 0.93 and v. 0.97 are INCOMPATIBLE

- MANA needs individual coordinate interface

- Global trigger state personality change option would be beneficial

- MANA should allow the user to select the desired MAP file instead of producing a MAP for each scenario. This would reduce disk space needed to hold scenarios.

- Possibility of some way to simulate artillery

- More variety of terrain which would automatically reduce movement speed and increase stealth

??QUESTIONS??

“Curiouser and Curiouser”