Ammunition Reload Requirements of LAV Ammunition Reload Requirements - - PDF document

1

Defence Research and Development Canada Recherche et développement pour la défense Canada

Canada

Geoff Pond, Ph.D., P.Eng. Land Capability Development OR Team

Ammunition Reload Requirements of LAV Primary Weapon System Options Ammunition Reload Requirements of LAV Primary Weapon System Options

2

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Outline

• Introduction

– Current Medium-Weight Vehicle Capability – Weapon System Alternatives

• Reload Requirements

– Binomial Statistics – Monte Carlo Simulation – Wargaming

• Conclusions

3

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Light Armoured Vehicle (LAV) – III with Delco Turret

Entered in Service: 1998 Crew: 3 Dismounts: 7 Weapons: 25mm Dual Feed Cannon 7.62 Coaxial Machine-Gun 7.62 Pintle Mounted Machine-Gun

The third generation light armoured vehicle (LAV III) is the cornerstone of Canadian Army mounted operations. The LAV III is a wheeled infantry fighting vehicle as shown on this slide. This vehicle is crewed by three soldiers: a driver, gunner, and vehicle commander. In addition, seven soldiers occupy the rear passenger compartment. These soldiers may be dismounted to deliver more firepower on an opposition force, if required.

4

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

LAV III Inventory

33 LAV Less Kits (LKs) being converted to ISC w/ Nanuk Remote Weapon Stations 310 Infantry Section Carrier (ISC) Variant + 180 Command Post Variant + 50 Forward Observation Officer Variant 540 LAV IIIs with Delco Turrets Currently, approximately 540 LAV III vehicles exist in the army inventory. The majority of these vehicles entered into service in 1998. To date, they have endured ten years of intense use, both during training and military operations. To extend its life and upgrade the vehicle to perpetuate its battlefield significance, a mid-life refit is currently being planned. As part of this refit, consideration is being given to replacing the vehicle’s turret. The Delco turret which is mounted on the vehicle houses a 25mm cannon, a co- axial 7.62mm machine gun, a pintle-mounted 7.62 machine gun, and two clusters of four 76mm smoke grenade launchers. The vehicle’s primary weapon system is the 25mm cannon which can shoot either in single-shot, or burst

• modes. Various 25mm ammunition types are fired by this weapon system but
• nly two varieties are commonly employed on operations. To defeat lightly

armoured or hardened targets, an armour-piercing fin stabilised discarding-sabot (APFSDS) round is employed. The alternative is a High Explosive Incendiary (HEI) round where the fuze detonates on impact and distributes an incendiary compound which burns flammable materials within a 5m radius. Use of both rounds can be accommodated by a primary and secondary ammo bin which feed the cannon. For example, the primary bin may be loaded with HEI and the secondary bin loaded with APFSDS rounds. Depending on the target, the gunner chooses the type of round that is most appropriate, by the flick of a switch. The primary bin contains 150 ready-rounds and the secondary bin contains 60 ready-

• rounds. When all ammunition has been expended, the ammo bins are restocked

by the turret crew from inside the vehicle.

5

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

LAV III – Nanuk RWS

• 33 LAV LK’s ISC
• mounts a .50 cal machine-gun
• first unit delivered, full delivery expected fall 2009
• to be deployed on Taskforce 1-10 as the Force

Protection unit for the Provincial Reconstruction Team Nanuk RWS LAV III - Nanuk Recently, a limited number of Canadian LAV IIIs have been fitted with a remote weapon system named the Nanuk. An image of this system mounted on the vehicle is shown on this slide. A remote weapon station such as the Nanuk is controlled remotely by crew from under the protection of the vehicle armour. The major disadvantage of this type of system is that reloading the system is done from outside the vehicle, exposing the crew.

6

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Study Question

Given the ammunition usage and reload characteristics of each vehicle type, how many vehicles within a platoon will be available to engage opposition forces at any given time?

7

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Op Medusa

Strength Casualties Captured 2,000 1,200 12 512 136 Longest Threat Period 4 hrs Largest Ammo Expenditure 600 rounds A threat period of 4 hours with 600 rounds expended per vehicle was selected in

• rder to replicate the worst case scenario as described by returns from an inquiry

made in theatre through the Army Lessons Learned Centre (ALLC). For this engagement, friendly and opposition force strength and casualties are detailed in this slide’s table.

8

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

• No mechanical failures (stoppages, jams, etc)
• No casualties
• Reload requirements are randomly distributed throughout

engagement

• No ammunition management

Simplifying Assumptions

The following slides detail a binomial probability model for the reload of the LAV III having either a Delco turret or a RWS. This basic model requires several simplifying assumptions. The most relevant of those assumptions are detailed on this slide. In addition to those, the following assumptions are also made:

• The threat period is 4 hours
• A LAV with RWS must withdraw 1km at a speed of 60km/hr prior to reloading
• A LAV with RWS requires 10 minutes to reload
• A turreted LAV reloads in place
• A turreted LAV requires 2 minutes to reload
• Each vehicle expends 600 rounds of ammunition during the engagement
• The time for a withdrawn vehicle to regain situational awareness and reacquire

a target when returning to the engagement is negligible

• The ammo expenditure rate of each vehicle is independent of that of all other

vehicles

9

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Reload Time

• RWS must have comparable calibre
• RWS based on the ARX20 (Nexter)
• 20mm
• 100 round ammo box
• Assumed to fire at same rate as 25mm

Delco

• Must withdraw 1km away from

engagement in order to reload (to provide sufficient protection to gunner)

Some RWS include a 20mm or 25mm cannon, which is more similar to that currently part of the Delco turret. The ARX20 20mm RWS ammo bin fits 100 ready-rounds. Unlike the Delco turret, the weapon system is fed by a single ammo bin. This is typical of most RWS having a 20-to-25mm calibre cannon. Therefore, once 100 rounds have been fired, the system must be reloaded from

• utside the vehicle. To reload, the vehicle is expected to withdraw from an

engagement to an area were the crew can safely exit the vehicle to access the RWS ammo bin.

10

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Reload Probability

% 5 . 2 min 240 ) reload min/ 2 )( reloads 3 ( = =

turret

p

% 25 min 240 ) reload min/ 12 )( reloads 5 ( = =

RWS

p A threat period of 4 hours with 600 rounds expended per vehicle was selected in

• rder to replicate the worst case scenario as described by returns from an inquiry

made in theatre through the Army Lessons Learned Centre (ALLC). The 2 minute reload time for the turreted LAV is the standard reload time provided that a few rounds from the previous upload remain so that the new upload can be linked to it. Ten minutes to reload an RWS seemed appropriate provided by feedback from the ALLC and in consultation with staff who represent Army users in the development of future equipment acquisition. The probability of one of the vehicles reloading, p can be calculated as follows. For the Delco turret, 150 rounds can be uploaded to the primary bin. Assuming the vehicle's primary bin is full at the onset of the threat period, the vehicle requires three more uploads to reach the 600 round expenditure assumed to take

• place. The amount of time spent reloading is: 3 uploads × 2 minutes per upload

= 6 minutes. Recall that the threat period is assumed to last four hours. The fraction of time spent reloading is therefore 6 minutes divided by 240 minutes = 0.025, i.e., 2.5%. This can be interpreted as “each LAV III having a Delco turret will spend 2.5% of the threat period reloading the ammo bins”. This process is repeated for the specifications of the RWS.

11

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Calculations

x N x

p p x N x P

−

−

• =

) 1 ( ) (

( )

1 1 = ∪ =

R D

R R P

( )

( )

1 1 1 1 = ∩ = − = ∪ =

R D R D

R R P R R P

The binomial probability distribution may be used to model the expected number of vehicles available to engage targets. This model is based on a binary response indicating whether a vehicle is in the process of reloading, or not, where: N denotes the number of vehicles (N = 4 for a platoon) x denotes the number of vehicles reloading p denotes the probability of one of the vehicles reloading Special caution must be taken when performing the calculations for a platoon having a mix of RWS vehicles and Delco Turret vehicles. In the Venn diagram

• n the left side, the union probability depicts the cases where either a) a vehicle

with a Delco turret is reloading, b) a vehicle with an RWS is reloading, or c) both a vehicle with a Delco turret is reloading and a vehicle with an RWS is

• reloading. Since case c) corresponds to the scenario where two vehicles are

reloading (not just one), we must remove this intersection area as depicted in the Venn Diagram on the right side.

12

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Vehicle Availability

Number of Vehicles Reloading (Binomial Dist.)

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% LAV-Delco LAV-RWS 50/50 Mix 0 Vehicles Reloading 1 Vehicle Reloading 2 or More Vehicles Reloading

Number of Vehicles Reloading (Binomial Distribution) Number of Vehicles Reloading (Binomial Distribution) Number of Vehicles Reloading (Binomial Distribution) Number of Vehicles Reloading (Binomial Distribution)

These are the results of probabilities as determined using the binomial probability distribution. As shown, the baseline LAV-Delco platoon could expect to be short 1 vehicle approximately 10% of the time while that vehicle

• reloads. Conversely, if the platoon consists entirely of RWS, the platoon can be

expected to be short a minimum of 1 vehicle throughout the majority of the threat period. The platoon will be short 2 vehicles (at half capacity) for more than 20% of the time. A 50/50 mix of RWS / Delco turrets results in a blend of the two extremes.

13

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Exponential Cumulative Distribution Function

CDF for Reloading

0.2 0.4 0.6 0.8 1 20 40 60 80 100 120 140 160 180 200 220 240 Time (minutes) Cumulative Probability LAV-Delco (=60) LAV-ARX20 (=40)

Results based on a binomial probability distribution are a reasonable first

• estimate. However, the fidelity of the results may be improved by running

Monte Carlo simulations. A problem constraint is missing when using the binomial distribution. Use of this distribution, as described, assumes that the time to reload can be parsed into one-minute intervals randomly distributed throughout the threat period. Of course, in reality, this is not the case. Instead, reloads are done over a continuous time period. For example, reloading a Delco turret requires two sequential minutes – not two minutes randomly distributed throughout the threat period. If a Monte Carlo simulation were used instead, this constraint can be included in the program formulation. Monte Carlo simulations operate by random sampling of a predefined probability distribution. If the sampled pseudo-random number is below/above a specific threshold, some event occurs. For example, in this case, a random number is drawn from a distribution which indicates the number of time-steps until the vehicle is expected to reload. Often, the probability of machine survival is modeled as an exponential decay. It is assumed that this is also an appropriate representation of the reload

• probability. In addition, the exponential distribution is continuous (rather than a

discrete distribution such as the binomial model) and is therefore more amenable to a Monte Carlo simulation over a continuous time scale. Using the analogy between machine life and ammunition expenditure, the figure shown here depicts the cumulative probability that either the LAV-Delco or the LAV- ARX20 will require reloading, as function of time. As expected, the probability that the LAV-ARX20 requires reloading is slightly higher. The plot can be interpreted as follows: after fighting for 20 minutes (along the x-axis), there is a 40% probability (along the y-axis) that the ARX20 will require reloading.

14

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Coded Gene of Platoon Vehicle Availability

Vehicle is in the reload process Vehicle is available to engage opposition forces The results of a single Monte Carlo simulation are depicted here. In this figure, each row of the array corresponds to a vehicle. The top two rows correspond to LAVs having Delco turrets whereas the bottom two rows correspond to LAVs having an RWS. Where cells of the array are shaded in green, the vehicle is engaging the enemy. Where cells are shaded in red indicate the vehicle is either reloading or withdrawing in order to reload. Each cell of the array corresponds to a single minute of the simulated threat period.

15

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Vehicle Availability

Number of Vehicles Reloading (Monte Carlo Sim.)

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% LAV-Delco LAV-RWS 50/50 Mix 0 Vehicles Reloading 1 Vehicle Reloading 2 or More Vehicles Reloading

Number of Vehicles Reloading (Monte Carlo Simulation) Number of Vehicles Reloading (Monte Carlo Simulation) Number of Vehicles Reloading (Monte Carlo Simulation) Number of Vehicles Reloading (Monte Carlo Simulation)

Running the simulation only a single time is not sufficient to obtain reliable

• results. For example, the expected value for the number of reloads of a LAV

with Delco turret is four. Neither of the LAVs in Figure 6 reload the expected number of times. In MATLAB, on a PC, one thousand iterations of the simulation can be completed in less than 0.2 seconds. During each iteration, time series histories such as that shown in Figure 6 are recorded. At the end of one thousand iterations, the mean amount of time where 0,1,2,… vehicles of the platoon are reloading is calculated. The results are displayed here. The resemblance to the figure shown earlier based on the simpler binomial approximation is clear.

16

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Experiment: Wargame Options

The Canadian military employs virtual simulations which satisfy higher standards of verification, validation and accreditation (VV&A). The physics engines behind these programs have been carefully tested to assure the results produced in the game are an appropriate reflection of real life events. Currently, Virtual Battlespace 2 (VBS2), is the virtual simulation of choice employed by the Army. A screenshot of a platoon of LAV IIIs is depicted here. In this type of serious game, each entity in the game can be controlled by a real

• player. A real LAV driver can play the role of driving the LAV using a system
• f steering wheels and pedals. The view that player would have on their monitor

is from the first person perspective. Communication between players can be done through radio headsets as in reality. Furthermore, to closer replicate reality, white noise can be injected over the communications network to further enhance the realism. In a simulation such as this, vehicles have the opportunity to work cooperatively to engage a single target, i.e., ammunition expenditure rates are no longer assumed to be independent. The fidelity of results would be further increased by incorporating simulated vehicle kills, mechanical failures, friendly dismounted soldiers, and the influence of human factors (e.g., confusion, panic, and fog of war).

17

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Experiment

Strength Vehicles Dismounts 4 × LAV III 1 × BMP2 28 7 Longest Threat Period 1 hour For this experiment, friendly and opposition force strengths are detailed in the table above. The opposition force was told to protect specified strong points. The friendly forces were told the general location of the red force and were instructed to advance to contact, then close-in and destroy enemy forces. The terrain was a suburban area surrounded by desert.

18

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Wargame Experiment

Operators during play After-Action Review / video replay

Following each of the trial runs of the experiment, an after-action review was conducted in a theatre setting. This allowed players to obtain perfect awareness

• f how the engagement played out.

19

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Wargame Experiment

Crew Commander (CC) hatches up in turret Gunner looking through day sight CC looking through viewports from inside the turret

In practice, the crew commander typically operates with his head outside the turret hatch. This allows the crew commander to perform quick direct visual

• bservation of his surroundings. To replicate this in simulation, the crew

commander was equipped with three side-by-side monitors to provide a wide field-of-view. In addition, the crew commander wore a motion tracking headset. As the crew commander looked left or right, the image on the monitors would pan accordingly.

20

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Extrapolated Results

Number of Vehicles Reloading (Empirical)

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% LAV-Delco LAV-RWS 50/50 Mix 0 Vehicles Reloading 1 Vehicle Reloading 2 or More Vehicles Reloading

Each of the vignettes considered here, although similar in nature, required varying lengths of time to complete. The results were then extrapolated to a 4- hour time period for comparison to the results obtained earlier through simpler

• methods. The results of this extrapolation are depicted above.

21

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Summary

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% LAV-Delco LAV-RWS 50/50 Mix 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% LAV-Delco LAV-RWS 50/50 Mix

Method Investment Results

Binomial Statistics Monte Carlo Simulation Virtual Wargaming 1 × Junior Analyst (1 wk) 1 × Junior Analyst (2 wks) 1 × Junior Analyst (2 wks) 8 × Contracted Interactors (1 wk) 3 × Simulation Support Staff (1.5 wks) 39 × Enlisted Soldiers (1 wk) 1 × Jr. Officer (1 wk)

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% LAV-Delco LAV-RWS 50/50 Mix

The investment required to perform the binomial approximation or Monte Carlo simulation is very modest. Conversely, the investment required for a platoon- sized virtual simulation, is much greater. The investment in personnel costs are

• utlined in the summary table depicted on this slide. In total, the virtual

simulation required approximately 2000 personnel hours of work. The binomial model required approximately 40 personnel hours and the Monte Carlo model required approximately 75 personnel hours. In addition to human resources, the virtual simulation also requires the infrastructure to support it, i.e., a large simulation space; computer hardware; and software licenses.

22

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Conclusions

• Analysis using basic methodologies and

simplifying assumptions requires limited staff and resource investments

• Conversely, running high-fidelity wargames is

resource-intensive

• Limited improvements in the fidelity of ammo

expenditure results are obtained by running wargame experiments (results typically vary by <10%)

23

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Questions?

24

Defence R&D Canada – CORA • R & D pour la défense Canada – CARO

Contact Information

Geoff Pond Geoff.Pond@forces.gc.ca 1-613-541-5010 x.2469 Centre for Operational Research and Analysis Canada