History

Since release R2.0SR1.00, MaRVs have been capable of adjudicating against ships, and can either sink it, or inflict a mobility or radar kill. In January 07, a change was integrated to allow MaRVs inflicting radar kills on ships to also kill aircraft located onboard. Unfortunately, there were some shortcomings with this implementation, most notably that the aircraft kills were not logged in the Adj KVS instrument, but only in the Adj A2G KVS. Corrections have been made via overlay and will be discussed below. This discussion won’t include details of the implementation of MaRVs themselves, but rather will cover MaRV adjudication against ships and their associated aircraft.

Target Selection

When a MaRV attempts to adjudicate, it will look inside its target area (radius determined by data) and will attempt to find the largest ship within that area. If there is a landmass within that area, the MaRV may be tricked by that landmass instead of adjudicating against the ships in the area. By design, there is a user-entered probability that MaRVs will be deceived by the land; however, as of R2.0SR2.01, that probability is used incorrectly: Currently, if there is any land within the search radius, MaRVs will be misled into missing the ships.

Once the MaRV detects its target, it will determine whether it is a maritime target or a land target. If it is a maritime target, the MaRV will use the Surface Adjudicator as shown in the figure below.

Target Adjudication

Once a ship is chosen for adjudication, the outcome can be either No Damage, Sink, Mobility Kill or Radar Kill. Only in the case of Radar Kill can aircraft onboard the ship be killed.

First, the number of aircraft onboard is determined by taking the number of aircraft in each squadron (accounting for the ones which have already been killed) and subtracting those which are currently airborne.

Aircraft are split into two groups: sheltered and unsheltered. Aircraft are considered sheltered up to the number of AssetAircraftShelters on the ship; any aircraft in excess of that are considered unsheltered. Of the number on board, aircraft are killed proportional to the PCT_SHELTERED_AIRCRAFT_KILLED and PCT_UNSHELTERED_AIRCRAFT_KILLED provided by user data, depending on whether they are sheltered or unsheltered.

Kills are deterministic, found by multiplying those factors by the number of aircraft in that condition, and are rounded up. Kills are then applied by multiplying those factors by the number onboard for each squadron (in alphabetical order), again rounded up. This successive rounding up and applying to squadrons in the same order every time will penalize those squadrons earliest in alphabetical order. In the example below, the first figure shows the total “number to kill” for sheltered/unsheltered aircraft. That total number of kills is then applied to each squadron in the second figure.

Total Number To Kill

MaRV Kills Applied to Squadrons

Looking at Squadron_A, it has 9 aircraft on the ship at the time of adjudication. 75% of them (rounded up to 7) are unsheltered. Since 50% of unsheltered aircraft are killed by the MaRV, that is rounded up to 4. It has 2 aircraft sheltered, and 25% of them killed rounds up to 1. You may expect Squadron_C and D to also have 1 sheltered aircraft killed, but since there are only 2 total sheltered kills, they run out after Squadron_B.

Due to rounding the number of kills up, it is possible for more aircraft to be killed than should have been. Finally, there is no stochasticity to it, in terms of the number or type of aircraft killed (other than the last squadron once all the kills have been recorded by rounding); the kills are always applied in alphabetical order by squadron ID.

Design Shortcomings

• Aircraft can’t be killed on a ship by a mobility kill; only for a radar kill or sink
• Aircraft kills are instrumented only in Adj A2G KVS, not in Adj KVS
• Kills are assigned alphabetically by squadron Air Unit ID
• It is possible to kill more aircraft than exist
  • When a squadron has 0 aircraft killed (ex. due to rounding), there is a random draw to determine whether the squadron in fact had 1 killed
    • Though counter intuitive, the logic is this: If there are 3 squadrons of F18Cs, and 1 aircraft is killed by the MaRV, that leaves a 0.33 probability of any particular squadron losing an aircraft. Since that would round down to 0 for each squadron, as the algorithm goes through the list of squadrons it does a random draw until it picks the losing squadron
      • Since there is no check to confirm that the squadron actually has aircraft remaining, more can be killed than were possessed
  • Kills are reported with holes in the instrumentation
  • A2G ID is always reported as “1”
    • ATO ID for the MaRV is always reported as “1”
  • Kills for sinks aren’t instrumented at all (for aircraft, or embarked land units or supplies)
• It is not clear which aircraft are sheltered; that is, which squadrons or in what order the aircraft are sheltered
  • There is no way to assign high value aircraft to shelters
    • Aircraft killed are not instrumented as being sheltered or unsheltered

SAC Overlays

Several enhancements have been implemented in the SAC via overlays:

• Aircraft killed by MaRVs are reported in the Adj KVS
• Corrected “more can be killed than were possessed” issue by only doing the random number draw if the squadron has aircraft remaining
• Change to Adj KVS to track resources killed when a ship is sunk:
  • One line in the Adj KVS will report the kill of the ship platform itself (killer, victim, Asset ID, Kills = 1, etc)
    • Subsequent lines in the Adj KVS will report the assets the ship BSE had or was carrying which have been killed as well (troops, trucks, radars, aircraft, etc.) The ship platform itself has been filtered out to not report twice.
    • Note that this overlay change applies only to ships sunk (assuming that everything onboard was killed), but not to radar or mobility kills (since adjudication data doesn’t support what resources were killed).
  • Minor improvements (capitalized ABL and ASW)