OODA Turn Engine

The 10-step OODA cycle that drives each simulation turn, from observation through action and feedback.

OODA Cycle — 10 Steps

Always automated Human or auto (mode-dependent) Always human in manual mode OBSERVE ORIENT DECIDE ACT OBSERVE 1 Partial Tactical Pictures + Global Indicators Fog of war (VC1/VC2 detection), control panels, logistics reinforcements, initiative determination AUTO 2 Staff Assessment / Situation Appreciation Human review in manual mode, automated analysis in auto mode H/A ORIENT 3 Strategic Decision — Maneuver Concept Choose maneuver concept: human decision or C2AUTO module (simplified strategic culture model) H/A 4 Translate to TABORD + INFOPS Orders Human with 15-min timeout, or auto-generated. Produces tactical and information operations orders H/A DECIDE 5 Disseminate FRAGOs down C2 Chain Apply signal loss based on VE3 (C2 connectivity). Some units may NOT receive orders if connectivity is degraded. AUTO ACT 6 Form Unit Groups + Execute Movements VC6 speed + terrain modifiers on H3 hex grid AUTO 7 Apply Local Orders + Fill Combat Table Test tactical behavior codes, resolve local orders AUTO 8 Resolve Combat (Monte Carlo) Attacker VC3 vs defender VC4 x terrain x uncertainty. Sub-turns: initiative holder attacks first, then defender AUTO 9 Determine Consequences Unit status, geographic occupation, population morale. Push infrastructure effects to MIDAS AUTO 10 Information Feedback to C2 Chain With signal loss. Update control panels. Check victory/defeat conditions. Loop back to Step 1. AUTO NEXT TURN

Three Simulation Modes

ASPECT MANUAL SEMI-AUTO AUTO Duration Days Hours Minutes Human Phases O, O, D, A (all 4 phases) O, D only (observe + decide) None (initial setup only) Order Refresh Every turn Every 5-20 turns At initial setup only Best For Decision training Doctrine development Staff exercises (CPX) Command post exercises Hot planning Rapid option testing
Historical basis. The OODA model comes from John Boyd (1976). It ensures the simulation mirrors how real military staffs operate — observe the situation, orient understanding, decide on a course of action, and act — making it a natural interface for wargaming. Each simulation turn maps directly to this cognitive loop, so players interact with the system the way they would with a real command post.

Victory / Defeat Conditions

The simulation checks three end-state conditions at the end of each turn (Step 10):

  • Will below threshold — the belligerent population's will to fight drops below a configurable minimum, triggering political collapse.
  • Capability below threshold — combat capability (remaining forces, logistics) falls below the level needed to sustain operations.
  • War objectives achieved — geographic or strategic objectives are met as evaluated by the arbiter module.

Signal Loss Model

The probability of an order not reaching a subordinate unit depends on VE3 (C2 connectivity value):

  • High VE3 — reliable communications, nearly all FRAGOs are delivered.
  • Degraded VE3 — telecom infrastructure damaged (via MIDAS cascade), increasing order loss probability.
  • Low VE3 — severe signal loss; many units operate on last received orders or default tactical behaviors.

This directly links MIDAS infrastructure modeling to tactical outcomes — destroying a telecom node degrades enemy C2.

Initiative Determination

Which belligerent attacks first each sub-turn is determined by comparing C2 capability scores:

  • Step 1 computes initiative as a function of each side's overall C2 capability and operational tempo.
  • Step 8 uses initiative to sequence combat: the initiative holder resolves attacks first, gaining a structural advantage.
  • Shift conditions — initiative can shift between turns if C2 capability changes (e.g., from infrastructure damage or reinforcement).
Limitation — Fully Automatic Mode. In fully automatic mode, the C2AUTO module uses a simplified strategic culture model. Real-world strategic decision-making is far more nuanced, involving political constraints, coalition dynamics, media pressure, and cognitive biases that cannot be fully captured algorithmically. Auto mode is for rapid option screening, not definitive analysis.