GESA × 6D Foraging

Dimension-Level Episodic Memory

In the 6D business context, GESA operates at the dimension level. Each of the six business dimensions maintains its own episode memory, and cascade paths are stored with full origin-to-propagation context.

---

The 6D Framework

The 6D Foraging Framework maps business events across six dimensions:

Dimension	Domain
D1	Customer
D2	People & Culture
D3	Revenue
D4	Regulatory & Compliance
D5	Quality
D6	Operational

A cascade analysis identifies which dimension originates an event and which downstream dimensions it propagates to. GESA stores not just the outcome, but the full cascade path.

---

Dimensional Episode Memory

interface CascadeEpisode extends Episode {
  originDimension:    string      // e.g., 'D3_Revenue'
  cascadePath:        string[]    // e.g., ['D3', 'D6', 'D1', 'D5']
  cascadeDepth:       number      // How many dimensions affected
  fetchScore:         number      // FETCH score of the triggering event
  fetchTier:          string      // 'EXECUTE' | 'CONFIRM' | 'QUEUE' | 'WAIT'
  dimensionScores:    Record<string, number>  // Per-dimension impact scores
}

This enables dimensional retrieval: not just "what happened in similar situations" but "what happened in D6 Operational cascades that reached D1 Customer."

---

Generating Dimension-Specific Recommendations

GESA generates intervention recommendations per dimension, ranked by historical effectiveness for that dimension type:

GESA.generate({
  dimension: 'D6_Operational',
  drift: 42,
  cascadePath: ['D6', 'D5', 'D1'],
  temperature: 0.73
})
→ [
    { strategy: 'Reduce WIP limits', confidence: 0.84, episodicSupport: 12 },
    { strategy: 'Add buffer capacity at bottleneck', confidence: 0.71, episodicSupport: 7 },
    { strategy: 'Cross-train adjacent team member', confidence: 0.58, episodicSupport: 3 }
  ]

The cascade path is a first-class retrieval dimension. An episode where D3→D6→D1 cascaded is more relevant to a current D3→D6 situation than an episode where D2→D4 cascaded — even if the DRIFT magnitude is similar.

---

The 8-Agent Orchestrator

In StratIQX, the 6D framework runs through an 8-agent AI orchestrator. Each agent produces an episode at its boundary:

Agent	6D Dimension	Episode Tag
Executive Summary	Cross-dimensional synthesis	synthesis
Financial Analysis	D3 Revenue	D3
Operational Analysis	D6 Operational	D6
Market Analysis	D1 Customer	D1
Strategic Recommendations	D1 + D3 + D6	multi_dim
Implementation Roadmap	D6 Operational	D6_execution
Risk Assessment	D4 + D5	D4_D5
Competitive Analysis	D1 Customer (external)	D1_external

8 episodes per report, each tagged to its dimension. Over hundreds of reports, GESA accumulates dimensional episode memory: not just "what happened in this run" but "what happened in D3 Revenue cascades across all runs."

---

Cascade Pattern Learning

GESA learns which cascade patterns are most predictive of downstream impact:

HighImpactCascades = GESA.analysePatterns({
  dimension: 'D3_Revenue',
  minEpisodicSupport: 10
})
→ [
    { path: ['D3', 'D6', 'D1'],   avgImpact: 72, frequency: 34 },
    { path: ['D3', 'D1'],         avgImpact: 58, frequency: 21 },
    { path: ['D3', 'D6', 'D5', 'D1'], avgImpact: 89, frequency: 8 }
  ]

This allows the framework to prioritize monitoring and intervention for cascade patterns that historically produce the most downstream impact — not just the ones with the highest origin DRIFT score.

---

→ GESA × HEAT