Jira progress: loading…

AGGR

Aggregation Engines

1. Description

Deterministic aggregation of validated computation outputs across scope, structure, and time

2. Overview

Aggregation Engines (AGGR) are deterministic micro-engines within the ZAYAZ Computation Hub. Their responsibility is to aggregate already-computed and validated outputs into higher-order, consolidated results.

AGGRs are first-class members of the Micro Engines (MICE) family and plays a critical role in:

• multi-level ESG roll-ups (asset → site → company → group),
• temporal consolidation (monthly → quarterly → annual),
• scope and category aggregation,
• audit-grade provenance and replayability.

AGGRs does not compute new primary values. They derive secondary results strictly by combining existing outputs.

3. Design Principles

AGGRs are governed by the following non-negotiable principles:

1. Determinism
   Given the same inputs and aggregation rules, AGGRs always produces the same output.

2. No Assumptions
   AGGRs never estimates, extrapolates, or fills gaps.

3. Post-Compute Only
   AGGRs operates exclusively on validated computation outputs — never on raw inputs.

4. Full Lineage Preservation
   Every aggregated result retains a complete reference to its source executions.

4. Scope of Responsibility

4.1. What AGGRs Does

AGGRs performs mathematical and structural aggregation across explicitly defined dimensions:

• Structural
  • asset → site
  • site → company
  • company → group
• Temporal
  • day → month
  • month → quarter
  • quarter → year
• Logical
  • scope-level consolidation (e.g. Scope 1 + Scope 2)
  • category-level roll-ups
  • portfolio or region summaries

4.2. Supported Aggregation Modes

Depending on the metric type and rule definition, AGGRs supports:

• Sum
• Weighted sum
• Average
• Weighted average
• Minimum / maximum (where explicitly allowed)
• Count

Aggregation rules are explicitly declared and versioned.

5. What AGGRs Does Not Do

AGGRs explicitly does not perform any of the following:

• ❌ Estimation or extrapolation (handled by SEM)
• ❌ Validation of raw inputs (handled by VALI)
• ❌ Unit transformation (handled by TRFM)
• ❌ Policy enforcement or routing (handled by TRPG / ZADIF)
• ❌ Narrative generation or interpretation (handled by ZARA)

This separation is critical for explainability and assurance.

6. Inputs

AGGRs consumes:

• A collection of validated computation output payloads
• Aggregation context, including:
  • scope or hierarchy definition
  • time window
  • aggregation rule reference
• Optional weighting parameters (e.g. revenue, area, production volume)

All inputs must conform to their declared output schemas.

7. Outputs

AGGRs produces:

• A consolidated output payload (schema-validated)
• Aggregation metadata, including:
  • aggregation rule ID
  • aggregation dimensions
  • engine version
• Provenance references:
  • list of source execution IDs
  • source method identifiers and versions

The aggregated output itself can be further consumed by:

• additional AGGR passes (higher-level roll-ups),
• reporting pipelines,
• verification workflows.

8. Audit & Provenance

Every AGGR execution generates an auditable aggregation record.

This record links:

• all source execution IDs,
• the aggregation rule definition,
• the AGGR engine version,
• timestamps and execution context.

This enables:

• full replay of aggregated results,
• verifier inspection,
• regulatory audit readiness.

AGGRs are fully integrated with ALTD / DAL logging.

9. Position in the Computation Hub

AGGRs sits after primary computation and validation in the execution chain:

Input → CALC → VALI → TRFM → (optional SEM) → AGGR → Reporting / Disclosure

This ordering ensures that aggregation is always applied to trusted, normalized data.

10. Canonical Identification

• Component ID: AGGR
• USO Code: AGGR
• Engine Type: Micro Engine (MICE)
• Status: Stable
• Layer: Computation Hub

AGGR identifiers are used consistently across:

• execution logs,
• provenance records,
• USO naming,
• explainability surfaces.

11. Design Rationale

By elevating aggregation to a dedicated engine, ZAYAZ ensures:

• clear semantic distinction between computed, estimated, and aggregated values,
• deterministic behavior suitable for assurance and verification,
• long-term scalability as reporting complexity increases.

Aggregation is not an implementation detail —
it is a semantic operation, and AGGR makes it explicit.

12. Related Components

• CALC — Primary computation engine
• VALI — Validation engine
• TRFM — Transformation engine
• SEM — Stochastic Extrapolation Module
• ZARA — Explainability and orchestration
• USO — Universal Signal Ontology

13. Example AGGRs

Example AGGRs and identifiers:

• MEID_AGGR01_v1 (Org hierarchy roll-up)
• MEID_AGGR02_v1 (Time-window aggregation)
• MEID_AGGR03_v1 (Intensity weighted average)

---

Version: 1.0
Status: Stable
Owner: Computation Hub / MICE