SEA™ Case-Based Project Management Handbook

Purpose & Context

SEA™’s case-based PM adapts CMMN (Case Management Model and Notation) for knowledge-intensive work. Unlike prescriptive workflows (Waterfall/Agile), cases enable discretionary, event-driven orchestration with AI agent support.

Actionable: Each section answers “How do I manage knowledge work?” with CMMN primitives, artifact pipelines, and PM-Agent integration.

1 CMMN Fundamentals

1.1 Core Primitives

1.2 Case Lifecycle

1
2
3
4
5
6
7
Created → Active → Completed/Terminated
    ↓
  Stages (parallel/sequential)
    ↓
  Tasks (mandatory + discretionary)
    ↓
  Sentries (entry/exit gates)

1.3 Discretionary vs Mandatory

Mandatory Tasks: Must execute (e.g., “Submit final report”)
Discretionary Tasks: Activate based on events (e.g., “Additional research” if hypothesis unclear)

2 Artifact Pipeline

2.1 Four Reification Stages

1
2
3
4
5
6
7
8
9
10
11
12
13
┌─────────────────────────────────────────────────────────────────┐
│                    ARTIFACT PIPELINE                            │
├─────────────────────────────────────────────────────────────────┤
│                                                                 │
│  Cognitive Artifacts (notes, sketches)                          │
│         ↓ Synthesis                                             │
│  Intellectual Artifacts (specs, SOPs, concepts)                 │
│         ↓ Refinement                                            │
│  Information Products (software, campaigns, reports)            │
│         ↓ Capitalization                                        │
│  Intellectual Capital (enterprise assets, reusable IP)          │
│                                                                 │
└─────────────────────────────────────────────────────────────────┘

2.2 Stage Characteristics

2.3 Transition Tokens

Every artifact progression generates a TransitionToken:

Flow "Synthesis" from "CognitiveArtifact" to "IntellectualArtifact"
@rationale "Raw notes become formal specifications"

Provenance Gating: No artifact can skip stages (“teleportation” prevented).

3 PM-Agent Orchestration

3.1 PM-Agent Role

The Project Manager Agent is a built-in AI conductor that:

• Assigns tasks to qualified CaseMembers (human or AI)
• Evaluates sentry conditions for stage transitions
• Manages cognitive ergonomics (workload balancing)
• Integrates with Knowledge Base for contextual decisions

3.2 Hybrid Teams

3.3 Task Assignment Logic

1
2
3
4
5
1. Identify available tasks (sentries satisfied)
2. Match task requirements to member capabilities
3. Check SDS-031 authority rules
4. Assign to qualified member
5. Log assignment to IFL

4 Case Lifecycle Management

4.1 Case Creation

Steps:

1. Define desired outcome
2. Select case template (research, development, consulting, creative)
3. Identify initial CaseMembers
4. Configure PM-Agent with Knowledge Base context
5. Activate initial stage

Example (Research Case):

1
2
3
4
5
Case: "Evaluate ML Framework Performance"
Desired Outcome: "Comparative analysis with recommendations"
Template: research_case_pattern
Initial Stage: Literature Review
PM-Agent Context: ML frameworks, performance metrics

4.2 Stage Progression

Entry Sentry: Condition to activate stage
Exit Sentry: Condition to complete stage

Example:

1
2
3
Stage: Analysis
Entry Sentry: "All data collected AND quality validated"
Exit Sentry: "Findings documented AND peer reviewed"

4.3 Task Activation

Mandatory: Automatically active when stage starts
Discretionary: PM-Agent activates based on events

Example:

1
2
Mandatory: "Conduct statistical analysis"
Discretionary: "Additional data collection" (if initial results inconclusive)

5 Knowledge Work Patterns

5.1 Research Pattern

5.2 Development Pattern

5.3 Consulting Pattern

6 Provenance and Governance

6.1 TransitionToken Generation

Every state change generates an immutable token:

1
2
3
4
5
6
7
8
9
PlanItemTransitionRecord {
  caseId: "case-123"
  planItemId: "stage-analysis"
  fromState: "Available"
  toState: "Active"
  timestamp: "2026-01-04T21:00:00Z"
  actor: "pm-agent-001"
  transitionToken: "tt-abc123"
}

6.2 IFL Integration

All TransitionTokens logged to Internal Federated Ledger for:

• Immutable audit trail
• Provenance verification
• Capital valuation
• Compliance evidence

6.3 Governance Policies

Policy "NoTeleportation" per Constraint Obligation priority 10
@rationale "Artifacts require lineage"
as: forall p in resources where p.name = "ProjectionToken":
    exists t in resources where t.name = "TransitionToken"

Policy "SovereignGate" per Constraint Obligation priority 10
@rationale "Capital conversion requires human approval"
as: forall r in flows where r.name = "Capitalization":
    exists s in resources where s.name = "SignatureToken"

7 Implementation Roadmap

Phase 1: CMMN Foundation (Weeks 1-2)

• Team training on CMMN primitives
• Select case template
• Create first pilot case

Phase 2: Artifact Pipeline (Weeks 3-4)

• Define artifact types per stage
• Configure TransitionToken generation
• Establish Exchange Rate Policies

Phase 3: PM-Agent Integration (Weeks 5-6)

• Configure task assignment rules
• Define custom sentries
• Integrate Knowledge Base

Phase 4: Scale (Ongoing)

• Create domain-specific templates
• Refine sentry patterns
• Measure artifact reuse and capital formation

Conclusion

This handbook delivers a unified, actionable case management system that leverages SEA™’s CMMN implementation. By treating knowledge work as discretionary and event-driven, organizations achieve flexible orchestration, artifact capitalization, and AI-augmented project management.