← Back to catalogue
Published

Dependencies and Impact Analysis

EM-WRK-06 · em-wrk-06

Support change decisions by identifying direct and permitted downstream impacts while making evidence limits and missing dependency coverage visible.

Enterprise profiles Enterprise subject EnterpriseWRK

Bundle → Layer → Finding → Questions Filled

2 bundles · 2 layers · 4 findings · 11 questions

Dependency edges This bundle describes directed dependencies and the rules governing their interpretation.

Type and direction

This layer identifies dependency meaning and permitted traversal.

Typed dependency

This finding records the meaning, scope and evidence of one dependency.

  1. Which dependency type does this edge record?
  2. Which object depends on which other object within the scope and validity period of this edge?
  3. What evidence and attribution support this edge?

Type rules

This finding records permitted traversal and composition for one dependency rule.

  1. Which direction and conditions does this rule permit for assessing direct impact through one edge?
  2. Which ordered edge-type combinations does this rule permit for repeated or mixed-type chaining?
  3. Under which conditions does this rule permit a cycle?
Impact analysis This bundle describes potential impacts and the limits of an analysis.

Impact scenario

This layer relates a trigger to a bounded assessment of affected objects.

Impact set

This finding separates direct impacts from impacts reached through permitted chains.

  1. Which trigger, scope and graph snapshot does this impact scenario assess?
  2. Which objects does this impact scenario identify as potentially affected through direct dependencies, including non-transitive types?
  3. Which additional objects does this impact scenario reach through chains supported by explicit composition rules?

Completeness

This finding states the coverage limits of one impact scenario.

  1. Which unknown or unverified relationships are recorded for this impact scenario?
  2. How do the recorded coverage gaps limit the conclusions of this impact scenario?

Classifiers Filled

Family
Enterprise profiles
Category
Enterprise subject
Entry kind
subject
Domain
EnterpriseWRK
Industry
Cross-industry
Tags
EM-WRK-06W2subjectDependencyDependencyTypeCriticalityAssessmentImpactScenarioDependencyEvidence

What it is Filled

Typed cross-object dependencies, prerequisites and constraints with scope and evidence. Does not imply automatic causality.

Why it exists Filled

Support change decisions by identifying direct and permitted downstream impacts while making evidence limits and missing dependency coverage visible.

Distinguishing features Filled

  • Every dependency has a type and direction, with rules for direct traversal, repeated chaining, mixed-type composition and cycles.
  • A direct dependency can indicate a potential impact even when its type does not permit transitive chaining.
  • A path in the dependency graph is not proof of causation, and causal claims require their own evidence.
  • The absence of an edge shows only that nothing was recorded and does not establish independence.
  • Schedule links can express prerequisites, while this model also covers dependencies across other objects and scopes.

What robots and AI may and may not do Filled

Must not

  • Declare any path in the graph a proven causal chain.
  • Treat a missing edge as evidence of independence.
  • Exclude a potential direct impact solely because its dependency type is non-transitive.
  • Infer permission for mixed-type chaining from the transitivity of individual edge types.
  • Record a dependency without type and direction.
  • Present an impact set as complete when parts of the graph are unknown.

Only with a human decision

  • A responsible human must decide whether to proceed with a change whose impact set is incomplete.
  • A responsible human must approve using a causal claim as the basis for blame or corrective action.
  • A responsible human must approve changes to dependency traversal and composition rules.

May

  • Assess potential direct impacts through relevant dependencies, including non-transitive types, in the direction permitted by the applicable rule.
  • Extend an impact assessment through chains only when explicit rules permit the ordered edge-type combinations.
  • Report the coverage and unknown areas of the graph with every impact answer.
  • Flag cycles that the applicable dependency rules do not allow.
  • Ask owners to confirm unverified dependencies.

Moral aspects Filled

  • Overstated causal chains can wrongly assign blame to teams or people.
  • Hidden dependencies can lead to outages or missed obligations that harm customers.

Who is affected

  • Owners of dependent systems and projects
  • Teams named in impact or causal analyses
  • Customers of affected services

Owners Filled

Steward

The project management office or execution lead who answers for dependency records and impact analyses.

Master systems

  • Task tracker
  • Project portfolio management system
  • Plans and acceptance records

Links to other meta-models Filled

neighbor

  • EM-TEC-01
  • EM-TEC-04
  • EM-WRK-01
  • EM-WRK-02

references

  • WM-XCT-037 - conceptual-candidate

What else AI and robots need to interact with it Filled

Identity and identifiers required Filled

  • A dependency is identified by source object, target object, type and validity period.
  • An impact scenario is identified by its trigger, the graph snapshot and the time of analysis.

Direct properties not applicable Not applicable

Not applicable

Enterprise record contour: physical properties belong to referenced world models.

Recognition optional Filled

  • A dependency record has a source, a target, a type, a direction and evidence.
  • Often confused with a schedule link between tasks or with a causal finding from an incident review.

Capabilities and actions required Filled

  • Impact sets can be recomputed when the graph changes.
  • Different triggers yield different impact sets over the same graph.

Hazards and failure modes required Filled

  • False confidence from impact answers that hide unknown edges.
  • Explosion of results when all edge types are chained transitively.
  • Stale edges that no longer reflect the systems or plans they describe.

Standards and interfaces required Filled

  • W3C RDF and SHACL for typed edges and constraints on them.
  • W3C PROV-O for provenance of dependency evidence.
  • ISO 8601 for validity periods of dependencies.

Context of use required Filled

  • Used in change planning, release coordination, portfolio planning and incident review.
  • Typical sources are trackers, architecture repositories and configuration databases.

Sources Filled

  1. PMI: project/program/portfolio
  2. Process-based and flow-based organization of work; compare Scrum/Kanban against primary sources
  3. Tracker object practice: container, issue, workflow and binding to the business initiative
  4. The Scrum Guide - Checked live (HTTP 200) on 2026-10-07.
  5. The Kanban Guide - Checked live (HTTP 200) on 2026-10-07.

Open questions

  • Which edges permit transitivity?
  • How are blocking, usage, funding and a causal claim distinguished?
  • How is graph incompleteness accounted for in impact analysis?
  • Установить границу и решение reuse/extend/new по действующим спецификациям.
  • Подтвердить semantic crosswalk, права и source mastership.
  • Выбрать immutable refs; провести проверки fixtures до заявления о публикационной готовности.

Machine files

Provenance

enterprise research programme · published-partial

Built from: enterprise/models/em-wrk-06/brief.json, ver-cy/world-models/card-supplements/em-wrk-06.json

Still in Russian: suggested owner, blocking decisions, vercy candidates. Translate in research/enterprise/i18n/units.en.json.