2.1 Activity Definition and Logic Relationships

2.1 Activity Definition and Logic Relationships

AACE RP 23R-02 concerns identification of activities, and RP 24R-03 concerns development of activity logic. Together they support a schedule that represents the execution plan in measurable work and defensible dependencies.

Define useful activities

An activity should have a clear scope, responsible party, identifiable start and finish, appropriate duration, calendar, and status evidence. It should be detailed enough to measure and forecast but not split merely to increase activity count. A milestone has zero duration and marks an event such as notice to proceed, design approval, delivery, system turnover, or completion.

Activity identification draws from the WBS, estimate, work packages, procurement plan, contract milestones, stakeholder interfaces, methods, and phase plan. Include enabling work such as approvals, access, inspections, testing, and turnover when those steps can control downstream work.

Relationship endpoints in PDM

Precedence Diagramming Method relationships connect activity endpoints:

• Finish-to-Start (FS): successor start is constrained by predecessor finish.
• Start-to-Start (SS): successor start is constrained by predecessor start.
• Finish-to-Finish (FF): successor finish is constrained by predecessor finish.
• Start-to-Finish (SF): successor finish is constrained by predecessor start; this is uncommon and should be checked carefully.

An SS relationship may protect the start of an overlapping process, while an FF relationship protects its finish. Used together, they can model a production ladder. Do not judge the pair only by asking whether an FS link exists. Review whether both the start and finish endpoints are adequately constrained and whether the overlap is physically feasible.

Logic categories

Physical or technical logic follows the method of work: a foundation must support steel, and an energized system may require completed testing. Contractual or external logic reflects approvals, access, permits, owner-furnished information, or third-party interfaces. Preferential logic represents a chosen sequence that could be changed, such as moving a crew area by area. Resource limits may be modeled through explicit crew-flow logic, resource-constrained scheduling, or leveling, depending on the model purpose. Document the method so users know which relationships are changeable during optimization.

Lags and leads

A positive lag can represent a real waiting period, but a separate activity is usually clearer when the period has scope, responsibility, calendar, or status of its own. A lag labeled “cure seven days” hides whether curing began, whether the correct calendar applies, and who confirms completion.

Negative lag allows a successor endpoint to occur before the referenced predecessor endpoint. Some specifications prohibit it, and many quality checks flag it because it can hide the point at which overlap becomes possible and complicate updating. It does not make every schedule mathematically impossible, but activity decomposition or SS/FF logic is generally more transparent. Follow the governing specification and document justified exceptions.

Arrow diagrams and dummy activities

In Arrow Diagramming Method, arrows represent activities and nodes represent events. A zero-duration dummy arrow may be needed to preserve unique activity identity or express logic that cannot otherwise be shown without creating a false dependency. The dummy consumes no time or resources. Modern PDM generally avoids this device because activities are nodes and relationships are explicit.

Network review

Check for circular logic, missing interfaces, open endpoints, redundant links, excessive lags, invalid actual dates, and unexplained constraints. Many specifications expect every non-start activity to have an appropriate predecessor and every non-finish activity to have an appropriate successor, but external-interface milestones or intentionally bounded subnetworks may require documented treatment. A diagnostic flag begins the review; it does not replace professional judgment.

Trace important milestones backward through continuous driving relationships and then forward from major inputs. Ask whether a delay to a predecessor produces the expected effect and whether every material handoff is visible. The objective is an explainable execution model, not mechanical compliance with one relationship type.

Applied review: test both endpoints

For every overlapping pair, ask what permits the successor to start and what permits it to finish. An SS relationship may protect the first endpoint while an FF relationship protects the second. If only one endpoint is constrained, a duration change can create an unintended result. Use a small hand calculation or schedule copy to test the relationship under shorter and longer durations.

Also distinguish relationship lag from work. A review period, cure, transport, or mobilization step often deserves its own activity because it has a calendar, owner, status, and evidence. Where a lag remains appropriate, document its unit, calendar, and basis. This makes later updates and path explanations far more reliable.