Foundations of Project Network Scheduling

Introduction

Projects convert an intended change into a defined result within agreed constraints of scope, time, cost, quality, risk, and resources. Because project work is interdependent, managers cannot control time effectively by listing tasks alone. They must understand the logical relationships among activities, identify the sequence that governs completion, and determine where delay will affect the project’s promised finish date. Network scheduling provides this structure.

A project network is a logical model of project work. It represents activities, their dependencies, and the points at which work begins, ends, or achieves a significant condition. The Critical Path Method (CPM) uses this network to calculate dates, identify schedule flexibility, and focus managerial attention on activities that may determine the project duration.

Network scheduling is therefore more than a computational technique. It is a way of thinking about project time: work is decomposed, relationships are made explicit, assumptions are exposed, and completion is examined as a system rather than as a collection of isolated tasks.

Learning Objectives

By the end of this chapter, the reader should be able to:

• explain the purpose and historical development of network scheduling;
• distinguish a project, activity, event, milestone, deliverable, work package, and schedule;
• describe the logic and assumptions of CPM;
• calculate basic early and late schedule dates and total float;
• relate CPM to the project life cycle, Work Breakdown Structure (WBS), and schedule management plan;
• evaluate the practical limitations and governance risks of network-based schedules; and
• interpret a network schedule for managerial decision-making.

Why Network Scheduling Matters

A simple task list answers the question, “What work exists?” A network schedule also answers:

• Which activities must precede others?
• Which activities can occur concurrently?
• What is the earliest feasible project completion date?
• Which activities have no scheduling flexibility?
• Where should management allocate attention, contingency, or recovery effort?

The network approach supports several important decisions. It helps a project manager sequence procurement and construction, coordinate design interfaces, assess the effect of a delayed approval, compare alternative execution strategies, and communicate the consequences of schedule change. It also provides a basis for schedule performance measurement and time-related risk analysis.

A network schedule is a model, not a prediction of reality. Its value depends on the quality of the underlying scope definition, duration estimates, dependency logic, calendars, resource assumptions, and progress information.

The Evolution of Network Scheduling

Early project planning relied heavily on bar charts, particularly the Gantt chart developed in the early twentieth century. Bar charts remain useful because they communicate planned timing clearly, but they do not inherently show the logical basis for that timing. A bar may indicate that an activity begins on a particular date without explaining whether the date is caused by a predecessor, a resource constraint, a contractual rule, or a managerial preference.

During the 1950s, two influential network methods emerged. The Program Evaluation and Review Technique (PERT) was developed for the U.S. Navy’s Polaris missile programme and emphasized uncertainty in activity durations. The Critical Path Method was developed through work associated with DuPont and Remington Rand and initially supported planning and control of maintenance and construction activities. CPM used deterministic duration estimates and focused on identifying the longest path through a network.

Modern scheduling software integrates network logic with calendars, resource data, cost information, baselines, progress updates, risk analysis, and reporting. Contemporary practice also uses critical-chain ideas, probabilistic simulation, earned value management, constraint analysis, and integrated digital models. Nevertheless, the underlying principles remain the same: define work, establish logic, estimate durations, calculate schedule dates, and manage changes transparently.

Fundamental Vocabulary

Precision in terminology is essential because different project elements serve different purposes.

Project

A project is a temporary endeavour undertaken to create a unique product, service, or result. It has a defined beginning and end, although the end may occur through completion, cancellation, or transition into operations. A project is distinguished from ongoing operations, which repeatedly produce or maintain outputs.

Activity

An activity is a discrete item of work that consumes time and, usually, resources. Examples include “prepare foundation,” “approve design,” and “install control panel.” An activity should be sufficiently defined to permit duration estimation, responsibility assignment, progress measurement, and logical linkage.

An activity is not necessarily the same as a WBS element. A WBS element describes scope; an activity describes the work required to produce or verify that scope. In practice, a work package is decomposed into one or more schedule activities.

Event

In traditional Activity-on-Arrow network notation, an event is a point in time representing the start or completion of activities. It consumes no time and no resources. Events are useful for expressing logical conditions, but they are not deliverables and should not be confused with work.

In Activity-on-Node notation, which is common in contemporary software, the node represents an activity and arrows represent dependencies. The distinction between activities and events must therefore be interpreted according to the notation being used.

Milestone

A milestone is a significant point or achievement in a project, normally with zero duration. Examples include “contract signed,” “design approved,” “system commissioned,” and “client acceptance.” A milestone may mark an event, a decision, a control point, or an external commitment. It should have clear acceptance criteria; otherwise it becomes a vague reporting label.

Deliverable

A deliverable is a verifiable product, result, or capability produced by project work. A deliverable may be internal, such as an approved design package, or external, such as an operational facility. Deliverables are primarily scope and acceptance concepts, whereas activities are schedule and execution concepts.

Work package

A work package is the lowest WBS component at which cost, duration, responsibility, and performance can be planned and controlled. It is not necessarily a single activity. For example, the work package “Install electrical systems” may contain activities for procurement, cable installation, testing, inspection, and handover.

Schedule

A schedule is an integrated time-based model showing planned activities, logical relationships, dates, milestones, constraints, calendars, and often resources and costs. A schedule is more than a list of dates. It expresses an execution strategy.

Dependency and constraint

A dependency is a logical relationship in which the timing or completion of one element affects another. A constraint is a restriction imposed by a contractual date, law, resource availability, technology, policy, or management decision. Dependencies describe relationships; constraints limit possible dates. Confusing the two can produce an artificially rigid schedule.

Network Logic and Notation

The most widely used contemporary convention is Activity-on-Node (AON), also called precedence diagramming. Each activity appears as a node, and arrows show logical relationships. The principal relationship types are:

Finish-to-start is the most common relationship. Leads allow a successor to begin before the normal relationship point; lags impose a waiting period. For example, curing concrete may require a lag after placement. Leads and lags should be used only when they represent real logic, not merely to force desired dates.

Figure: An activity-on-node network with parallel paths. The highlighted path has the longest duration and is critical under the stated assumptions.

The diagram illustrates that a network may contain parallel work. The project cannot complete until all required predecessors of the final activity are complete. Consequently, the longest path—not necessarily the path with the greatest number of activities—determines the earliest project completion date.

CPM: Logic, Calculations, and Interpretation

CPM generally assumes that each activity has one chosen duration, relationships are logically valid, and the network is sufficiently defined. It calculates early dates by moving forward through the network and late dates by moving backward from the required completion date.

For an activity with duration d, the principal variables are:

For the common finish-to-start relationship with zero lag:

The forward-pass rule for an activity with several immediate predecessors is:

The maximum is required because the activity cannot start until every required predecessor has finished. For the backward pass:

For an activity with several immediate successors, the usual rule is:

The minimum is required because the activity must finish early enough to permit every successor to begin on time. Total float is:

All four dates use the same time unit. These equations apply to the stated AON, finish-to-start, zero-lag case. Other relationship types, calendars, resource constraints, and negative lags require corresponding scheduling logic and should not be treated as if the simple equations were universally sufficient.

An activity with zero total float is commonly described as critical. More precisely, it is critical relative to the network, calendars, imposed completion date, and calculation assumptions being used. A schedule may contain more than one critical path. Criticality can also change when durations, logic, constraints, calendars, or progress are updated.

Worked example

Consider four activities:

Set the project start at time 0. The forward pass gives:

• A: ES = 0, EF = 0 + 3 = 3
• B: ES = 3, EF = 3 + 5 = 8
• C: ES = 3, EF = 3 + 2 = 5
• D: ES = max(8,5) = 8, EF = 8 + 4 = 12

The earliest completion is therefore 12 days. The path A–B–D has duration 3 + 5 + 4 = 12, while A–C–D has duration 3 + 2 + 4 = 9. The latter has three days of path-level flexibility under the simple assumptions.

Using a required project completion at day 12, the backward pass gives:

• D: LF = 12, LS = 12 − 4 = 8
• B: LF = 8, LS = 8 − 5 = 3
• C: LF = 8, LS = 8 − 2 = 6
• A: LF = min(3,6) = 3, LS = 3 − 3 = 0

Thus, A, B, and D have zero total float. C has TF = 6 − 3 = 3 days. The calculation does not mean that C is unimportant; it means that, under these assumptions, C may be delayed by three days without delaying the project finish.

CPM within the Project Life Cycle

The project life cycle commonly includes initiation, planning, execution, monitoring and controlling, and closing, although organizational and delivery approaches vary. Network scheduling has a different role in each stage.

During initiation, the schedule is high-level. The sponsor may need a milestone roadmap, major dependencies, and a feasibility view rather than a detailed activity network. During planning, the team progressively decomposes scope, defines activities, estimates durations, develops logic, assigns calendars and resources, and establishes a schedule baseline. During execution, actual progress is compared with the baseline and forecasts are updated. During monitoring and controlling, the team analyzes variances, assesses impacts, manages changes, and may re-sequence work. During closing, the schedule records completion, acceptance, outstanding actions, and lessons learned.

In predictive approaches, the full network may be developed before execution, although it should still be refined through approved change control. In adaptive or hybrid approaches, the near-term work may be planned in greater detail while later work remains at a higher level. This is sometimes called rolling-wave planning. Network logic remains useful, but the level of detail and certainty changes over time.

Relationship to the WBS

The WBS is a hierarchical decomposition of the total project scope. It answers, “What must be delivered?” The network schedule answers, “In what logical order will the work be performed, and when?” These are complementary structures rather than alternatives.

A sound transformation from WBS to schedule normally follows this sequence:

1. identify deliverables and control accounts;
2. decompose work packages into activities;
3. define activity attributes, owners, assumptions, and acceptance conditions;
4. estimate durations and required resources;
5. sequence activities using technically valid dependencies;
6. add milestones, calendars, constraints, and external interfaces;
7. calculate the network and review its feasibility; and
8. approve the time-phased baseline through governance procedures.

The WBS should not be copied mechanically into the activity list. A single work package may require many activities, and some schedule activities—such as project management, inspection, or coordination—may span several deliverables. Traceability is nevertheless essential: every significant activity should map to a defined scope element, and every important deliverable should have activities that produce and verify it.

The Schedule Management Plan

The schedule management plan defines how the schedule will be developed, maintained, controlled, and reported. It should establish, as appropriate:

• scheduling methodology and notation;
• planning levels and required detail;
• calendars, working time, and time zones;
• estimation rules and assumptions;
• activity coding and responsibility structures;
• acceptable relationship types, leads, lags, and constraints;
• baseline approval and change-control procedures;
• progress measurement rules and data dates;
• critical-path and float-reporting conventions;
• update frequency and schedule-quality checks; and
• required reports, escalation thresholds, and record retention.

The plan prevents the schedule from becoming a private spreadsheet owned by one person. It establishes a common language and makes schedule information auditable. For example, one organization may define progress by physical quantities, another by weighted milestones, and another by earned value. Unless the measurement rule is explicit, reported completion may be inconsistent or misleading.

Applications in Professional Practice

CPM is valuable in construction, engineering, information systems, product development, pharmaceutical projects, transformation programmes, events, and public infrastructure. Typical uses include:

• evaluating alternative sequences and contract strategies;
• identifying dates that require executive or regulatory decisions;
• coordinating suppliers and interfaces;
• prioritizing recovery actions after delay;
• testing whether a target completion date is feasible;
• supporting time-related claims and change analysis; and
• integrating schedule information with cost and risk management.

A project manager should not manage only the critical path. Near-critical activities may have little float and may become critical after a small delay. Activities with substantial float may still be important because they consume scarce resources, affect quality, or create safety and reputational risks. The network is a decision aid, not a substitute for professional judgment.

Illustrative Case: Hospital Diagnostic Wing

Background

A hospital plans a new diagnostic wing. The project includes design approval, equipment procurement, building work, system installation, regulatory inspection, and clinical commissioning. The opening date is publicly announced, so schedule credibility is important.

Problem

The initial schedule is a bar chart. It shows procurement beginning three months after design, but it does not reveal that some equipment has a long manufacturing lead time or that regulatory inspection requires completed test records. Senior managers therefore believe that adding contractors to construction will guarantee an earlier opening.

Analysis

The project team develops a WBS and converts work packages into activities. The network reveals two principal paths: a building path and an equipment path. Equipment procurement begins after design approval, manufacturing overlaps with building work, and installation cannot finish until both the equipment and appropriate rooms are ready. Commissioning follows regulatory inspection.

The team identifies the equipment path as critical. Building work has some float, but equipment manufacturing and regulatory approval have almost none. The schedule management plan requires weekly updates, named owners for external approvals, and a formal rule that inspection dates cannot be treated as confirmed until the regulator accepts the submission.

Approach and solution

The team takes four actions. First, it obtains a supplier commitment and separates procurement into design release, purchase order, manufacture, delivery, and inspection activities. Second, it adds explicit approval milestones rather than hiding approval time in a general activity. Third, it develops a risk response for late equipment delivery, including an approved alternative supplier. Fourth, it performs a schedule-quality review to detect open ends, excessive constraints, and undocumented lags.

Outcome

The revised network shows that adding construction labour would not improve the opening date unless it addressed the actual controlling path. Management instead accelerates design release, assigns a senior owner to regulatory coordination, and protects the equipment manufacturing window. The project opens close to the approved baseline, and the organization retains a traceable explanation of key decisions.

Learning points

The case demonstrates that CPM improves decisions only when the network reflects real dependencies. It also shows why milestones, external approvals, procurement activities, and risk responses should be visible. A critical path is not a permanent property of the project; it is a current result of the model and its assumptions.

Limitations, Risks, and Governance

Deterministic duration assumptions

Traditional CPM uses one duration for each activity. Real work is uncertain because of rework, weather, productivity variation, approval delays, and changing requirements. A deterministic network can therefore create false precision. Probabilistic methods such as three-point estimating and Monte Carlo simulation may be appropriate for high-risk projects, but they depend on credible distributions, correlations, and data.

Resource limitations

Basic CPM often assumes that resources are available when needed. If several activities require the same specialist, the logically feasible network may be operationally impossible. Resource leveling may extend the completion date or change the critical path. Resource-constrained schedules should be distinguished from purely logic-driven CPM results.

Poor logic and excessive constraints

Missing predecessors, circular relationships, unexplained date constraints, and extensive use of “must start on” dates can make a schedule appear stable while concealing risk. A constraint may suppress the visible effect of delay rather than solve its cause. Professional review should ask whether each relationship reflects a real requirement.

Progress and status bias

Schedule forecasts are only as reliable as status data. Teams may report activities as complete when only a small portion of work is finished, or may move remaining duration into the future without explaining the cause. Governance should require objective completion rules, a defined data date, documented forecast changes, and separation of baseline, current forecast, and actual performance.

Ethical and contractual issues

Schedules may influence funding, public commitments, supplier claims, and safety decisions. Deliberately optimistic durations, hidden contingency, manipulation of progress, or selective reporting can create ethical and contractual harm. Schedule information should be traceable, version-controlled, and communicated with assumptions and uncertainty. A schedule must never be used to justify unsafe acceleration or suppression of required quality and regulatory activities.

Software and artificial intelligence

Scheduling software can calculate dates rapidly, identify logic errors, compare scenarios, and connect schedule data to cost and risk systems. AI-supported tools may suggest relationships, estimate durations from historical data, detect anomalies, or summarize schedule variance. These capabilities do not remove the need for domain expertise. AI-generated logic may reflect historical bias, omit novel dependencies, or reproduce poor-quality source data. Human approval, explainability, access control, privacy protection, and an auditable change history remain necessary.

Best-Practice Principles

1. Begin with approved scope and a traceable WBS.
2. Define activities as measurable, assignable units of work.
3. Use logic to represent real technical and managerial dependencies.
4. Minimize artificial constraints and document unavoidable ones.
5. Use consistent calendars, units, coding, and status dates.
6. Review the critical and near-critical paths after every material update.
7. Integrate resource, cost, risk, procurement, and approval information where relevant.
8. Maintain separate views of baseline, actuals, and current forecast.
9. Validate schedule quality with the people who perform the work.
10. Treat the network as a living model subject to controlled refinement.

Chapter Summary

Network scheduling provides a structured representation of project work, dependencies, timing, and completion conditions. Its development from bar-chart planning to CPM, PERT, integrated scheduling systems, and probabilistic analysis reflects the growing complexity of projects. The central conceptual distinction is between scope structures and time structures: deliverables and work packages describe what must be produced, while activities and relationships describe how and when the work will be performed.

CPM calculates early and late dates, total float, and the path or paths that currently govern project completion. Its basic equations are reliable for a defined AON network with stated duration, relationship, calendar, and completion assumptions. CPM does not automatically account for uncertainty, resource contention, changing scope, or poor-quality data. These limitations require professional judgment, risk analysis, schedule governance, and transparent updating.

A credible schedule is connected to the WBS, embedded in a schedule management plan, aligned with the project life cycle, and supported by measurable progress rules. Its managerial value lies not simply in identifying a critical path but in improving decisions about sequencing, approvals, resources, risk responses, and recovery.

Key Terms

Project: A temporary endeavour undertaken to create a unique product, service, or result.

Activity: A defined item of work that consumes time and usually resources.

Event: A point in time marking a start or finish condition, especially in activity-on-arrow notation.

Milestone: A significant zero-duration point or achievement used for control or communication.

Deliverable: A verifiable product, result, or capability produced by a project.

Work package: The lowest WBS component planned and controlled for scope, cost, duration, and responsibility.

Schedule: A time-based model of activities, relationships, dates, calendars, and milestones.

Dependency: A logical relationship between activities or project elements.

Constraint: A restriction that limits permissible timing or execution choices.

Activity-on-Node: A network notation in which nodes represent activities and arrows represent relationships.

Finish-to-Start: A relationship in which a successor starts after its predecessor finishes.

Critical Path Method: A deterministic network scheduling method used to calculate dates and float.

Forward pass: Calculation of earliest start and finish dates through a network.

Backward pass: Calculation of latest start and finish dates from a required completion date.

Total float: Time an activity can be delayed without delaying the calculated project completion date.

Baseline: An approved version of the schedule used for performance comparison.

Rolling-wave planning: Detailed planning of near-term work while future work remains at a higher level.

Schedule management plan: The document defining how the schedule will be developed, updated, controlled, and reported.

Near-critical path: A path with limited float that may become critical after delay or change.

Review Questions

Short Answer Questions

1. What is the principal difference between a project activity and a deliverable?
2. Why is a network schedule more informative than a task list?
3. Define total float in the context of CPM.
4. What is the purpose of a schedule management plan?
5. Why can the critical path change during project execution?

Long Answer Questions

1. Explain the evolution from bar charts to CPM and contemporary integrated scheduling systems.
2. Distinguish the WBS from the network schedule and explain how the two should be integrated.
3. Describe the four principal precedence relationships and give an example of each.
4. Explain the assumptions and limitations of deterministic CPM.
5. Discuss ethical and governance requirements for credible schedule reporting.

Analytical / Application Questions

1. A project contains two parallel paths: A–B–D with durations 4, 6, and 3 days, and A–C–D with durations 4, 2, and 3 days. Calculate the earliest completion time and identify the critical path, assuming finish-to-start relationships and no lags.
2. A project manager proposes adding workers to an activity with 10 days of float while a supplier activity on the critical path is forecast to be 15 days late. Analyze whether the proposal will recover the project completion date and recommend a more appropriate investigation.
3. Review a hypothetical schedule containing many “must start on” constraints, several activities with no predecessors, and progress reported as 100% when testing records are incomplete. Identify the governance weaknesses and propose corrective actions.

Multiple Choice Questions with Answers

Question 1

What is the primary purpose of a project network schedule?

A. To replace the project scope statement

B. To show logical relationships and timing among project activities

C. To eliminate all project uncertainty

D. To record only project costs

Correct Answer: B. To show logical relationships and timing among project activities

Question 2

Which element normally represents a significant zero-duration achievement?

A. Work package

B. Deliverable

C. Milestone

D. Resource calendar

Correct Answer: C. Milestone

Question 3

In a conventional finish-to-start relationship, when may the successor begin?

A. Before the predecessor begins

B. When the predecessor finishes, subject to any lag

C. Only after the entire project finishes

D. At the same time as the predecessor, regardless of status

Correct Answer: B. When the predecessor finishes, subject to any lag

Question 4

For an activity with duration d, which equation gives its early finish in the simple CPM case?

A. EF = ES + d

B. EF = ES − d

C. EF = LF + d

D. EF = LS − d

Correct Answer: A. EF = ES + d

Question 5

If an activity has ES = 7 and LS = 10, what is its total float?

A. 3 time units

B. 7 time units

C. 10 time units

D. 17 time units

Correct Answer: A. 3 time units

Question 6

Why is the maximum predecessor early finish used to calculate an activity’s early start?

A. The activity can start only after all required predecessors finish

B. The shortest predecessor controls the start

C. It removes the need for duration estimates

D. It guarantees unlimited resources

Correct Answer: A. The activity can start only after all required predecessors finish

Question 7

Which statement best distinguishes a WBS from a network schedule?

A. The WBS describes scope decomposition; the network describes execution logic and timing

B. The WBS contains only milestones; the network contains only deliverables

C. The WBS is used after closing; the network is used only during initiation

D. They are different names for exactly the same structure

Correct Answer: A. The WBS describes scope decomposition; the network describes execution logic and timing

Question 8

In the worked network A–B–D and A–C–D, with durations A = 3, B = 5, C = 2, and D = 4 days, what is the earliest completion?

A. 9 days

B. 10 days

C. 12 days

D. 14 days

Correct Answer: C. 12 days

Question 9

Which condition is a limitation of basic deterministic CPM?

A. It always models resource contention perfectly

B. It assumes one selected duration for each activity

C. It cannot represent any dependency

D. It requires every activity to be a milestone

Correct Answer: B. It assumes one selected duration for each activity

Question 10

Why should near-critical paths be monitored?

A. They always have zero duration

B. They may become critical after a relatively small delay

C. They are excluded from the approved schedule

D. They contain no resources

Correct Answer: B. They may become critical after a relatively small delay