Planning, Scheduling and Resource Management

Introduction

Infrastructure projects convert designs, approvals, materials, labour, equipment and finance into public assets. Their success depends not only on technical design but also on the sequence and timing of work. A road, bridge, water-treatment plant or rail extension may be constructible in principle yet fail in practice if activities are poorly sequenced, resources are unavailable, interfaces are ignored or the schedule is treated as a fixed promise rather than a controlled management model.

Planning defines what must be done and how the work will be organized. Scheduling translates that plan into time-phased activities, milestones and logical relationships. Resource management ensures that labour, plant, materials, money, information and workspace are available when required. These functions are interdependent: changing a construction method changes resource requirements; resource shortages change activity dates; and a changed completion date may affect contractual obligations and public disruption.

This chapter presents activity-based network scheduling, the critical path method (CPM), precedence relationships, milestones, critical-chain concepts, resource allocation and levelling, schedule uncertainty, and practical schedule baselines.

Learning objectives

After studying this chapter, a student should be able to:

1. Develop a work breakdown structure and a logically connected network schedule.
2. Calculate early and late dates, float and the critical path using CPM.
3. Apply precedence relationships, lags, leads and milestones appropriately.
4. Explain differences between CPM and critical-chain scheduling.
5. Allocate and level labour, plant and material resources without violating project logic.
6. Analyze uncertainty using three-point estimates and probabilistic reasoning.
7. Establish a credible baseline and control changes through disciplined governance.

1. From scope to a schedule model

A schedule is a model of planned work, not merely a list of dates. A reliable model begins with a clear scope definition and a work breakdown structure (WBS). The WBS decomposes the project into deliverables and work packages that can be estimated, assigned and verified. For example, a highway interchange may be divided into enabling works, earthworks, drainage, structures, pavement, electrical systems, testing and handover.

A work package is divided into activities that have a measurable start and finish. An activity should be sufficiently detailed to assign responsibility and duration, but not so detailed that the schedule becomes difficult to maintain. “Construct pier” may be an appropriate activity, whereas “move worker’s hand” is not. Activity definition should consider:

• the required output;
• predecessor and successor relationships;
• responsible organization;
• expected duration and resources;
• inspection, approval and access constraints; and
• the evidence needed to confirm completion.

A schedule dictionary records the meaning of activity names, calendars, assumptions, coding, responsibility and measurement rules. This prevents different team members from interpreting “substantially complete” or “installation complete” differently.

Logical relationships

The most common relationship is finish-to-start (FS): the successor cannot start until the predecessor finishes. Other relationships are:

A lag is a deliberate waiting or overlapping period between activities. For an FS relationship with lag L, the successor start is constrained by predecessor finish plus L. A lead is an allowed overlap and should be used cautiously because it can conceal incomplete work or unsafe assumptions. Physical constraints, contractual requirements and safety rules should be expressed as logic rather than informal notes.

A milestone is a zero-duration event used to mark an important point, such as “permit issued,” “bridge deck complete” or “section available for traffic.” Milestones are useful for governance and reporting, but they do not replace the activities required to achieve them.

2. Critical Path Method

CPM is a deterministic network technique. Each activity has an assumed duration, and logical relationships determine the earliest and latest times at which it can occur. The critical path is the continuous path through the network with the greatest total duration under the stated assumptions. Activities on that path generally have zero total float, although calendars, constraints and progress updates can produce more complex results.

Forward pass

For activity i, let ESᵢ be early start, EFᵢ early finish and dᵢ duration. Under a simple finish-to-start network:

For an activity with predecessors:

where j represents all immediate predecessors. For the first activity, ES is normally the project data date or time zero. The maximum is required because the activity must wait for its latest-finishing predecessor.

Backward pass

Let LFᵢ be late finish and LSᵢ late start. If T is the required project completion time, the final activity has LF = T. For an activity with successors:

where k represents all immediate successors. Then:

Total float is:

Float is expressed in the schedule’s time unit, commonly working days. These equations assume a consistent time origin, valid relationships, fixed durations and a network without contradictory constraints. Calendar effects must be handled consistently; a duration of ten working days is not the same as ten calendar days.

Worked CPM example

Consider the following activities. Durations are in working days and all relationships are finish-to-start.

Forward pass: A has ES 0 and EF 3. B has ES 3 and EF 7. C has ES 3 and EF 8. D must await both B and C, so ES = max(7, 8) = 8 and EF = 14. E runs from 14 to 22; F runs from 14 to 18. G starts at max(22, 18) = 22 and finishes at 24. The planned duration is therefore 24 working days.

Backward pass: G has LF 24 and LS 22. E has LF 22 and LS 14. F has LF 22 and LS 18. D has LF = min(14, 18) = 14 and LS 8. B has LF 8 and LS 4, giving float 1 day because ES = 3 and LS = 4. C has LF 8 and LS 3, giving zero float. A has LF = min(4, 3) = 3 and LS 0. The critical path is A–C–D–E–G, with duration 24 days. B has one day of total float, while F has four days of float.

The result is not a guarantee. It is conditional on the durations, logic, calendars and constraints being credible. If C takes six days rather than five, the project completion date increases by one day unless recovery action is taken.

Float interpretation

Total float is the amount an activity can be delayed without delaying the calculated project completion date, assuming other activities remain as scheduled. Free float is the delay possible without delaying the early start of any successor. Float is a property of the network, not necessarily an entitlement of one party. In contracts, ownership and use of float should be addressed explicitly because consumption of float can create disputes.

3. Precedence scheduling in practice

Activity-on-node precedence diagrams represent activities as nodes and relationships as connecting arrows. Modern scheduling software commonly uses this method. Good practice is to minimize arbitrary date constraints such as “must start on” unless they represent a real external condition. Excessive constraints can hide the true critical path and make recovery analysis unreliable.

Logic should reflect construction methods. For example, a long pipeline may be divided into work fronts: clearing, trenching, bedding, pipe laying, joint testing, backfilling and reinstatement. SS relationships may represent the controlled movement of crews between fronts, while testing may be FF-related to installation. The schedule must still respect safe access, inspection hold points and environmental requirements.

A schedule should be checked for:

• open ends, where activities lack legitimate predecessors or successors;
• circular logic;
• excessive lags and constraints;
• negative or unexplained float;
• activities longer than the organization’s control threshold;
• missing approvals, procurement and commissioning activities; and
• logic that conflicts with the method statement.

4. Critical-chain concepts

CPM focuses primarily on logical sequence and deterministic durations. Critical-chain project management (CCPM) adds resource dependence and emphasizes protection against uncertainty. The critical chain is the longest chain of dependent activities considering both technological precedence and resource constraints.

CCPM commonly uses shorter, focused activity estimates rather than embedding large safety allowances in every task. Protection is then placed in buffers:

• a project buffer protects the completion date;
• feeding buffers protect the critical chain from non-critical paths; and
• resource buffers provide warning that a critical resource will soon be needed.

The approach also discourages harmful multitasking and encourages starting work when the preceding work and required resources are actually ready. It is especially relevant where a scarce specialist crew, crane or commissioning engineer creates resource dependencies. However, CCPM is not a substitute for realistic quantities, sound logic or contractual planning. Buffer sizing and behavioural assumptions must be transparent, and many projects use hybrid approaches combining CPM reporting with resource-focused management.

5. Resource allocation and levelling

Resources include labour trades, equipment, materials, money, work areas, design capacity and management attention. Resource allocation assigns resources to activities. Resource levelling adjusts activity timing when demand exceeds availability. Resource smoothing attempts to reduce fluctuations while preserving the planned completion date and usually uses available float.

Suppose a project has two activities planned concurrently, each requiring three electricians, but only four are available. The schedule is infeasible as planned. Options include changing the sequence, splitting activities where technically safe, adding labour, subcontracting, changing work hours, using alternative methods or accepting a later completion. The decision should be based on cost, safety, productivity, contractual obligations and the value of earlier completion.

A simple resource profile totals demand by time period:

where Rₜ is total demand for a resource in period t and rᵢ,ₜ is the amount required by activity i in that period. A period is overloaded when Rₜ exceeds the available capacity Cₜ:

This comparison assumes that resource units are comparable and that availability, calendars and productivity assumptions are correctly defined.

Common levelling heuristics include delaying activities with float, prioritizing activities with lower float, preserving contractual milestones, and avoiding fragmentation that causes inefficient crew mobilization. Software optimization can assist, but the planner must validate the result against site access, constructability and safety. A mathematically level profile may be operationally impossible.

Resource-loaded schedules also support cash-flow forecasting, procurement planning and earned value management. They should distinguish committed, forecast and available resources; otherwise apparent capacity may be overstated.

6. Schedule uncertainty and risk

A single duration conceals uncertainty. Weather, ground conditions, design changes, productivity, supply interruptions and approvals can affect activity outcomes. A three-point estimate uses optimistic time a, most likely time m and pessimistic time b.

For the traditional PERT beta approximation, the expected duration is:

The approximate variance is:

Here all three time estimates must use the same units, and the estimates should describe a consistent probability range. The formula is an approximation, not a law of nature; it assumes a suitable unimodal distribution and should not be treated as precise when estimates are poorly informed.

For example, if an excavation activity has a = 4 days, m = 6 days and b = 10 days:

t_e = (4 + 4×6 + 10) ÷ 6 = 38 ÷ 6 ≈ 6.33 days

and σ² = ((10 − 4) ÷ 6)² = 1 day². The standard deviation is approximately 1 day. The expected project duration is not necessarily the sum of expected durations on every path because paths may merge and activity durations may be correlated.

Risk analysis should identify uncertainty drivers, assign owners and define responses. Schedule risk analysis may use simulation, in which activity durations are sampled repeatedly from specified distributions. The output can estimate the probability of meeting a target date. Results are only as credible as the logic, distributions, correlations and assumptions. A false sense of precision is a major danger.

Schedule contingency should be distinguished from management reserve. Contingency addresses identified uncertainty within the approved scope; management reserve is controlled by governance for unforeseen work and should not be hidden in arbitrary activity durations.

7. Establishing a practical baseline

A baseline is the approved reference against which actual performance and forecast completion are compared. It should be established only after scope, logic, durations, calendars, resources, cost interfaces and contractual milestones have been reviewed.

A credible baseline contains:

1. a scope-linked WBS and activity coding system;
2. logical relationships and documented assumptions;
3. calendars for work, holidays, shifts and weather allowances;
4. activity durations supported by quantities, productivity or supplier information;
5. resource and cost loading where required;
6. procurement, design, approvals, commissioning and handover activities;
7. identified milestones and acceptance criteria; and
8. a change-control procedure.

The baseline should be statused at regular intervals. Actual starts and finishes, remaining durations, progress evidence and approved changes must be recorded. Updating actual progress while silently changing the baseline destroys the ability to measure variance.

Useful indicators include schedule variance in time, milestone variance, critical-path movement, float consumption, planned-versus-actual quantities and forecast completion. Earned value measures may also be used. In that system, schedule performance index is:

where EV is earned value and PV is planned value, both expressed in the same monetary or equivalent value units. SPI greater than 1 suggests more work has been earned than planned at the status date; less than 1 suggests less. Because earned value can obscure actual time effects, it should be supplemented with network analysis and milestone forecasting.

A baseline is not immutable. Approved changes to scope, sequence, access or contract time may justify a controlled rebaseline. Rebaselining to conceal poor performance is unethical and undermines governance.

8. Applied case: illustrative municipal water-treatment upgrade

Background

A municipality plans to upgrade an operating water-treatment plant while maintaining continuous water supply. The work includes a new clarifier, pipework, electrical systems, commissioning and operator training. The contract completion milestone is fixed, but the site has limited laydown space and only one specialist electrical commissioning team is available.

Problem

The initial contractor schedule showed civil construction, mechanical installation and electrical installation proceeding in parallel. Review identified missing approval activities, an unrealistic assumption that the plant could be isolated at any time, and simultaneous demands for the same crane and commissioning team. The schedule predicted completion in 14 months but did not demonstrate feasibility.

Analysis

The project team rebuilt the WBS around design approvals, procurement, enabling works, civil works, mechanical installation, electrical installation, testing, disinfection and handover. Interface workshops identified that electrical energization could not begin until the utility approval and cable tests were complete. Commissioning had to occur in controlled shutdown windows. A resource histogram showed a crane overload during months 7–8 and a specialist commissioning overload during months 11–12.

Three-point estimates were developed for excavation and process-equipment delivery because ground and supplier risks were material. A risk register linked each uncertainty to an owner and response: supplementary ground investigation, early vendor drawing review, a delivery inspection plan and a contingency shutdown window.

Approach and solution

The team used CPM to calculate the contractual forecast and identify the civil-to-commissioning critical path. Resource levelling moved non-critical pipe-support installation into available float and resequenced two work fronts. The crane was reserved for lifts with no practical alternative; smaller lifts used mobile equipment. A feeding buffer was placed before commissioning, and a project buffer was reported separately from the deterministic contractual dates.

The approved baseline included measurable milestones: civil structure watertight, equipment mechanically complete, electrical approval, dry testing complete, wet commissioning complete and operator acceptance. Weekly updates used photographic evidence, test certificates and quantity records rather than subjective percentage claims.

Outcome

The revised plan forecast 15 months rather than 14, but it was resource-feasible and included the required shutdown window. Early procurement and design reviews reduced the probability of late equipment delivery. During construction, one delivery slipped by three weeks; the feeding buffer absorbed most of the delay, while an approved resequencing action protected the public-service milestone.

Learning points

The case demonstrates that a shorter schedule is not necessarily a better schedule. Logic, resources, uncertainty, operational constraints and acceptance requirements must be integrated. A baseline earns credibility through evidence and transparent assumptions, not through optimistic dates.

9. Professional, ethical and technological considerations

Planners have a professional duty to report realistic forecasts. Deliberately shortening durations to win work, hiding negative float, suppressing known risks or changing baselines without approval can expose workers, the public and project sponsors to harm. Schedule pressure must never justify unsafe concurrent operations or bypassing inspections.

Digital tools can improve collaboration through common data environments, 4D building information modelling, automated clash detection, mobile progress capture and schedule dashboards. Artificial intelligence can assist with anomaly detection, productivity forecasting and risk pattern recognition. However, AI outputs depend on data quality and may reproduce historical bias. A professional must verify logic, protect confidential information, document model assumptions and retain human accountability for contractual and safety decisions.

Contemporary practice increasingly links schedules with digital twins, geographic information systems and reality capture. These technologies are valuable when they answer a management question—such as whether installed quantities support a claimed milestone—not when they merely produce visual complexity.

Best-practice checklist

Before approving a schedule, ask:

• Is every major scope item represented by measurable work?
• Are design, procurement, approvals, interfaces and handover included?
• Does the logic reflect the construction method and safe access?
• Are calendars, shifts, weather and working-time rules consistent?
• Are durations supported by quantities, productivity or supplier evidence?
• Is the critical path stable and understandable?
• Are resource overloads visible and resolved or explicitly accepted?
• Are uncertainty, contingency and management reserve distinguished?
• Can progress be verified objectively?
• Is there a controlled process for updating and changing the baseline?

Chapter Summary

Planning converts project scope into a structured WBS and defined work packages. Scheduling connects activities through precedence relationships and uses CPM forward and backward passes to calculate dates, float and the critical path. Milestones communicate important events but do not replace the work needed to achieve them. Critical-chain thinking adds resource dependence and buffers to the analysis.

Resource allocation assigns labour, plant, materials and other constrained inputs; levelling resolves over-allocation, while smoothing uses float to reduce demand fluctuations. Schedule uncertainty should be represented through realistic ranges, risk ownership and, where justified, probabilistic analysis. A practical baseline integrates scope, logic, calendars, resources, costs, risks and measurable acceptance criteria. Ethical schedule management requires honest forecasts, controlled changes and protection of safety and public interests.

Key Terms

Work breakdown structure: Hierarchical decomposition of project scope into manageable deliverables and work packages.

Activity: A defined piece of work with a duration, output and logical position in a schedule.

Predecessor: An activity or event that constrains a successor.

Successor: An activity or event that follows another activity in the network.

Finish-to-start: Relationship in which the successor starts after the predecessor finishes.

Lag: Deliberate waiting or overlap time between logically related activities.

Milestone: Zero-duration event representing a significant achievement or decision point.

Critical Path Method: Deterministic network technique for calculating dates and float.

Critical path: Longest-duration logical path governing the calculated project completion.

Total float: Time an activity may be delayed without delaying calculated project completion.

Free float: Time an activity may be delayed without delaying a successor’s early start.

Resource allocation: Assignment of available resources to scheduled activities.

Resource levelling: Adjustment of activity timing to resolve resource over-allocation.

Resource smoothing: Reduction of resource fluctuations while preserving key dates where possible.

Critical chain: Longest chain considering both precedence and resource dependencies.

Buffer: Time protection placed to absorb uncertainty or protect a chain.

Three-point estimate: Duration estimate using optimistic, most likely and pessimistic values.

Baseline: Approved reference plan used to measure performance and control change.

Rebaseline: Controlled replacement or revision of an approved baseline.

Schedule risk analysis: Systematic assessment of uncertainty affecting dates and completion probability.

Review Questions

Short Answer Questions

1. What is the purpose of a work breakdown structure in schedule development?
2. Distinguish between total float and free float.
3. What is the difference between resource levelling and resource smoothing?
4. Why are milestones not sufficient as a complete project schedule?
5. State two conditions that should be documented when establishing a baseline.

Long Answer Questions

1. Explain the forward-pass and backward-pass procedures in CPM, including the assumptions behind the calculations.
2. Compare CPM and critical-chain project management in terms of durations, resources and buffers.
3. Discuss how procurement, approvals and commissioning should be represented in an infrastructure schedule.
4. Explain the ethical risks of unrealistic schedules and uncontrolled rebaselining.
5. Describe how a resource-loaded schedule can support construction planning and cost management.

Analytical / Application Questions

1. Construct a CPM network for a small building project with at least six activities, calculate early and late dates, and identify the critical path.
2. A project has two concurrent activities requiring a total of eight welders, but only five are available. Propose and justify at least three feasible responses, considering time, cost, safety and productivity.
3. For an activity with optimistic, most likely and pessimistic durations of 5, 8 and 17 working days, calculate its PERT expected duration and approximate variance. Explain how the result should be used in schedule-risk discussions.

Multiple Choice Questions with Answers

Question 1

In a finish-to-start network, an activity has predecessors finishing at days 12 and 15. What is its earliest possible start, ignoring lags?

A. Day 12

B. Day 13

C. Day 15

D. Day 27

Correct Answer: C. Day 15

Question 2

Which relationship means that a successor cannot start until its predecessor has finished?

A. Start-to-start

B. Finish-to-start

C. Finish-to-finish

D. Start-to-finish

Correct Answer: B. Finish-to-start

Question 3

An activity has ES = 6 days and LS = 9 days. Its total float is:

A. 3 days

B. 6 days

C. 9 days

D. 15 days

Correct Answer: A. 3 days

Question 4

Which statement best describes a critical path?

A. The path with the most activities

B. The path with the greatest total duration under stated assumptions

C. The path using the most labour

D. The path with the highest cost

Correct Answer: B. The path with the greatest total duration under stated assumptions

Question 5

If resource demand is 10 workers and available capacity is 7 workers in a period, the schedule is:

A. Under-allocated by 3 workers

B. Balanced exactly

C. Over-allocated by 3 workers

D. On the critical path

Correct Answer: C. Over-allocated by 3 workers

Question 6

Using t_e = (a + 4m + b) ÷ 6, what is the expected duration when a = 5, m = 8 and b = 17 days?

A. 8 days

B. 9 days

C. 9.67 days

D. 12 days

Correct Answer: B. 9 days

Question 7

What is the principal purpose of a project buffer in critical-chain scheduling?

A. To conceal missing activities

B. To protect the planned completion date from chain uncertainty

C. To increase every activity duration

D. To replace resource planning

Correct Answer: B. To protect the planned completion date from chain uncertainty

Question 8

Which action most directly protects baseline integrity?

A. Changing baseline dates whenever progress is late

B. Removing negative float from reports

C. Recording approved changes through formal change control

D. Using only milestone activities

Correct Answer: C. Recording approved changes through formal change control

Question 9

A milestone normally has:

A. A zero duration

B. A guaranteed zero cost

C. No predecessor

D. No relationship to acceptance

Correct Answer: A. A zero duration

Question 10

For an activity with a = 4, m = 6 and b = 10 days, the PERT variance is approximately:

A. 0.5 day²

B. 1 day²

C. 2 day²

D. 6 day²

Correct Answer: B. 1 day²