Construction Planning and Scheduling

Introduction and Context

Construction projects convert design information, resources and contractual commitments into a physical facility within constraints of time, cost, quality, safety and environmental performance. Planning determines what must be done, how work will be organised and what resources are required. Scheduling translates that plan into a time-phased sequence of activities, milestones and control points.

At postgraduate level, scheduling should not be understood merely as preparing a bar chart. It is a decision system that integrates scope definition, production logic, quantities, productivity, resources, procurement, risk, contract obligations and progress measurement. A technically elegant schedule can still be unrealistic if it ignores access constraints, workface congestion, design maturity, approvals, resource limitations or the production behaviour of repetitive work.

A professional schedule therefore has four functions: it communicates the intended execution strategy, provides a basis for coordinating parties, supports forecasting and serves as a baseline against which actual performance is evaluated. This chapter develops activity definition, network analysis, CPM, PERT, resource planning, line-of-balance, location-based methods, schedule compression, rolling-wave planning and baseline control.

Learning Objectives

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

1. Develop a logically complete work breakdown structure and activity list.

2. Construct and analyse precedence networks using CPM and PERT.

3. Interpret float, critical paths, calendars and schedule uncertainty.

4. Apply resource loading, resource smoothing and resource levelling.

5. Select suitable methods for repetitive and location-based construction.

6. Develop time–location diagrams and identify production conflicts.

7. Evaluate crashing, fast-tracking and other compression strategies.

8. Apply rolling-wave planning and establish a defensible baseline.

9. Use professional scheduling software responsibly for planning and control.

10. Critically evaluate schedule quality, governance, ethics and limitations.

1. From Scope to Schedule

1.1 Work breakdown and activity definition

The work breakdown structure (WBS) decomposes the project scope into progressively smaller deliverables. A good WBS is deliverable-oriented rather than merely organised by company department. At lower levels, work packages should be sufficiently specific to estimate, assign, schedule and measure, but not so detailed that administration overwhelms control.

An activity is a measurable piece of work that consumes time and, usually, resources. Activity definition should consider:

• the physical or managerial output;
• responsible organisation and work crew;
• quantity and productivity basis;
• predecessor and successor relationships;
• location or workfront;
• required information, permits and materials;
• planned duration and calendar;
• acceptance or completion criteria.

Milestones are zero-duration events representing significant conditions, such as approved design, possession of a site, completion of a floor or beneficial occupancy. They do not perform work, but they provide contractual and managerial control points.

A schedule should distinguish activities from constraints. “Complete excavation” is an activity; “must finish by 30 June” is a constraint. Excessive use of imposed dates can conceal weak logic and reduce the schedule's ability to forecast naturally.

1.2 Precedence relationships

The common logical relationships are:

• Finish-to-start (FS): the successor cannot start until the predecessor finishes.
• Start-to-start (SS): the successor cannot start until the predecessor starts, possibly after a lag.
• Finish-to-finish (FF): the successor cannot finish until the predecessor finishes, possibly after a lag.
• Start-to-finish (SF): the successor cannot finish until the predecessor starts; this is uncommon and should be justified.

A lag is a time separation inserted between linked events. For example, waterproofing may start two days after concrete placement. Lags should not replace activities when the intervening work requires management attention or progress measurement. A curing activity, inspection, curing period or approval is often clearer as an explicit activity.

Network logic must reflect physical reality. Relationships should not be added merely to force desired dates. Circular relationships, dangling activities, excessive open ends and unexplained constraints are indicators of poor schedule quality.

2. Critical Path Method

2.1 Basic model and assumptions

The critical path method (CPM) represents activities as nodes or arrows and calculates the longest time path through a logically connected network. In the activity-on-node convention, each activity has a duration and relationships connect activities.

The basic deterministic CPM assumes that activity durations are sufficiently represented by single estimates, relationships are known, resource limits are either incorporated separately or do not alter the calculated sequence, and the network is logically complete. These assumptions are important: the calculated critical path is a model result, not an unavoidable physical law.

For a simple finish-to-start network without leads or lags, the forward pass is:

EF_i = ES_i + d_i

ES_i = max(EF_j) for all immediate predecessors j

where ES is early start, EF is early finish and d is activity duration. If the project starts at time zero, an activity with no predecessor has ES = 0.

For the backward pass, if project duration is T:

LF_i = min(LS_k) for all immediate successors k

LS_i = LF_i - d_i

where LS is late start and LF is late finish. For an activity with no successor, LF is normally set equal to T.

Total float is:

TF_i = LS_i - ES_i = LF_i - EF_i

Activities with zero total float form one or more critical paths under the stated network and calendar. A path is critical only relative to the current data date, logic, durations, constraints and resource assumptions. Near-critical activities may become critical through small changes and therefore deserve management attention.

Free float is the delay possible without delaying the early start of any immediate successor. For a simple FS network:

FF_i = min(ES_k) - EF_i

where k represents immediate successors. Total float may be shared among activities, whereas free float is generally more locally available.

2.2 Worked CPM example

Consider the following activities, with durations in working days:

A, site establishment, 3 days, no predecessor;

B, excavation, 5 days, predecessor A;

C, reinforcement and formwork, 4 days, predecessor A;

D, foundation concrete, 2 days, predecessors B and C;

E, curing and foundation inspection, 3 days, predecessor D;

F, backfilling, 2 days, predecessor E.

Forward pass:

A: ES 0, EF 3.

B: ES 3, EF 8.

C: ES 3, EF 7.

D: ES max(8, 7) = 8, EF 10.

E: ES 10, EF 13.

F: ES 13, EF 15.

The planned project duration is therefore 15 working days. The controlling path is A–B–D–E–F, with duration 3 + 5 + 2 + 3 + 2 = 15 days. Activity C has one day of total float because it finishes at day 7 while D cannot start before day 8. If B increases by two days, the project finish increases by two days unless mitigation changes the logic or durations. If C increases by one day, it uses its float; a further increase threatens completion.

2.3 Schedule quality and interpretation

A CPM schedule should be tested for:

• complete scope coverage;
• logical continuity from project start to finish;
• realistic calendars and working hours;
• appropriate activity durations;
• valid constraints and lags;
• procurement, design, approvals and commissioning;
• resource and access feasibility;
• measurable progress rules;
• absence of unexplained negative float;
• consistency with contract milestones.

Negative float indicates that the calculated completion is later than a required date, usually because of a constraint or imposed deadline. It is a warning requiring analysis, not a reason to manipulate logic.

3. PERT and Schedule Uncertainty

CPM uses a single duration estimate. The program evaluation and review technique (PERT) uses three estimates for an activity: optimistic time a, most likely time m and pessimistic time b. The commonly used expected duration is:

te = (a + 4m + b) / 6

The associated variance is:

σ² = [(b - a) / 6]²

These expressions are based on a beta-like distribution and are most useful when estimates represent genuine uncertainty rather than arbitrary optimism and pessimism. The expected project duration is obtained by adding expected durations along a selected path. Under the simplifying assumption that activity durations are independent, the path variance is the sum of activity variances and the path standard deviation is the square root of that sum.

For a target completion time X on a path with mean μ and standard deviation σ, a normal approximation uses:

Z = (X - μ) / σ

The probability is then estimated from the standard normal distribution. This approximation must be treated cautiously. Activity durations may be correlated, multiple paths may compete to become critical, and the critical path can change as uncertainty is realised. Monte Carlo simulation is generally more informative for complex schedules because it samples duration and risk distributions repeatedly and reports completion percentiles rather than a single deterministic date.

4. Resource Loading, Levelling and Production Feasibility

A time-only schedule may claim that several activities can occur simultaneously even though the project has one crane, a limited workforce, restricted access or insufficient working space. Resource loading assigns labour, equipment, materials or cost quantities to activities and displays demand over time.

Resource smoothing adjusts noncritical activities within available float to reduce peaks while preserving the planned completion date. Resource levelling allows activity dates to move, potentially extending the project, so that demand does not exceed availability. The distinction is managerial: smoothing protects the completion date where possible; levelling prioritises feasible resource demand.

Resource planning should consider:

• crew composition and productivity;
• shift patterns, calendars and fatigue;
• equipment utilisation and maintenance;
• material delivery and storage;
• workface congestion and safety separation;
• subcontractor interfaces;
• mobilisation and demobilisation;
• learning effects and weather exposure.

Software resource levelling is not a substitute for professional judgement. Automated algorithms may delay activities based on mathematical priority rules without understanding contractual obligations, access strategy or constructability. The planner must review the resulting sequence and document assumptions.

5. Planning Repetitive and Location-Based Work

Traditional CPM is effective for many unique activities, but repetitive projects—such as roads, pipelines, tunnels, apartment floors and housing units—are strongly influenced by production continuity. A crew that completes one location should move efficiently to the next without waiting, excessive interference or loss of learning.

5.1 Line of balance

The line-of-balance (LOB) method represents production rates through repeated locations. A line's slope expresses the rate at which a crew progresses through locations. If a crew must complete n units in a time interval, its average production rate is:

r = n / Δt

where r is units per unit time. Rates must be expressed consistently, such as floors per week or metres per day.

LOB is particularly useful for detecting crossings. If the line for a successor trade intersects or overtakes the predecessor line, the planned sequence may produce a workface conflict. Buffers can be introduced to protect safety, quality, curing or inspection requirements.

5.2 Location-based planning and time–location diagrams

Location-based management divides a project into zones or production units and plans crew movement through them. A time–location diagram places time on the horizontal axis and location on the vertical axis. The slope of a work line indicates production speed; a steeper line means more locations are completed per unit of time when axes are scaled conventionally.

The method makes visible features that a bar chart can hide: crew interruption, reverse movement, congestion, idle time and inconsistent production. It is especially valuable where the same work sequence repeats. However, location-based planning still requires a logic model for nonrepetitive work, design release, procurement, commissioning and project-wide interfaces.

A practical workflow is to define zones, estimate quantities and production rates, establish crew sequences, insert technological and safety buffers, plot the lines, check for conflicts and then reconcile the result with the CPM schedule. The resulting integrated schedule can contain both network activities and location-based production activities.

6. Schedule Compression

Compression is used when the forecast completion exceeds a contractual or management target. It should begin with diagnosis: the delay may arise from incorrect logic, late design, insufficient resources, poor productivity, procurement or an unrealistic original plan.

6.1 Crashing

Crashing adds resources, overtime, additional shifts, alternative methods or higher-cost procurement to reduce activity duration. A basic cost slope is:

Cost slope = (Crash cost - Normal cost) / (Normal duration - Crash duration)

The units are cost per unit of time saved. The expression is meaningful only when the relationship between time and direct cost is reasonably understood and the activity is technically crashable.

A rational crashing analysis generally identifies activities on the current critical path, estimates feasible minimum durations and incremental costs, and selects the least-cost measures that achieve the required reduction. After each change, the network must be recalculated because another path may become critical. Indirect savings, disruption, safety effects, quality risk and subcontract conditions should be included in the decision.

6.2 Fast-tracking and alternative methods

Fast-tracking overlaps activities that were originally sequential, such as beginning procurement from an approved design package before all design is complete. It can reduce duration without directly increasing crew size, but increases rework, coordination and interface risk. It requires clearly defined information-release rules and decision authority.

Other strategies include prefabrication, modularisation, design standardisation, additional workfaces, changed construction methods, resequencing and selective overtime. Compression should never compromise statutory safety requirements, inspection, quality assurance or worker welfare.

7. Rolling-Wave Planning and Baseline Control

Rolling-wave planning develops near-term work in greater detail while keeping future work at an appropriate planning level. The near-term window may include detailed activities, crew assignments, constraints and daily targets; later work may remain as higher-level activities until design, procurement or access information matures.

This approach is appropriate in uncertain environments, but it must not become a mechanism for hiding delay. Each planning cycle should identify constraints, assign owners, establish release criteria and record changes. Look-ahead planning commonly examines the next few weeks and tests whether planned activities are genuinely ready to start.

The baseline is the approved time-phased plan against which performance is measured. Baseline control requires:

1. a documented schedule basis and assumptions;

2. approval by authorised parties;

3. controlled calendars, coding and work breakdown;

4. a formal change procedure;

5. periodic status dates and progress records;

6. preservation of historical baselines;

7. explanation of variance and forecast change.

Progress should be measured using objective rules. Physical progress may use installed quantities, weighted milestones or earned-value methods, but planned percentage complete should not be confused with actual accomplishment. A schedule update should record actual starts and finishes, remaining durations, logic changes, delays, constraints and forecast completion.

In earned value terminology, schedule performance can be summarised by schedule variance, SV = EV - PV, and schedule performance index, SPI = EV / PV, where EV is earned value and PV is planned value. These indicators are dimensionless or value-based, but they do not explain causation and can be misleading when activities are assigned unsuitable weights. They should supplement, not replace, critical-path and production analysis.

8. Professional Scheduling Software

Software such as Primavera P6, Microsoft Project and equivalent systems can perform network calculations, calendars, resource loading, baselines, progress updates, filters, histograms and reports. The software does not create a valid plan automatically. Its output depends on the quality of scope, logic, durations, calendars, coding and status data.

A disciplined software workflow is:

• establish the WBS and coding structure;
• define calendars and exceptions;
• enter activities and milestones;
• apply logic and review open ends;
• estimate durations and resources;
• calculate the schedule;
• inspect critical and near-critical paths;
• conduct constructability and resource reviews;
• approve and save a baseline;
• update at a fixed data date;
• compare current forecast with baseline;
• report causes, decisions and corrective actions.

Data exchange requires attention to calendars, activity IDs, time zones, coding dictionaries, actual-date conventions and ownership of the schedule. A visually attractive report may conceal invalid logic or manipulated progress. Audit trails, permissions and version control are therefore governance requirements.

Artificial intelligence can assist with schedule risk identification, natural-language extraction of activities from documents, anomaly detection, progress recognition from images and forecast modelling. These uses require validated data, explainable outputs, human review, cybersecurity and protection of confidential project information. AI should support—not replace—the planner's accountability for logic and professional judgement.

9. Illustrative Applied Case: Urban Hospital Construction

Background

This fictional case concerns a six-storey urban hospital with a constrained site, a tower crane, repeated floor layouts and a fixed opening date. The project includes structural floors, façade, mechanical and electrical services, medical installations, testing and commissioning.

Problem

The initial CPM schedule forecast completion 11 weeks later than the required opening. The schedule showed several activities with zero float, but it did not represent floor-by-floor crew movement, inspection release or commissioning interfaces. Resource loading also showed simultaneous demand for the crane and a limited services workforce.

Analysis

The project team first validated the WBS and removed summary activities that were being used as substitutes for detailed work. Each floor was divided into structural, façade, services rough-in, partitions, finishes and testing work packages. A time–location diagram revealed that the services crew was repeatedly interrupted because structural and façade sequences released workfaces unevenly. The crane histogram showed a demand peak during façade installation and material distribution. A review also found that commissioning had been represented as one final activity rather than progressive systems testing.

Approach and solution

The team adopted a hybrid schedule. CPM controlled design approvals, procurement, structural completion, major interfaces and commissioning. Location-based planning controlled repetitive floor work. A two-floor buffer was created between selected crews, and services design was released in packages aligned with floor zones. Prefabricated service modules reduced workface duration. Noncritical procurement activities were rescheduled within float to reduce the crane peak. Selected activities were overlapped only after issuing defined design-release criteria.

The updated plan used rolling-wave detail for the next six weeks and maintained higher-level activities for later fit-out zones. A controlled change request documented added prefabrication cost, reduced duration, access assumptions and safety controls. Progressive testing was added to the baseline rather than relying on a single end-of-project commissioning activity.

Outcome

The revised forecast recovered eight of the eleven weeks without assuming unsafe overtime. The remaining three weeks were addressed through an agreed change to the opening-readiness sequence and earlier partial occupancy of nonclinical areas. Weekly time–location reviews identified crew stoppages earlier than the original bar chart had done. The project team retained the original baseline and separately recorded the approved revised baseline, preserving accountability for the causes and decisions.

Learning points

The case demonstrates that schedule recovery is not simply a matter of shortening critical activities. Accurate location modelling, resource feasibility, design-release control, progressive commissioning and transparent baseline changes are equally important. Hybrid planning is often more reliable than forcing a repetitive project into a purely activity-based network.

10. Challenges, Limitations and Governance

Schedules are models and therefore contain uncertainty. Common limitations include inaccurate quantities, optimistic productivity, incomplete design, correlated risks, weather variability, unreliable progress data and changing site conditions. CPM can imply false precision, while PERT can suggest statistical confidence that is unsupported by the quality of estimates. Resource levelling can produce a mathematically feasible but operationally poor sequence. LOB diagrams can oversimplify unique constraints and nonrepetitive work.

Ethical scheduling requires truthful status reporting, transparent assumptions and avoidance of manipulating logic, calendars or progress percentages to conceal delay. Planners should not create artificial critical paths, insert unjustified constraints or report completed work that has not met acceptance criteria. Contractual notices, records of delay, access conditions and instructions should be preserved. Safety and legal working-hour requirements override schedule pressure.

Good practice includes independent schedule reviews, documented schedule basis, risk workshops, constructability reviews, regular look-ahead meetings, objective progress rules, change control, audit trails and integration of schedule with cost, procurement, quality and safety systems. The strongest schedule is not the one with the most activities; it is the one that provides a credible, transparent and actionable representation of how the project can be delivered.

Chapter Summary

Construction planning converts scope into deliverables, work packages, activities, resources and execution logic. CPM calculates early and late dates, float and critical paths under stated assumptions, while PERT represents uncertain durations using three-point estimates. Resource loading and levelling test whether a time plan is operationally feasible. For repetitive construction, line-of-balance and time–location methods reveal production rates, workface conflicts and crew continuity that conventional bar charts may hide.

Schedule compression should be based on diagnosis and should consider cost, risk, quality, safety and changing critical paths. Rolling-wave planning provides detail where information is mature while preserving flexibility for later work. Baseline control requires approved assumptions, objective progress measurement, controlled change and a retained audit trail. Professional software is a calculation and communication tool, not a substitute for sound planning. Effective postgraduate practice integrates network analysis, production thinking, resources, uncertainty, governance and professional judgement.

Key Terms

Activity: A defined piece of work with duration, logic and usually resource requirements.

Baseline: An approved time-phased plan used for performance comparison.

Critical path: The longest logical path governing calculated project completion.

Crashing: Reducing duration by applying additional or more costly resources or methods.

Data date: The status date up to which actual project information has been recorded.

Float: Schedule flexibility available before a defined activity or project date is delayed.

Fast-tracking: Overlapping activities that were previously planned sequentially.

Lag: A time separation applied to a logical relationship.

Line of balance: A production-oriented method showing progress rates through repeated locations.

Location-based planning: Planning work by zones or production units and crew movement.

Milestone: A significant zero-duration event or control point.

PERT: A technique using optimistic, most likely and pessimistic duration estimates.

Precedence relationship: A logical dependency between activities.

Resource levelling: Adjusting activities so resource demand does not exceed availability.

Resource smoothing: Adjusting activities within available float to reduce demand peaks.

Rolling-wave planning: Detailed near-term planning with progressively less detailed future planning.

Schedule compression: Strategies used to reduce forecast project duration.

Time–location diagram: A chart showing work progress simultaneously against time and physical location.

Total float: Delay available before an activity delays the calculated project completion.

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

Review Questions

Short Answer Questions

1. Distinguish an activity from a milestone and give one construction example of each.

2. What is the difference between resource smoothing and resource levelling?

3. Define total float and explain why zero float does not necessarily mean physical inevitability.

4. State two reasons for using a time–location diagram on a repetitive project.

5. What information should be recorded when updating a baseline schedule?

Long Answer Questions

1. Explain the complete process of developing a construction schedule from the WBS through baseline approval.

2. Describe the CPM forward-pass and backward-pass calculations, including the assumptions behind them.

3. Compare CPM, PERT, line-of-balance and location-based planning in terms of strengths and limitations.

4. Discuss crashing and fast-tracking, including cost, risk, quality and safety implications.

5. Explain how professional scheduling software should be governed to ensure reliable project control.

Analytical / Application Questions

1. A contractor has a fixed completion date, a limited crane and repeated floor work. Propose a hybrid planning and control approach, explaining how CPM, resource analysis and time–location planning would be integrated.

2. A schedule reports 95% completion, but commissioning activities remain largely unfinished and several installed systems have not passed acceptance tests. Critically evaluate the progress measurement and recommend corrective actions.

3. A project is forecast twelve days late. Evaluate whether crashing, fast-tracking, resequencing, prefabrication or improved design release should be used, and identify the data required before selecting a recovery strategy.

Multiple Choice Questions with Answers

Question 1

In a basic CPM forward pass, the early start of an activity with multiple predecessors is normally determined by:

A. The minimum early finish of its predecessors

B. The maximum early finish of its predecessors

C. The average duration of its predecessors

D. The latest finish of the project

Correct Answer: B. The maximum early finish of its predecessors

Question 2

For an activity with ES = 6 days and LS = 10 days, assuming the same float convention, total float is:

A. 4 days

B. 6 days

C. 10 days

D. 16 days

Correct Answer: A. 4 days

Question 3

Using PERT, if optimistic time is 2 days, most likely time is 5 days and pessimistic time is 8 days, the expected duration is:

A. 4 days

B. 5 days

C. 6 days

D. 15 days

Correct Answer: B. 5 days

Question 4

Which method primarily represents crew production through repeated physical locations?

A. Line of balance

B. Cash-flow forecasting

C. Design matrix analysis

D. Inspection test planning

Correct Answer: A. Line of balance

Question 5

Resource smoothing generally attempts to reduce resource peaks while:

A. Ignoring activity logic

B. Preserving the planned completion date where possible

C. Eliminating all project float

D. Replacing the baseline with actual cost

Correct Answer: B. Preserving the planned completion date where possible

Question 6

Which action is an example of fast-tracking?

A. Adding a second crane without changing logic

B. Increasing the concrete strength specification

C. Starting procurement from an approved design package before all design is complete

D. Extending every activity by one week

Correct Answer: C. Starting procurement from an approved design package before all design is complete

Question 7

The main purpose of a schedule baseline is to:

A. Prevent all future changes

B. Provide an approved reference for performance comparison

C. Guarantee that the critical path never changes

D. Replace the project cost estimate

Correct Answer: B. Provide an approved reference for performance comparison

Question 8

In a time–location diagram, a sudden change in the slope of a crew line most directly indicates a change in:

A. Production rate

B. Contract currency

C. Project insurance

D. Structural design load

Correct Answer: A. Production rate

Question 9

A cost slope for crashing is calculated as:

A. Normal cost divided by normal duration

B. Crash cost divided by crash duration

C. Crash cost minus normal cost, divided by normal duration minus crash duration

D. Normal duration minus crash duration, divided by total project cost

Correct Answer: C. Crash cost minus normal cost, divided by normal duration minus crash duration

Question 10

Which statement best describes rolling-wave planning?

A. All activities are planned to the same level of detail at project start

B. Only completed activities are included in the schedule

C. Near-term work is detailed while future work remains at a suitable higher level

D. The baseline is deleted after every update

Correct Answer: C. Near-term work is detailed while future work remains at a suitable higher level