Module 6: Guided Workshop Activities and Integrated Project Output

Module 6: Guided Workshop Activities and Integrated Project Output

CPM, Pert, and S-Curve Training Module

Module 6 converts the preceding lessons into one progressive two-day workshop. It gives the facilitator the lecture sequence, participant inputs, centralized Google Sheet structure, graph requirements, step-by-step procedures, quality gates, required outputs, reporting standards, and AI-validation rules. The companion worked example contains the complete fictional dataset and all required simulations.

Learning Objectives

Establish one controlled and traceable project dataset as the source of truth for the workshop.
Gather and validate project information without inventing missing values.
Convert approved activity logic into reproducible CPM forward-pass, backward-pass, float, and critical-path outputs.
Apply PERT three-point estimates to communicate schedule uncertainty and target-date confidence.
Prepare and interpret activity-weight, CPM network, PERT probability, S-Curve, Earned Value, and recovery visualizations.
Connect project variance to activities, field evidence, constraints, corrective actions, and management decisions.
Produce a concise and defensible integrated project-control report.
Use AI only for bounded review and drafting under documented human validation.
Apply formal quality gates before proceeding to each succeeding activity.

Progressive Output Rule

Each group must use one project and one reporting cutoff throughout the workshop. Every activity consumes the verified output of the preceding activity. A group must not proceed when a required input is incomplete, inconsistent, unsupported, or unreviewed.

Module Navigation

Section	Main question	Required output
Workshop design	How will the two days be delivered?	Timetable and assigned group roles
Workbook architecture	Where will controlled inputs and calculations be stored?	Protected centralized workbook
Activity 1	Is the project dataset complete and traceable?	Validated project-control dataset
Activity 2A	Which activities and paths control completion, and how uncertain are they?	CPM and PERT outputs with graphs
Activity 2B	Is actual progress within the planned envelope?	S-Curve and optional Earned Value outputs
Activity 3	What actions and decisions are required?	Integrated report and recovery screening
Activity 4	Which AI statements are technically acceptable?	Validation log and corrected report
Final presentation	Can the group communicate a defensible management position?	Nine-minute presentation package

I. Two-Day Workshop Delivery Blueprint

Day and official period	Introductory instruction	Guided activity	Primary output
Day 1, 8:30–10:00	Lecture 1 and workbook orientation	Demonstration and role assignment	Participants understand the data requirements and controls
Day 1, 10:15–12:00	Guided data collection	Activity 1: Centralized Project Data Gathering and Validation	Validated project dataset
Day 1, 1:15–2:00	Lecture 2A: CPM and PERT	Schedule and probability demonstration	Analysis method understood
Day 1, 2:00–3:00	Guided computation	Activity 2A: CPM and PERT Analysis	Network, float, paths, risk outputs, and graphs
Day 1, 3:15–3:45	Lecture 2B: S-Curve Monitoring	Planned-versus-actual demonstration	Progress-control method understood
Day 1, 3:45–4:30	Guided progress analysis	Activity 2B: S-Curve Analysis	Progress curves, variance, and status finding
Day 2, 8:45–9:30	Lecture 3: Management Reporting	Integrated-report demonstration	Reporting method understood
Day 2, 9:30–10:00	Output-template orientation	Group task assignment	Report work plan
Day 2, 10:15–12:00	Guided report development	Activity 3: Integrated Project Output	Management-ready report and recovery graphic
Day 2, 1:15–2:00	Lecture 4: Responsible AI	Prompt and validation demonstration	AI boundaries understood
Day 2, 2:00–3:00	AI-supported drafting	Activity 4A: AI-Assisted Review and Drafting	Original AI output
Day 2, 3:15–4:00	Human verification	Activity 4B: AI Validation and Correction	Validation log and corrected report
Day 2, 4:00–4:30	Group presentation	Final project review	Engineering judgment and recommendation

Required Instructional Pattern

Introductory lecture for 30–45 minutes: discuss the concepts, required inputs, procedures, visual outputs, and expected decisions.
Facilitator demonstration: use the fictional irrigation-project dataset in the companion example.
Guided group work: apply the same method to an actual, recently completed, or facilitator-provided project.
Quality-gate review: require the group and facilitator to validate the output.
Controlled submission: preserve the output as an input to the next activity and the final report.

Participant Roles

Role	Required responsibility
Project Data Lead	Controls the project profile, coordinates completion, and maintains the Change Log
Schedule Lead	Encodes activities, relationships, durations, dates, milestones, calendars, and resources
Quantity and Progress Lead	Encodes quantities, weights, planned values, and verified accomplishment
Documentation Lead	Verifies approved records, inspections, photographs, tests, and correspondence
Quality Reviewer	Checks completeness, arithmetic consistency, traceability, graphs, and data status
Reporter	Develops report-ready statements, tables, graphs, and the final presentation

II. Facilitator and Participant Preparation

Participant Readiness

Can read a Program of Work and Bill of Quantities.
Can distinguish quantity, unit cost, total cost, and activity weight.
Can interpret an approved construction schedule and progress report.
Can use spreadsheet formulas, filters, charts, and protected ranges.
Understands that physical accomplishment must be supported by field evidence.
Can identify basic construction sequencing, testing, and inspection requirements.

Facilitator Pre-Workshop Setup

STEP-BY-STEP

Create one master Google Sheet under the authorized training account.
Create one protected group template or separate group workbook from the master.
Protect formula, dashboard, data-dictionary, validation-list, and baseline ranges.
Grant edit access only to designated participants and view access to observers.
Prepare the fictional dataset and simulations from the companion example.
Test formulas, dropdowns, filters, protected ranges, links, charts, and dark-mode rendering.
Prepare a local spreadsheet copy for an internet-outage contingency.
Assign a group code and Project Data Lead to every group.
Establish file naming, versioning, and reporting-cutoff conventions.
Confirm that confidential or prohibited information will not be entered into an unapproved system.

File and Version Naming

Artifact	Naming convention	Example
Group workbook	`GROUP##_PROJECTID_CUTOFF`	`GROUP03_LC3_WD40`
Project report	`PROJECTID_ProjectControlReport_v##`	`LC3_ProjectControlReport_v02`
S-Curve graph	`PROJECTID_SCurve_CUTOFF`	`LC3_SCurve_WD40`
CPM graph	`PROJECTID_CPMNetwork_v##`	`LC3_CPMNetwork_v02`
AI validation log	`PROJECTID_AIValidation_v##`	`LC3_AIValidation_v01`
Presentation	`GROUP##_PROJECTID_FinalPresentation`	`GROUP03_LC3_FinalPresentation`

III. Data Governance and Evidence Rules

Rules Applicable to All Activities

Use one project for the entire workshop.
Define one reporting cutoff and apply it consistently.
Use approved project records as the primary source.
Distinguish baseline, approved revision, actual, forecast, and proposed information.
Mark unavailable information as unavailable or pending verification; never guess.
Record a source reference for every material date, quantity, percentage, duration, cost, and cause statement.
Keep calculations deterministic, transparent, and reproducible.
Separate verified facts, calculations, interpretations, assumptions, forecasts, contractual questions, and recommendations.
Do not treat a graph as stronger evidence than its controlled source table.
Do not treat AI-generated text as verified engineering output.
Maintain a Change Log and preserve superseded values and the reason for each change.

Evidence Hierarchy

Evidence level	Typical source	Permitted use
Level 1 — Approved record	Contract, Notice to Proceed, approved POW, BOQ, baseline, or revision	Establish official scope, amount, duration, dates, logic, and weights
Level 2 — Verified field record	Signed inspection, accepted measurement, test, dated photograph, equipment log	Establish actual events, quantities, quality status, and field conditions
Level 3 — Submitted but unverified record	Contractor report, preliminary schedule, draft claim, unsigned log	Record as submitted, but do not treat as verified
Level 4 — Verbal or unsupported statement	Recollection, unreferenced explanation, assumed cause	Record only as requiring verification

Data-Status Codes

Code	Meaning	Required action
`VER`	Verified	May be used in analysis
`PEN`	Pending verification	Assign an owner and due date
`MIS`	Missing	Request the source record
`INC`	Inconsistent	Reconcile before use
`UNS`	Unsupported	Exclude from conclusions
`NAP`	Not applicable	State why it does not apply

IV. Visual and Graph Standards

Graph Rule

Every graph must be generated from the controlled workbook, use the same reporting cutoff as the report, show the units and data series clearly, and include a written engineering interpretation. Decorative or manually reshaped curves are not acceptable.

Mandatory and Conditional Visuals

Activity	Required graph or diagram	Purpose
Activity 1	Activity weight versus earned-contribution bar graph	Locate incomplete or unearned project value
Activity 2A	CPM network diagram	Show logic, critical paths, non-critical paths, and current control
Activity 2A	PERT completion-probability curve	Communicate confidence for selected target durations
Activity 2B	Early-start, late-start, and actual S-Curve	Show the planned envelope, boundary breach, variance, and trend
Activity 2B	PV-EV-AC bar graph when verified cost data exist	Compare schedule and cost status at one cutoff
Activity 3	Recovery-screening timeline when recovery is proposed	Communicate assumptions and possible remaining-duration effect
Activity 4	Validation table; graph optional	Show accepted, revised, unsupported, and rejected AI statements

Visual Quality Checklist

Use a descriptive graph title.
Label the horizontal and vertical axes and state units.
Include a legend when more than one series is displayed.
Show the reporting cutoff and important milestones.
Annotate the first S-Curve boundary breach and current variance.
Distinguish baseline critical paths from the current controlling activity.
Show the PERT mean, target duration, and completion probability.
State all recovery assumptions beside the recovery graph.
Do not use smoothing that changes the apparent data path.
Do not use a three-dimensional chart that distorts comparison.
Use labels and sufficient contrast so meaning does not depend on color alone.
Place the source table immediately before or after the graph.
Label every fictional-data visual as simulated.

Four-Part Graph Interpretation

Describe the display: state what the graph contains and the reporting cutoff used.
Identify the current condition: state the present value, variance, breach, or path position.
Identify the controlling factor: name the activity, path, cost value, or assumption that drives the conclusion.
State the required response: identify the immediate action, decision, or missing information.

V. Centralized Google Sheet Architecture

Tab	Primary owner	Minimum purpose
`00_Instructions`	Facilitator	Data dictionary, formats, status codes, naming rules, and privacy reminders
`01_Project_Registry`	Project Data Lead	Identity, dates, amount, cutoff, and status
`02_Source_Inventory`	Documentation Lead	Source availability, reference, evidence level, and reviewer
`03_Activity_Inputs`	Schedule Lead	Activities, relationships, durations, quantities, weights, and resources
`04_PERT_Estimates`	Schedule Lead	Three-point estimates, expected duration, variance, and basis
`05_Progress_Updates`	Quantity and Progress Lead	Planned and actual progress by activity and period
`06_Constraints`	Documentation Lead	Delay events, affected activities, evidence, status, and response
`07_CPM_Calculations`	Schedule Lead	ES, EF, LS, LF, Total Float, Free Float, and path status
`08_SCurve_Data`	Quantity and Progress Lead	Early-start, late-start, actual, and optional cost-control values
`09_Dashboard`	Facilitator-protected	Critical path, variance, risk flags, and graphs
`10_Report_Builder`	Reporter	Verified report statements and source references
`11_AI_Validation`	Quality Reviewer	Prompt, original response, validation, correction, and reviewer
`12_Change_Log`	Project Data Lead	Timestamp, old value, new value, source, reason, and editor

Minimum Activity-Input Columns

Column	Requirement	Validation rule
Project ID	Required	Must match Project Registry
Activity ID	Required	Unique and nonblank
WBS code	Recommended	Consistent hierarchy
Activity description	Required	Specific and measurable
Relationship	Required when predecessor exists	Use an approved relationship code
Immediate predecessor	Required except authorized start activities	Must reference an existing Activity ID
Lag or lead	Optional	State unit and approval basis
Baseline duration	Required	Positive numeric value
Calendar	Required	Identify working-day convention
Baseline start and finish	Required	Consistent with logic and calendar
Quantity and unit	Required when measurable	Nonnegative and traceable
Cost or physical weight	Required	Total equals 100 percent
Planned crew and equipment	Recommended	Use standard resource names
Milestone	Required when applicable	Zero duration or defined event
Source reference	Required	Approved record reference
Data status	Required	Valid status code only

Core Spreadsheet Formulas

Purpose	Example formula
Weight total	`=SUM(weight_range)`
Weight check	`=IF(ABS(SUM(weight_range)-1)<0.0001,"PASS","REVIEW")`
Earned contribution	`=activity_weight*verified_completion_fraction`
PERT expected duration	`=(optimistic+4*most_likely+pessimistic)/6`
PERT variance	`=((pessimistic-optimistic)/6)^2`
Early finish	`=early_start+duration`
Multiple-predecessor early start	`=MAX(predecessor_early_finish_range)`
Late start	`=late_finish-duration`
Total Float	`=late_start-early_start`
Free Float	`=MIN(successor_early_start_range)-early_finish`
Progress variance	`=actual_cumulative-planned_cumulative`
Simplified SPI	`=actual_or_earned_progress/planned_progress`
Completion probability	`=NORM.S.DIST(z_score,TRUE)`

Workbook Control Checklist

Facilitator owns the master workbook.
Formula, baseline, dashboard, and validation ranges are protected.
Each group edits only its assigned workbook or tabs.
Source links use controlled access.
Every revised value includes an approval reference.
Superseded values are preserved in the Change Log.
A versioned backup is saved after every activity.
Confidential information is excluded from unapproved platforms.

VI. Activity 1 — Centralized Project Data Gathering and Validation

Activity Card

Item	Requirement
Introductory lecture	45 minutes
Workbook demonstration	45 minutes
Guided group work	105 minutes
Main input	Approved project records and verified field records
Main output	Controlled dataset and activity-weight graph
Progression condition	Dataset Readiness Checklist passes

Lecture 1 — 45-Minute Facilitator Guide

Minute	Discussion and facilitator action
0–5	Explain why unreliable inputs produce unreliable CPM, PERT, and S-Curve conclusions
5–12	Explain planning, data gathering, monitoring, analysis, action, and reporting
12–20	Distinguish baseline, approved revision, actual, forecast, and proposed information
20–28	Explain project, activity, quantity, progress, resource, and constraint inputs
28–34	Explain evidence levels and data-status codes
34–40	Demonstrate workbook tabs, protected ranges, source references, and Change Log
40–45	Explain the activity-weight graph, required outputs, and quality gate

Reporting Cutoff

The exact date or reporting period through which status is measured. Planned progress, actual progress, constraints, evidence, and graphs must refer to the same cutoff.

Activity 1 Required Inputs

Project identity, contract number, location, implementing office, contractor, funding source, and amount.
Notice to Proceed, approved duration, original completion, approved revision, and reporting cutoff.
Activity ID, WBS code, description, predecessor, relationship, lag, duration, calendar, dates, and milestone.
Quantity, unit, approved cost or physical weight, crew, equipment, and source reference.
Actual dates, accepted quantity, verified completion, remaining quantity, inspection, test, and photograph references.
Constraint event, affected activity, duration or productivity effect, evidence, response, status, and follow-up owner.

Activity 1 Workflow

STEP-BY-STEP

Select one project and assign group roles.
Establish one reporting cutoff.
Inventory source records and assign evidence levels and status codes.
Complete the Project Registry using approved records only.
Build the activity list from the POW, BOQ, construction method, and approved schedule.
Assign unique IDs and measurable activity descriptions.
Enter relationships, lags, and milestones; flag broken logic.
Enter durations, dates, quantities, units, weights, crews, equipment, and sources.
Confirm that activity weights total 100 percent and document the weighting basis.
Enter actual accomplishment only when supported by accepted evidence.
Record each constraint and connect it to an activity and source.
Record missing or inconsistent data instead of guessing.
Generate the activity-weight versus earned-contribution graph.
Complete a two-person validation and record corrections in the Change Log.
Submit the workbook for the quality gate.

Earned Activity Contribution

\text{Earned Contribution}=\text{Activity Weight}\times\text{Verified Completion Fraction}

Activity 1 Required Output

Completed Project Registry.
Source inventory with evidence levels.
Validated activity and predecessor table.
Quantity and activity-weight table.
Planned and actual accomplishment table.
Activity-weight and earned-contribution graph.
Constraint and delay register.
Missing-data and inconsistency register.
Change Log and group validation record.

Common Errors

Error	Facilitator response
Broad activity such as “civil works”	Require measurable work packages
Proposed date entered as approved	Reclassify it and request approval evidence
Percentage entered without quantity support	Require an accepted measurement or inspection record
Weights forced to total 100 percent	Return to the approved weighting basis
Cause entered without an affected activity	Require Activity ID and evidence
Missing value replaced with an estimate	Restore `MIS` or `PEN` status
Graph values typed manually	Link the graph series to controlled cells

Dataset Readiness Checklist

Project identity and reporting cutoff are complete.
Source records are inventoried and classified.
Activity IDs are unique and descriptions are measurable.
Durations, calendars, and immediate predecessors are available.
Relationships contain no obvious circular or disconnected logic.
Weights total 100 percent and use a documented basis.
Actual accomplishment is supported by accepted evidence.
The activity-weight graph matches the controlled table.
Constraints are linked to activities and evidence.
Missing values are assigned for follow-up.
Data classes are not mixed.
At least two members reviewed the dataset.
No value was invented.

VII. Activity 2A — CPM and PERT Schedule Analysis

Activity Card

Item	Requirement
Introductory lecture and demonstration	45 minutes
Guided computation	60 minutes
Main input	Validated activity logic, duration, calendar, and risk estimates
Main output	CPM network graph and PERT probability graph
Progression condition	CPM-PERT Readiness Checklist passes

Lecture 2A — 45-Minute Facilitator Guide

Minute	Discussion and facilitator action
0–6	Review input requirements and scheduling conventions
6–14	Demonstrate logic audit, missing links, milestones, and procurement paths
14–23	Demonstrate forward pass and multiple-predecessor logic
23–31	Demonstrate backward pass, Total Float, Free Float, and critical paths
31–38	Explain PERT three-point estimates, expected duration, and variance
38–43	Explain path switching, project variance, Z-score, and probability curve
43–45	Explain graph annotations, assumptions, and the quality gate

Schedule Convention Rule

State whether the workbook uses a day-zero or day-one convention, whether durations are working or calendar days, and how milestones, holidays, weather allowances, lags, leads, and partial days are handled.

CPM Analysis Workflow

STEP-BY-STEP

Review every relationship for constructability and contractual logic.
Identify disconnected activities, missing links, circular logic, excessive lags, and missing milestones.
State the scheduling convention and calendar.
Perform the forward pass.
Use the maximum predecessor Early Finish for multiple finish-to-start predecessors.
Determine the earliest project completion.
Perform the backward pass.
Use the minimum successor Late Start for multiple successors.
Compute Total Float and Free Float.
Trace every zero-float critical path and near-critical path.
Identify the current controlling activity using actual status and remaining logic.
Generate the CPM network graph with critical, non-critical, and current-control distinctions.

CPM Calculations

EF=ES+d

ES_{successor}=\max(EF_{predecessors})

LS=LF-d

LF_{predecessor}=\min(LS_{successors})

TF=LS-ES=LF-EF

FF=\min(ES_{successors})-EF

PERT Risk Analysis Workflow

STEP-BY-STEP

Select activities with meaningful uncertainty.
Define optimistic, most likely, and pessimistic durations.
Record the uncertainty driver, source, date, and assumptions.
Challenge optimism, anchoring, and consensus bias.
Calculate expected duration and variance.
Replace baseline durations with expected durations and recalculate the network.
Identify whether the PERT path differs from the deterministic path.
Sum variances along the selected PERT critical path and state the simplifying assumptions.
Calculate project standard deviation and target Z-scores.
Convert Z-scores into probabilities.
Generate the PERT probability curve with mean and target markers.
Translate the result into monitoring and contingency actions.

PERT Calculations

t_e=\frac{a+4m+b}{6}

\sigma^2=\left(\frac{b-a}{6}\right)^2

V_p=\sum\sigma_{cp}^2

\sigma_p=\sqrt{V_p}

Z=\frac{T_d-T_e}{\sigma_p}

PERT Approximation Limitation

The basic path-variance method assumes a selected critical path, independent activity durations, and an approximate normal distribution. It may understate risk when activity durations are correlated or several near-critical paths can become controlling.

Activity 2A Required Output

Logic-audit table.
CPM network diagram and interactive graph.
Forward-pass and backward-pass table.
Total Float and Free Float table.
Critical, parallel-critical, and near-critical paths.
Current controlling-path statement.
Three-point estimate and basis table.
Deterministic-versus-PERT path comparison.
Variance, standard deviation, Z-score, and probability calculations.
PERT completion-probability graph.
Schedule-risk register and assumptions statement.

CPM-PERT Readiness Checklist

Calculations use the verified Activity 1 dataset.
Scheduling convention and calendar are stated.
Network logic is complete and free of circular dependencies.
Forward and backward passes are consistent.
Total Float and Free Float are not confused.
Critical and parallel-critical paths are identified.
Current control uses actual status, not baseline status alone.
The CPM graph matches the calculation table.
Three-point estimates have documented sources.
The PERT path is recalculated using expected durations.
The probability graph shows mean, target, and probability.
Probability is not presented as certainty or entitlement.

VIII. Activity 2B — S-Curve Monitoring and Project-Status Analysis

Activity Card

Item	Requirement
Focused lecture and demonstration	30 minutes
Guided analysis	45 minutes
Main input	CPM early/late dates, weights, reporting periods, and verified accomplishment
Main output	S-Curve envelope and optional Earned Value graph
Progression condition	S-Curve Readiness Checklist passes

Lecture 2B — 30-Minute Facilitator Guide

Minute	Discussion and facilitator action
0–5	Explain early-start, late-start, baseline, actual, and forecast curves
5–11	Explain cost, quantity, and physical-accomplishment weighting
11–17	Demonstrate value allocation and cumulative calculations
17–22	Explain variance, simplified SPI, and trend interpretation
22–26	Explain the banana curve and first boundary breach
26–30	Explain optional Earned Value and graph-quality requirements

Approved Allocation Methods

Method	Appropriate use	Required disclosure
Linear	Work is reasonably uniform	State even distribution
Quantity-based	Planned production varies by period	Show planned quantity per period
Milestone or weighted step	Value is earned after defined acceptance	State the acceptance rule
Resource or productivity-based	Output follows planned resources	Show productivity assumptions

No Arbitrary Smoothing

Do not alter the planned series to create a visually pleasing S-shape. The curve must result from the approved schedule, activity weights, allocation method, and reporting periods.

S-Curve Analysis Workflow

STEP-BY-STEP

Confirm one reporting interval and cutoff.
Verify that activity weights total 100 percent.
Select and document the allocation method for each activity.
Allocate value across the early-start schedule.
Repeat the allocation using late-start dates.
Enter only verified actual accomplishment.
Plot early-start, late-start, and actual curves.
Calculate variance and simplified SPI for each period.
Identify the first late-start-boundary breach.
Review CPM to identify controlling and contributing activities.
Connect causes to the constraint register and evidence.
Complete the optional PV-EV-AC analysis when verified cost data exist.
Write the graph interpretation and immediate response.

S-Curve and Earned Value Indicators

\text{Progress Variance}=\text{Actual Progress}-\text{Planned Progress}

SPI=\frac{\text{Earned Progress}}{\text{Planned Progress}}

SV=EV-PV

CV=EV-AC

CPI=\frac{EV}{AC}

Status Interpretation

Condition	Required interpretation
Actual above early-start curve	Ahead of the aggressive plan; verify quality, quantity, sequence, and front-loading
Actual between early and late curves	Float is being consumed; completion may remain recoverable
Actual below late-start curve	Planned flexibility is exhausted; immediate corrective action is required
Variance worsening	The response is absent, late, or ineffective
Variance improving but negative	Recovery is occurring, but the project remains behind
Positive schedule with poor quality or cost	The gain may be unsustainable or create rework

Activity 2B Required Output

Early-start, late-start, and actual progress tables.
Combined S-Curve envelope graph.
Reporting-cutoff and first-breach annotations.
Variance and simplified SPI table.
Current project-status and trend statement.
Activity-level cause analysis.
Preliminary corrective actions.
Optional PV-EV-AC table and bar graph.
Data and method limitations statement.

S-Curve Readiness Checklist

Planned and actual values use the same cutoff.
Activity weights total 100 percent.
Allocation methods are stated and consistently applied.
Actual accomplishment is verified.
Early and late curves are derived from CPM dates.
Variance is stated in percentage points.
The first boundary breach and current trend are identified.
The S-Curve graph matches the source table.
Cause statements identify activities and evidence.
Forecasts are labeled as forecasts.
The optional cost graph uses verified PV, EV, and AC at the same cutoff.

IX. Activity 3 — Integrated Project Output Development

Activity Card

Item	Requirement
Introductory lecture	45 minutes
Report-template demonstration	30 minutes
Guided report development	105 minutes
Main input	Validated data, CPM-PERT outputs, S-Curve, evidence, and limitations
Main output	Integrated report, action matrix, decision register, and recovery graph
Progression condition	Report Quality Checklist passes

Lecture 3 — 45-Minute Facilitator Guide

Minute	Discussion and facilitator action
0–6	Distinguish raw data, calculation, interpretation, and management information
6–13	Demonstrate a concise executive summary
13–20	Explain how to report CPM and PERT findings
20–27	Explain precise S-Curve and Earned Value language
27–34	Demonstrate evidence-linked cause analysis
34–40	Demonstrate corrective-action and decision registers
40–45	Explain recovery-graph assumptions, peer review, and acceptance criteria

Required Report Structure

Section	Required content
Project Profile	Identity, scope, amount, duration, cutoff, and status
Purpose and Basis	Reporting purpose, sources, conventions, and boundary
Data Quality	Evidence inventory, missing data, inconsistencies, and limitations
CPM Analysis	Logic, float, paths, current control, and network graph
PERT Analysis	Estimate basis, path risk, probability, and probability graph
S-Curve Analysis	Early, late, actual, variance, trend, and S-Curve graph
Earned Value	PV, EV, AC, indicators, and graph when applicable
Root-Cause Analysis	Evidence-based causes linked to activities
Corrective-Action Plan	Action, owner, deadline, resources, indicator, and trigger
Decision Register	Decision, owner, due date, required information, and status
Recovery Screening	Assumptions, scenario graph, effect, and limitations
Recommendations	Immediate action, validation, escalation, and reserved decisions
Appendices	Tables, graphs, calculations, photographs, and logs

Integrated Report Workflow

STEP-BY-STEP

Copy only validated values and source-referenced statements into the report.
State the reporting cutoff, purpose, scheduling convention, and analysis boundary.
Prepare an executive summary with planned progress, actual progress, variance, current control, causes, immediate action, and reserved decision.
Present CPM findings and distinguish the baseline critical path from the current controlling path.
Present PERT findings as probabilities and ranges rather than guarantees.
Present S-Curve and optional Earned Value findings with their controlled graphs.
Link every root-cause statement to an activity, event, and source.
Classify responsibility without deciding entitlement prematurely.
Prepare measurable corrective actions and a separate decision register.
Create a recovery-screening graph when a recovery scenario is recommended.
State assumptions, limitations, and required follow-up.
Exchange reports for peer review.
Resolve findings and preserve corrections in the Change Log.
Submit the report for the quality gate.

Corrective-Action Matrix Fields

Field	Instruction
Issue	State one specific project-control problem
Affected activity	Use Activity ID and description
Required action	Begin with an action verb and measurable result
Responsible party	Identify the accountable person, contractor, or office
Deadline	State a date or defined period
Required resources	Identify labor, equipment, information, approval, or budget
Expected effect	State the schedule, quality, cost, or coordination improvement
Monitoring indicator	Define what will be checked and how often
Escalation trigger	State the condition requiring higher management action

Decision Register Fields

Field	Required entry
Decision required	Exact technical or management decision
Decision owner	Authorized person or office
Due date	Latest useful decision date
Information required	Missing evidence, calculation, or approval
Risk of no decision	Schedule, cost, quality, safety, or coordination consequence
Status	Open, pending information, decided, or closed

Activity 3 Required Output

Integrated project-control report.
Executive summary.
CPM, PERT, S-Curve, and optional Earned Value sections with graphs.
Root-cause table.
Corrective-action matrix.
Decision register.
Recovery-screening calculation and timeline graph.
Management recommendations.
Limitations and information-needs statement.
Peer-review log.
Presentation-ready visual package.

Report Quality Checklist

All values match the controlled calculation sheets.
Graphs match their source tables and reporting cutoff.
Critical and controlling-path statements match the network and actual status.
PERT findings are probabilities, not guarantees.
Root causes are evidence-based.
Corrective actions are measurable and assigned.
Decisions are separated from recommendations.
Recovery assumptions are visible beside the graph.
Contract entitlement is not concluded without the required analysis.
Forecasts are not presented as approved dates.
Limitations identify both effect and follow-up.
Peer-review findings are resolved or explicitly recorded.

X. Activity 4 — AI-Assisted Review, Drafting, and Validation

Activity Card

Item	Requirement
Introductory lecture	45 minutes
AI-assisted drafting	60 minutes
Human validation	45 minutes
Main input	Sanitized and verified report data
Main output	Original AI response, validation log, and corrected report
Progression condition	Responsible AI Use Checklist passes

Lecture 4 — 45-Minute Facilitator Guide

Minute	Discussion and facilitator action
0–7	Explain acceptable and prohibited AI tasks
7–13	Explain data classification, redaction, and approved tools
13–21	Demonstrate the six-part prompt structure
21–28	Demonstrate hallucination, arithmetic inconsistency, and unsupported causation
28–35	Demonstrate claim-by-claim verification
35–40	Demonstrate a challenge prompt and second-pass review
40–45	Explain output preservation, human sign-off, and presentation requirements

Appropriate AI Tasks

Identify blank, duplicated, or inconsistent fields for human review.
Check a supplied activity table for apparent missing links.
Convert verified calculations into plain-language draft text.
Organize documented causes and actions into a requested format.
Draft an executive summary or presentation outline.
Challenge a draft for unsupported claims, inconsistent values, or missing limitations.

Prohibited AI Tasks

Invent missing dates, durations, quantities, costs, causes, or accomplishment.
Replace CPM, PERT, S-Curve, Earned Value, quantity, cost, or contract calculations.
Approve a schedule, progress claim, extension, variation, or entitlement.
Determine physical accomplishment without accepted evidence.
Treat generic construction practice as proof of project events.
Upload confidential data to an unapproved tool.
Transfer AI output without claim-level human verification.

Six-Part Structured Prompt

Role: define a limited assistant role.
Verified source data: supply sanitized and source-referenced values.
Task: request one bounded review or drafting objective.
Output format: specify a table, checklist, narrative, or outline.
Restrictions: prohibit invention, recalculation, approval, entitlement, and unsupported causation.
Validation instruction: require missing, inconsistent, uncertain, and unsupported items to be listed separately.

AI-Assisted Review Workflow

STEP-BY-STEP

Select one limited AI task and identify the human reviewer.
Verify every supplied value against the controlled workbook.
Remove confidential and prohibited information.
Prepare the six-part prompt.
Save the prompt, supplied data, tool, date, and original response.
Divide the response into facts, calculations, interpretations, forecasts, and recommendations.
Verify every material statement against the project records and graphs.
Classify each statement as accepted, revised, unsupported, incorrect, requiring information, or rejected.
Run a challenge review as an additional aid.
Record the correction and verification source.
Transfer only validated content to the report.
Identify the accountable human reviewer and final engineering judgment.

AI Validation Log Fields

Field	Required entry
AI task	Limited purpose
Tool and date	Tool used and generation date
Prompt	Complete prompt
Supplied data	Sanitized verified inputs
AI statement	One claim, calculation, forecast, or recommendation
Statement type	Fact, calculation, interpretation, forecast, or recommendation
Verification source	Workbook, graph, report, contract, inspection, or test
Validation result	Accepted, revised, unsupported, incorrect, more information, or rejected
Correction	Final accurate wording
Reviewer	Accountable person or role

Activity 4 Required Output

Structured prompt and sanitized input package.
Original AI response.
Challenge-review response.
Claim-level validation log.
Corrected management narrative.
Final integrated report.
Presentation outline.
Human engineering-judgment statement.

Responsible AI Use Checklist

Source data were verified before prompting.
Confidential information was removed or handled through an approved system.
AI did not replace deterministic or contractual analysis.
The original prompt, supplied data, and response were preserved.
Every material statement was independently checked.
Unsupported causation, forecasts, and entitlement conclusions were removed.
Recommendations were reviewed for feasibility, safety, quality, resources, and contract consistency.
The final report identifies the accountable human reviewer.

XI. Final Presentation and Submission

Required Presentation Sequence

Project profile and reporting cutoff.
Data completeness, evidence quality, and limitations.
Activity-weight and earned-contribution graph.
CPM network, critical paths, and current controlling activity.
PERT probability graph and target confidence.
S-Curve envelope, boundary breach, variance, and trend.
Optional Earned Value graph.
Root causes, corrective actions, decisions, and recovery graph.
AI task, errors identified, and corrections.
Final engineering judgment.

Suggested Nine-Minute Allocation

Segment	Time
Project and data basis	1 minute
CPM and PERT findings	2 minutes
S-Curve and Earned Value	2 minutes
Causes, actions, decisions, and recovery	2 minutes
AI validation	1 minute
Final judgment	1 minute

Final Submission Package

Controlled centralized Google Sheet.
Project Registry and source inventory.
Activity, predecessor, duration, quantity, weight, and resource table.
Actual-accomplishment, constraint, missing-data, and Change Logs.
Activity-weight and earned-contribution graph.
Logic audit, CPM network, forward pass, backward pass, Total Float, and Free Float.
CPM network graph.
PERT estimates, path comparison, variance, standard deviation, and probabilities.
PERT probability graph.
Early-start, late-start, and actual S-Curve dataset and graph.
Progress variance and optional Earned Value indicators and graph.
Root-cause analysis.
Corrective-action and decision registers.
Recovery-screening calculation and timeline graph.
Integrated report and peer-review log.
AI prompt, original response, challenge response, and validation log.
Final corrected report and presentation.

XII. Evaluation Rubric

Criterion	Weight	Exemplary evidence	Needs major improvement
Data completeness, traceability, and control	20%	Values are source-referenced, status-coded, versioned, and consistent	Missing sources, mixed data classes, or invented values
CPM logic and calculation	20%	Constructible network, correct passes, both float types, all paths, and current control	Broken logic, arithmetic errors, or unsupported path
PERT risk analysis	15%	Estimate basis, recalculated path, assumptions, probability, and action implications	Arbitrary estimates or probability presented as certainty
S-Curve and optional Earned Value	15%	Documented allocation, clear graphs, precise variance, trend, and valid cost indicators	Arbitrary curve, mismatched cutoff, or unsupported actual progress
Root cause, recovery, and decisions	15%	Evidence-linked causes, measurable actions, triggers, and decisions	Generic actions, no owner, or premature entitlement conclusion
Report, graph quality, and presentation	10%	Clear, accessible, consistent, decision-useful, and limitation-aware	Unreadable graphs or raw tables without a conclusion
Responsible AI use	5%	Preserved outputs, claim-level validation, corrections, and human sign-off	AI text accepted without checking
Total	100%

XIII. Contingency Arrangements

No Suitable Actual Project

Use the complete fictional project from the companion example. Assign a controlled variation such as a different reporting cutoff, resource constraint, missing record, or target duration. Participants must still complete every validation and interpretation step.

Internet or Google Sheet Access Is Unavailable

Use the downloaded spreadsheet copy. Assign one group recorder, apply local file versioning, and upload the completed workbook after connectivity is restored.

Project Records Are Materially Incomplete

Do not fabricate inputs. Complete the source and missing-data registers, use only the verified subset for limited conclusions, state what cannot be concluded, and use the fictional dataset for calculation and graphing practice.

Activity Time Is Insufficient

Prioritize data validation, calculations, graph accuracy, and decision-useful outputs. Incomplete but traceable work is preferable to a polished report based on unsupported information.

XIV. Facilitator Debrief

Debrief Questions

Which input created the greatest uncertainty?
What changed after the schedule logic was reviewed?
Did the PERT path differ from the deterministic path?
When did actual progress first leave the late-start boundary?
Which graph most clearly communicated the management issue, and why?
Which cause had the strongest evidence, and which remained uncertain?
Which corrective action changes the controlling path rather than a non-critical activity?
What decision is required first, and what information is still needed?
Which AI statement sounded plausible but failed verification?
What should be updated at the next reporting cutoff?

Key Takeaways

Module 6 is one controlled workflow rather than a set of unrelated exercises.
Tables provide the controlled evidence and calculations; graphs communicate the resulting pattern and decision.
Activity 1 establishes a traceable dataset and activity-weight graph.
Activity 2A identifies logic, float, critical paths, uncertainty, and target confidence through CPM and PERT graphs.
Activity 2B measures variance magnitude, trend, and envelope position through the S-Curve and optional Earned Value graph.
Activity 3 translates technical findings into actions, decisions, and a clearly limited recovery screen.
Activity 4 uses AI only as a bounded drafting and review aid under claim-level human validation.
Every graph and conclusion must remain traceable to approved records, reproducible calculations, and verified field evidence.

Previous TopicAI Integration and Culminating Workshop - Worked Examples

Next TopicGuided Workshop Activities and Integrated Project Output - Complete Worked Example

Prev Next