15-Session Programme · 25 Minutes Per Day · Free · For the owners and operators of Nigeria’s energy assets
WEEK 3 OF 3 · LIFECYCLE, TRUST, CAPITAL AND THE LEADERSHIP MANDATE · SESSIONS 11–15 · WORK AT YOUR OWN PACE
Chapters 12 and 13 · Lifecycle ChoicesSESSION 11
Life Extension Must Be Earned
Why This Matters
Every asset you own will one day face two questions: can it earn more years of service, and what does the owner still owe when that service stops? Chapter 12 says life extension must be earned, not assumed because the facilities exist and the resource may still have value. Chapter 13 says retirement is an ownership decision: the end of production does not end the owner's responsibilities. Both reward the leader who starts early, while good alternatives still exist. Today you carry your decision in both directions: the future duty the asset would have to deliver, and the obligations, records and funding that remain when its service ends. Neither question is comfortable. Both belong in the asset review now.
25-Minute Module
0:00 – 5:00
Define the future duty before the economics
- The commercial case for continued use can be compelling. It does not establish whether the equipment can deliver the proposed future duty. Life extension is a technical, operational and ownership decision as well as an economic one.
- Specify the proposed future service: operating period, expected duty, configuration, environmental assumptions and supporting infrastructure. A facility can become a hub for other production or serve a changed feed or product; the assessment must address that service, not merely repeat the historical one.
- For upstream facilities, NUPRC's September 2026 Guidelines for Life Extension Management of Ageing Oil and Gas Facilities address a Life Extension Programme and the Commission's consent and approval process. It states that facilities should not continue beyond design life without the required consent and approval, and calls for planning and assessment no later than five years before anticipated design-life expiry. The scope includes onshore, swamp, shallow-water and deepwater upstream facilities.
- THE DOCTRINE
That is a specific upstream requirement, not a universal rule for all petroleum equipment. Establish the applicable position for your facility and retain the authoritative documents and advice behind it.
- Begin early enough to preserve choices. The cost of delay can be the loss of a good alternative, not simply the cost of another study.
If nobody can tell you when your asset's design life is expected to expire, that is your first finding.
5:00 – 10:00
Ageing is more than metal loss
- Physical deterioration may coexist with obsolete control equipment, unavailable spares, changed duty, incomplete modification records or declining specialist support. Ask each discipline what could prevent the future service and keep the answers in one lifecycle account. If structural work is feasible but a supporting control system cannot be maintained, the option must reflect both.
- Obsolescence, plainly: an obsolete component is not necessarily failed. The exposure may be that it cannot be repaired or replaced within the required restoration time. Identify the installed population, support status, spares, compatibility and upgrade path, and verify spare condition rather than counting boxes.
- Compare supported replacement, controlled stockholding, redesign or retirement, including integration, testing and workforce competence. An upgrade can also alter the conditions on which an analytics application was validated.
- THE DOCTRINE
Do not allow a supplier's sales timetable to become the owner's lifecycle strategy. The decision should remain understandable after that supplier's representative changes.
- AI can search inspection histories and modification records and highlight gaps. It cannot establish absent design evidence by generating plausible text.
A spare you have not verified is a hope, not a contingency.
10:00 – 14:00
Explicit gates, and economics after the scope is known
- The author proposes gates: agreed future duty, accepted evidence baseline, identified technical gaps, feasible interventions, a funded execution plan and the required approvals. They do not replace the regulator's process or the engineering assessment. At each, state what is established and what remains conditional.
This commitment Is not Approval to perform an assessment Approval to continue operating for the extended period Completion of a study Completion of the interventions it recommends An accounting provision Evidence that funds are available when the work must be performed Approval to develop a retirement plan Approval to execute the work - AT THE ASSET IN THE FIELD · THE WORKED CASE
- In the field. A fictional onshore operator's integrity team finds the vessels and piping feasible to extend. The instrument technicians report the supporting control system obsolete and its spares unverified. One option, two constraints, one lifecycle account.
IN THE BOARDROOM · THE PARALLEL- In the boardroom. The board pack reads life-extension assessment approved. The board asks which gates have passed, what remains conditional and what the alternative path is, and records the approval as exactly that.
- Keep an alternative path if extension is not justified; it reduces the pressure to read uncertain evidence in favour of continued operation. Once interventions are known, update costs, outage assumptions and residual obligations. Sunk acquisition cost does not make continued use economical.
Ask for the gate states on one page. The technical appendices can follow.
14:00 – 21:00
Retirement is an ownership decision
- The end of production does not end the owner's responsibilities: equipment can remain hazardous, structures deteriorate and records become harder to retrieve as teams disperse. Decommissioning, abandonment, removal, remediation and monitoring have specific meanings in the applicable legal and technical arrangements.
- The Petroleum Industry Act 2021 addresses abandonment, decommissioning and disposal in section 232 and a decommissioning and abandonment fund in section 233; the detailed obligations require facility-specific review. IAS 37 addresses provisions and related disclosures, including obligations involving restoration.
- THE DOCTRINE
An accounting provision and available cash are different things. A figure in the accounts does not by itself demonstrate that funds are available when the work must be performed.
- Have engineering, finance and legal reconcile scope, cost basis, timing and funding, explaining differences rather than forcing one number. Compare continued operation, changed use, mothballing and retirement on a common lifecycle boundary. Mothballing is not a condition with no maintenance, and a changed use does not inherit acceptance simply because the structure is available.
- Build an evidence inventory before the team leaves: configuration, materials, modifications, hazardous inventories, inspection histories, interventions and known uncertainties. AI can catalogue an archive; a generated summary should not replace it. Capture specialist knowledge, separating facts, interpretations and open questions.
- The retirement paper states why retirement is considered, the alternatives, any interim condition, the work, approvals and funding path. An assumed salvage credit needs evidence. Identify engagement responsibilities to people and communities, and never use AI-generated reassurance to replace an answer the organisation owes.
A disused installation should not become an orphan because the production organisation has moved on.
21:00 – 25:00
Carry both charges, then write Section 11
- Chapter 12's charge: begin while alternatives remain available. Require a defined future duty, accepted evidence, funded interventions and the applicable approvals before treating extended life as part of the business plan.
- Chapter 13's charge: bring end-of-life scope, records and funding into the asset review before cessation becomes urgent. Plan retirement with the same seriousness as the investment that began the asset's service.
- If your asset is far from end of life, apply both chapters to the system it serves. Today's assignment produces Section 11 of your plan.
The board that hears about end of life early still has choices.
Key Terms
Proposed future service
The period, duty, configuration, environment and supporting infrastructure an extension must deliver.
Obsolescence exposure
The risk that a component, not necessarily failed, cannot be repaired or replaced within the required restoration time.
Decision gate
The author's proposed checkpoint stating what is established and what remains conditional; not a substitute for the regulator's process.
Provision versus available cash
An accounting recognition of an obligation (IAS 37) does not show that funds will exist when the work is due.
Self-Check — Before You Submit
1. Which statement matches the book's account of NUPRC's September 2026 life-extension guideline?
- A. It applies to every petroleum installation in Nigeria
- B. It calls for planning and assessment no later than five years before anticipated design-life expiry for upstream facilities
- C. It sets the remaining-life calculation an engineer must use
- D. It allows operation beyond design life once the board approves
B. It calls for planning and assessment no later than five years before anticipated design-life expiry for upstream facilities — The book reports the five-year point and the upstream scope, and calls it a specific upstream requirement. Extending it to every installation is the tempting error.
2. A fictional terminal owner's finance director reports a decommissioning provision and concludes retirement is funded. What do you ask for?
- A. Nothing; the provision settles it
- B. A larger provision
- C. An AI summary of the obligations
- D. Reconciliation of scope, cost basis, timing and the funding account by engineering, finance and legal
D. Reconciliation of scope, cost basis, timing and the funding account by engineering, finance and legal — A provision and available cash are different things. A larger provision feels prudent but still says nothing about cash when the work falls due.
3. A critical control system is obsolete but working, and the supplier proposes an upgrade on its sales timetable. What is the owner's first move?
- A. Establish population, support status, verified spares and upgrade path, then compare the options
- B. Accept the supplier's timetable
- C. Wait until it fails
- D. Count the spares in the store and record the risk as covered
A. Establish population, support status, verified spares and upgrade path, then compare the options — Obsolete is not failed; the exposure may be restoration time. Counting boxes is the trap: the book requires spare condition and suitability to be verified.
Today’s Assignment
Write Section 11 of your plan: the lifecycle case for your decision's asset in both directions.
- Future duty. The proposed future service and where it differs from the original design. If the asset is upstream, note how NUPRC's life-extension guideline may apply, marked for confirmation by your advisers.
- Beyond metal loss. The ageing and obsolescence constraints each discipline would raise, in one lifecycle account.
- Gates and alternative. The six proposed gates with what is established and conditional at each, and the alternative path if extension is not justified.
- End of life. Obligations to keep visible; any provision against the funding path; options compared on a common lifecycle boundary; the evidence inventory; and the people and communities affected, with owners by role.
- Be candid about what your company gets wrong today. If your asset is far from end of life, apply this to the system it serves. Label assumptions.
YOUR 90-DAY PLAN
Section 11 — The future duty and the end of life: extension gates, obsolescence, retirement obligations and funding.
How this is marked — the rubric
| Defines the future service and constraints beyond metal loss in one account, treating the upstream guideline only as the book states it | 30 |
| Sets out the gates with established and conditional items and a credible alternative path | 25 |
| Separates provision from funding and compares options on a common lifecycle boundary | 25 |
| Builds the evidence inventory and names affected people and communities, with owners by role | 20 |
| Total · pass mark 70 | 100 |
Submitting session 11
Write your answer in Word or Google Docs. The first line must read
Write your answer in Word or Google Docs. The first line must read
Session: 11, or the submission is refused. Upload the Word file (.docx), or switch to the Google Doc tab and paste your document’s share link.Marked against the rubric above and returned as a written letter, saved to your progress page. Do not submit until you have passed the self-check above.Capabilities Covered
Future-duty definitionAgeing and obsolescenceExtension gatesProvision versus cashRetirement planning
Chapter 14 and Appendix C · Trust, Capital and PeopleSESSION 12
Govern the Intelligence You Rely On
Why This Matters
An AI application can enter asset work quietly: a summary pasted into a report, a question put to an assistant, a supplier's dashboard. Chapter 14 asks whether you could describe each application's purpose in one operational sentence. If the honest description is that it makes the company intelligent, its boundary is unclear, and so is the owner's exposure. Today you govern the application behind your decision, or the one proposed in Section 1: who owns it, what it may and may not do, how a competent person can disagree with it, how it fails safely and what record shows it was accepted. If your company has no AI in use yet, that is normal: govern the proposal before anyone buys it.
25-Minute Module
0:00 – 5:00
One operational sentence, named owners
- State the purpose in one operational sentence: it retrieves accepted inspection evidence for a named reviewer; it flags a defined condition for investigation; it checks whether a work package contains specified documents.
- NIST's AI Risk Management Framework (AI RMF 1.0) is a voluntary, cross-sector resource, not a Nigerian operating approval. Use it to support your own accountable arrangements; the asset's technical, legal and security obligations remain in force.
- Name a business owner, a technical owner and a review authority. State the equipment population, conditions, data sources, permitted users and actions, and the decisions for which the output must not be used.
- THE DOCTRINE
Keep the register proportionate to consequence. A document assistant and a safety-related system do not belong under one undifferentiated approval.
- The register holds version, validation record, known limitations, monitoring, access permissions and fallback. You should be able to withdraw the application without losing its records; a contract exit plan is part of operational resilience.
If two managers describe the same application's purpose differently, you have found your first governance task.
5:00 – 10:00
Source discipline, and a review that is more than a button
- NIST's Generative AI Profile describes confidently presented false content and excessive reliance on automated output. For an owner that can be a summary that drops a qualification, cites the wrong revision or supplies a value not in the report.
- Require source-linked answers for consequential document work. Test the absent answer: a confident response to an unanswerable question is a defect. Let the record holders build the test set, with conflicting revisions, negation, scanned tables, unit changes and proposed rather than accepted actions. A system can copy a value correctly and still use it in the wrong decision.
- THE DOCTRINE
If the reviewer cannot access the source, has no time to examine it or lacks authority to reject the output, the approval step is weak.
- Show evidence and uncertainty before inviting approval, keep the source visible, and record corrections and reasons for rejection. Do not punish a reviewer for declining a recommendation with a sound basis; ask why the disagreement occurred.
Ask to see the rejected outputs and the reasons. That record tells you whether the review is real.
10:00 – 15:00
Operational technology and information about people
- NIST SP 800-82 Revision 3 addresses OT security while recognising performance, reliability and safety needs. At the book's research date NIST identified Revision 4 as an initial public draft, not a final replacement. Read-only access reduces some risks but not all: a connection can still affect availability, disclose sensitive information or feed misleading data to a trusted workflow.
- AT THE ASSET IN THE FIELD · THE WORKED CASE
- In the field. A fictional gas-plant team proposes an advisory tool reading process data to flag a defined condition. Specialists test loss of data and loss of service; operators confirm the manual fallback. Any write action becomes a separate request with its own authority and testing.
IN THE BOARDROOM · THE PARALLEL- In the boardroom. The register shows the same tool as advisory and read-only, with named owners, a fallback and suspension conditions. The board hears plainly that it is not a route into control functions.
- THE DOCTRINE
Keep advisory tools from becoming informal routes into control functions. The application should fail in a way the operating team understands, with a known fallback.
- Asset records hold personal information: names, competence records, photographs, investigation material. NDPC identifies the Nigeria Data Protection Act 2023 as the national framework it implements. Limit data to the approved purpose, control access and retention, and examine provider terms on storage, training use and cross-border handling with qualified advisers.
A broad upload can include much more than the engineer intended. Ask what was actually sent.
15:00 – 21:00
The acceptance and change record
- Appendix C proposes management checks, not certification criteria; the competent owners set the thresholds for the actual use.
Test situation Evidence the owner should require The source has no answer Insufficient evidence identified, not a plausible value Two revisions conflict Conflict visible, source and revision preserved Units or decimals differ Values traceable; consequential extraction verified A report contains a limitation Negation, inaccessible coverage and conditions survive Data arrive late Staleness visible; permitted fallback followed Test situation Evidence the owner should require The service is unavailable Users continue through the accepted arrangement An unauthorised action is requested Permissions prevent it; request recorded Equipment duty changes Reviewed before reliance beyond validated conditions A supplier changes the version Change and any revalidation recorded An output is rejected Reason captured and reviewed by a responsible person - Gate states are pending, passed, conditional and failed. A conditional gate needs a limit, owner and review or expiry condition. A failed necessary gate is not cancelled by a high score elsewhere.
- A controlled prompt can confine retrieval to one equipment identity, show source and revision, separate observed values from estimates and list conflicts unreconciled. It is not a security control or proof of reliability.
A conditional pass without an expiry date is a permanent pass that nobody approved.
21:00 – 25:00
Review after change, then write Section 12
- A new model version, retrieval index, data source, extraction method, equipment duty or permitted action can affect performance; define which changes trigger revalidation. Keep a change record, a route to the last accepted configuration and a route to suspend. Record incidents and near misses; quiet correction deprives the owner of evidence. On retirement, keep needed records, remove access and name who does the work.
- The charge: authorise a specific use with named owners, tested evidence and a workable fallback. Make it possible for a competent person to disagree with the machine.
- Today's assignment produces Section 12 of your plan.
Govern the application you will actually run, not the one in the brochure.
Key Terms
Permitted use
The decision an application supports, its population, users and actions, and the decisions it must not inform.
Source-linked answer
An output that lets the reviewer open the source and revision behind each claim.
Gate state
Pending, passed, conditional or failed; a conditional state carries a limit, owner and review or expiry condition (Appendix C).
Revalidation trigger
A change, such as a new model version, data source or equipment duty, requiring re-testing before continued reliance.
Self-Check — Before You Submit
1. A fictional terminal owner's assistant gives a confident wall-thickness value the inspection report does not contain. The supplier calls it plausible. How should the owner treat it?
- A. As acceptable, because it was plausible
- B. As the supplier's matter alone
- C. As a defect to record and review: the application should identify insufficient evidence
- D. As a reason to add a disclaimer
C. As a defect to record and review: the application should identify insufficient evidence — A confident answer to an unanswerable question is a defect. Plausibility is the trap: Appendix C requires insufficient evidence to be identified, not a plausible value supplied.
2. Which description of NIST SP 800-82 matches the book?
- A. Revision 3 addresses OT security; Revision 4 was an initial public draft at the book's research date, not the final replacement
- B. Revision 4 is final and replaces Revision 3
- C. It is a Nigerian operating approval
- D. It shows that read-only access removes all risk
A. Revision 3 addresses OT security; Revision 4 was an initial public draft at the book's research date, not the final replacement — The book cites Revision 3 and notes Revision 4 as a draft at its research date. Calling the draft final is the easy error; read-only access reduces some risks, not all.
3. An application passes nine Appendix C tests but fails the unauthorised-action test. The sponsor proposes release because the overall result is strong. What do you decide?
- A. Release; nine of ten is strong
- B. Release as conditional with no expiry
- C. Average the results
- D. Hold release for that use: a failed necessary gate is not cancelled by a high score elsewhere
D. Hold release for that use: a failed necessary gate is not cancelled by a high score elsewhere — Appendix C says a failed necessary gate is not cancelled by a high score elsewhere. An open-ended conditional release looks like compromise but becomes permanent permission through inattention.
Today’s Assignment
Write Section 12 of your plan: govern the AI application your company would rely on for your decision, or the one proposed in Section 1.
- Its purpose in one operational sentence; business owner, technical owner and review authority by role; permitted and prohibited uses; and its register entry with fallback and exit arrangement.
- The source-discipline tests your record holders would set, and a human review the reviewer can use to reject output without penalty.
- The OT controls and manual fallback, the personal information involved and how it is limited, and the revalidation triggers.
- A completed Appendix C record for the ten test situations, with each gate's state and, for any conditional gate, its limit, owner and review or expiry condition.
- Say honestly where your company accepts AI output without a real review today. If no AI is in use, govern the proposal, and say whether a simpler method would do.
YOUR 90-DAY PLAN
Section 12 — The application governed: owners, permitted use, review, fallback and the acceptance and change record.
How this is marked — the rubric
| One-sentence purpose, named owners, permitted and prohibited uses and a proportionate register entry | 25 |
| Source-discipline tests and a human review a competent person can use to reject output | 25 |
| OT controls, fallback, personal information and revalidation triggers, citing guidance only with the status the book gives it | 20 |
| Appendix C record with a state for every gate and limit, owner and review or expiry condition for each conditional one | 30 |
| Total · pass mark 70 | 100 |
Submitting session 12
Write your answer in Word or Google Docs. The first line must read
Write your answer in Word or Google Docs. The first line must read
Session: 12, or the submission is refused. Upload the Word file (.docx), or switch to the Google Doc tab and paste your document’s share link.Marked against the rubric above and returned as a written letter, saved to your progress page. Do not submit until you have passed the self-check above.Capabilities Covered
Application ownershipSource disciplineMeaningful human reviewOT and privacy boundariesAcceptance and change control
Chapter 15 and Appendix D · Trust, Capital and PeopleSESSION 13
Capital Allocation That Survives the Downside
Why This Matters
Capital is where uncertain evidence most easily gets dressed up as a confident number. Chapter 15 makes a plain demand: compare only options the technical authority has accepted, on one consistent lifecycle basis, and put the downside and the funding beside the answer. Today you work through the book's fictional three-option case, check its arithmetic, and see why a ranking that survives a change in discount rate can still be fragile, and why the option with the lowest present cost may not be the one the business can pay for. Then you build the capital case for your own decision and run the Investment Committee Checklist against it. Pop Quiz 3 opens at the end.
25-Minute Module
0:00 – 4:00
Technical gates first, then a clear question
- The financial case begins after the technical gates. If a condition is unacceptable for continued operation, a positive return on doing nothing does not make that option feasible.
- Name the question: lowest lifecycle cost for a required service, investment in additional capacity, or continued operation against retirement. A calculation that answers one should not be presented as settling another.
- In plain language: present cost converts each future cost into today's terms, because money paid later weighs less than money paid now, and adds them up. The real discount rate is the conversion rate when amounts are in constant money, with inflation taken out. Residual value is an assumed or evidenced receipt at the end of the period, not a disposal obligation.
- THE DOCTRINE
Legal conditions, protection requirements and engineering acceptance are necessary gates. A large commercial benefit does not compensate for failing them.
If an option is in the spreadsheet, ask who accepted it technically, and for what duty.
4:00 – 10:00
The book's worked lifecycle decision
- The case is fictional: its costs, timings, residual values and rate are teaching assumptions, not Nigerian market quotations or observed performance. Each option is assumed technically accepted for its duty and duration. Five years of the same service: A repairs now and intervenes in year three; B replaces now; C performs an interim intervention now and replaces at the end of year one. Constant US dollars in millions, an assumed 10% real rate, a pre-tax, unlevered screen; costs positive, residual receipts negative.
Cash-flow timing A Repair then intervene B Replace now C Interim then replace Time zero 0.420 1.100 0.180 End of year 1 0.250 0.120 1.350 End of year 2 0.250 0.120 0.120 End of year 3 1.150 0.120 0.120 End of year 4 0.250 0.120 0.120 End of year 5 0.200 -0.030 -0.030 - A's year three adds a 0.900 intervention to 0.250; its year five subtracts a 0.050 receipt. B's year five is 0.120 less a 0.150 receipt: a net receipt, not negative maintenance. C's year one is a 1.120 replacement plus 0.230: a postponed investment obligation, not an avoided cost.
- Divide each amount by one plus the rate raised to the year number and sum. Appendix D's factors at 10%: 1.0000, then about 0.9091, 0.8264, 0.7513, 0.6830 and 0.6209. Round only the final result.
- At 10%: A USD 2.0128 million, B 1.4618, C 1.6599. B is lower than A by about USD 0.5511 million and than C by 0.1982: differences in assumed present cost, not demonstrated AI savings or approved returns.
Recalculate the preferred option yourself before you argue about it.
10:00 – 15:00
Examine the downside
Assumed real rate A present cost B present cost C present cost 0% 2.5200 1.5500 1.8600 10% 2.0128 1.4618 1.6599 20% 1.6684 1.3986 1.5036 - Zero percent is undiscounted, not a recommended commercial rate. The ranking holds, but that does not establish robustness to every uncertainty; scope and delivery can matter more.
- Break-even: at 10%, B's initial cost could rise from 1.100 to about 1.2982 before it equals C, all else fixed. At 1.300, B would be about 1.6618, slightly above C's 1.6599. That is for discussion, not a price forecast.
- THE DOCTRINE
A narrow difference should prompt scrutiny of assumptions rather than a confident selection based on rounding. The owner needs the most credible complete option.
- The screen recommends investigating B for sanction, subject to technical scope, execution, funding and the omitted cash flows. It is not permission to place an order.
Ask how far the winning option's cost must move before it stops winning.
15:00 – 21:00
Affordability, and what AI may contribute
- Affordability is a separate question. B needs 1.100 now; C needs 0.180 now and 1.350 at the end of year one. A business may rationally prefer C's profile, but the deferred commitment must be funded and its technical conditions maintained. C is not a low-cost permanent solution.
- For an actual Nigerian project, build currency-specific cash schedules on an approved conversion and escalation basis. A naira payment, a dollar quote and a real discount rate cannot be combined casually.
- AT THE ASSET IN THE FIELD · THE WORKED CASE
- In the field. Set the case at a fictional gathering system. Acceptance of C depends on the interim intervention holding its conditions until replacement; the field record must show them maintained.
IN THE BOARDROOM · THE PARALLEL- In the boardroom. The committee prefers C's lower first-year outlay. The paper shows the 1.350 commitment, its decision date and fallback, so it cannot vanish outside the current budget year.
- AI can assemble option dossiers, find missing scope and draft sensitivity questions. A controlled calculation engine does the arithmetic, and the analyst keeps the model version used at the meeting.
- THE DOCTRINE
Do not let a language model invent a failure probability or residual value because the financial model needs one. Where evidence is thin, present scenarios and judgement transparently.
State what would change the recommendation: a firm delivery commitment, a new engineering finding or a funding constraint.
21:00 – 25:00
Run the checklist, write Section 13, sit Pop Quiz 3
- Appendix D, the Investment Committee Checklist: read the recommendation before the model. Check the boundary (service, period, currency, real or nominal, rate, timing, tax, end of life); the arithmetic (inputs evidenced or labelled, signs, units, benefits counted once); uncertainty and funding (downside, delay case, funding trough); and the approval boundary (approved now, conditional, technical acceptance separate from sanction, model version kept).
- The charge: compare only feasible options on a common lifecycle basis. Put funding, downside and break-even assumptions beside the preferred result, and count benefits once. Today's assignment produces Section 13.
- Pop Quiz 3 opens now: five questions on Sessions 11 to 13, closed book. Sit it before Session 14.
A handwritten change that never reaches the model leaves two inconsistent approval records.
Key Terms
Present cost
An option's costs converted into today's terms at a stated rate and summed; time zero is not discounted.
Real discount rate
The rate for constant-money amounts, with inflation removed; never mixed casually with nominal figures.
Residual value
An assumed or evidenced receipt at the end of the period, not a disposal obligation (Appendix E).
Affordability
Whether the business can fund an option's commitments when due; a favourable present value does not establish it.
Self-Check — Before You Submit
1. In the book's fictional case, a committee member says B is clearly best because it ranks first at 0%, 10% and 20%. What is the right response?
- A. Agree; three rates prove robustness
- B. The ranking holds, but scope and delivery can matter more, and B's initial cost need only reach about 1.2982 to equal C
- C. Switch to C for its lower first-year outlay
- D. Use 0% as the commercial rate
B. The ranking holds, but scope and delivery can matter more, and B's initial cost need only reach about 1.2982 to equal C — An unchanged ranking does not establish robustness to every uncertainty. Treating three rate scenarios as proof is the trap; the break-even shows how narrow the margin over C is.
2. In the fictional case, what is C's year-one figure of 1.350?
- A. An avoided cost
- B. Negative maintenance expenditure
- C. A residual receipt
- D. A postponed investment obligation: a 1.120 replacement plus 0.230 of annual cost
D. A postponed investment obligation: a 1.120 replacement plus 0.230 of annual cost — The book calls C's year-one peak a postponed investment obligation, not an avoided cost. The negative values in the table are the year-five residual receipts.
3. Your analyst needs a failure probability for option A that the evidence does not support, and proposes asking a language model for one. What do you decide?
- A. Decline: present scenarios and judgement, or give any estimate a professional basis and stated uncertainty
- B. Accept it if it looks reasonable
- C. Average several model estimates
- D. Use it with a footnote
A. Decline: present scenarios and judgement, or give any estimate a professional basis and stated uncertainty — Chapter 15 forbids letting a language model invent a probability because the model needs one. A footnote looks like disclosure but still puts an invented number into the decision.
Today’s Assignment
Write Section 13 of your plan: the capital case for your decision on feasible options and one consistent lifecycle basis.
- Confirm the technical gates come first: which options are accepted for comparison, and which are excluded and why.
- Compare the feasible options on a common lifecycle boundary, stating the basis: service, period, currency, real or nominal, rate, timing, tax and end-of-life treatment.
- Show a downside and delay case and the assumption that could change the ranking; set the funding profile and affordability beside it.
- Name what AI may contribute and must not invent, and what would change the recommendation.
- Run the Appendix D checklist against your paper. Be honest about where your company compares ineligible options or treats present value as affordability. Label assumptions; no confidential figures needed.
YOUR 90-DAY PLAN
Section 13 — The capital case: feasible options on a common lifecycle basis, the downside, affordability and the committee checklist.
How this is marked — the rubric
| Technical gates confirmed first and a complete, consistent basis for feasible options on a common boundary | 30 |
| A downside and delay case and the assumption that could change the ranking | 25 |
| Affordability and funding profile separated from the economic ranking | 20 |
| AI limited to dossier and assumption work, change conditions stated and Appendix D run candidly | 25 |
| Total · pass mark 70 | 100 |
Submitting session 13
Write your answer in Word or Google Docs. The first line must read
Write your answer in Word or Google Docs. The first line must read
Session: 13, or the submission is refused. Upload the Word file (.docx), or switch to the Google Doc tab and paste your document’s share link.Marked against the rubric above and returned as a written letter, saved to your progress page. Do not submit until you have passed the self-check above.ASSESSMENT NOW OPEN
Pop Quiz 3 — Lifecycle, Governance and Capital opens now. Sit it before you start the next session — it is how you find out whether the doctrine landed, and the practice papers may be resat as often as you like.
Sit Pop Quiz 3 →Closed book, no notes, and Coach Ritchie switches himself off while you sit it.Capabilities Covered
Feasibility before financePresent-cost comparisonDownside testingAffordabilityCommittee scrutiny
Chapter 16 and Appendix B · Trust, Capital and PeopleSESSION 14
People, Contractors and the Right to Challenge
Why This Matters
Every section of your plan so far depends on people: the technician who reports a slip honestly, the planner who reads readiness, the contractor whose evidence you will rely on, the supervisor who says the access method will not work. Chapter 16 asks whether your organisation makes that work possible, and whether it is safe to report a real limitation when the forecast says otherwise. Appendix B then gives you board measures to watch it, four clocks that show where a decision actually waits, and a benefits ledger that keeps forecasts apart from cash. Today you write the people account for your decision and choose what your board will see. Practice Exam 3 opens at the end.
25-Minute Module
0:00 – 5:00
Design the work so people can succeed
- An asset receives care from people, not an organisation chart. HSE's maintenance-error guidance recognises that experienced, motivated technicians can still make slips and omissions; its shift-handover guidance emphasises communication when responsibilities pass between people or teams. These are international learning resources, not Nigerian law.
- Ask whether the approved procedure fits the available access, tools, staffing and time, and examine interruptions, competing tasks and information quality.
- THE DOCTRINE
A repeated error should prompt investigation of the work system rather than an automatic conclusion that more training is the only answer.
- AI can reduce searching, or flood specialists with alerts and drafts that take longer to verify than to write. Count verification time in the pilot; a system efficient on the executive screen may move work to the field. Let users flag output they distrust. Low adoption can reflect poor design, unclear authority or valid technical concern.
When the same mistake recurs, look at the job before you look at the person.
5:00 – 10:00
Competence and contractors, demonstrated
- Training attendance is not ability. By role: an executive challenges claims and understands authority; a planner interprets readiness and dependencies; an engineer evaluates the discipline's evidence and limitations; a data specialist needs access to domain expertise and understands the consequence of errors.
- Assess practically: give an incomplete record, a conflicting revision or a misleading model output and ask what decision is justified, and whether the person knows when to stop and seek help. Put knowledge transfer and handover evidence in the contract.
- NCDMB's February 2026 explanation of its NCEC guidance addresses demonstrable capability, documentary credibility and contracting delays; establish the applicable requirements with specialists. A prequalified contractor may still be unready: check task-specific competence, equipment, approved methods and mobilisation dependencies.
- Avoid incentives that reward speed while leaving quality and handover outside the payment account. Define accepted completion and its evidence.
Ask the contractor for the readiness record for this campaign, not the brochure.
10:00 – 14:00
Preserve the field warning; hand over what is unresolved
- AT THE ASSET IN THE FIELD · THE WORKED CASE
- In the field. The book's fictional supervisor reports that a proposed access method cannot reach an area under actual conditions, while the schedule shows work starting. The report preserves the observation, the consequence and the decision needed; an assistant may draft it, and the supervisor verifies the draft.
IN THE BOARDROOM · THE PARALLEL- In the boardroom. Leadership can ask for a truthful assessment and an alternative plan, or demand that the status remain green. The first improves the decision. The second teaches the team to conceal uncertainty until delay is unavoidable.
- THE DOCTRINE
No AI system can compensate for a reporting culture that systematically removes inconvenient information.
- A handover identifies what is incomplete, restricted or changed and who owes the next decision. Safety-critical information stays in the controlled operating and permit arrangements; an AI summary should not become a rival authoritative record.
- If one enthusiastic person is indispensable to every output, the capability is not yet dependable practice.
The warning you punish today is the one you will not hear next quarter.
14:00 – 20:00
Board measures with definitions
- Appendix B's measures are the author's proposed starting points. HSE's HSG254 is an established guidance resource for developing process-safety indicators in major-hazard settings; the table is not a reproduction of it.
Measure Definition and use Material decisions without a current disposition Material conditions lacking an accepted current position; review individually Critical work overdue Count and age against the approved due basis Evidence conflicts affecting decisions Open conflicts preventing a decision, with owner, age and route Restricted-operation conditions Current restrictions, validity, monitoring and approaching expiry Critical work readiness Prerequisites passed per critical package; failed gates shown separately Measure Definition and use Alert usefulness Confirmed relevant alerts divided by alerts reviewed, on agreed definitions Missed target events Defined events not flagged within the useful response window Decision-to-action time Accepted decision to executable action, with delay causes Accepted intervention completion Work accepted against criteria, not merely reported complete Benefit evidence Forecast, observed result and verified cash effect kept separate - Set asset-specific definitions first and no universal target: consequence, duty and approved obligations set the limits. For small populations, explain the cases rather than a percentage that swings with one event.
Choose the few measures that would have told you about your decision earlier.
20:00 – 25:00
Four clocks, a benefits ledger, Section 14 and Practice Exam 3
- Four clocks: when the condition was observed, assessed, decided by an authorised person and acted on. One closure time cannot show whether review, funding, procurement or execution caused the delay. Review reasons, not speed, and never reward closing a record before its evidence is accepted.
- Benefits ledger: baseline, comparator, method, period, owner, evidence and costs for each claim, with a state of forecast, observed operational change, financially verified or withdrawn. An avoided failure needs a stated counterfactual and uncertainty; a cash saving needs a real change in an outflow or receipt.
- The charge: make it possible to report a real limitation without penalty for spoiling the forecast. Assess competence through the work and require contractors to hand over usable evidence. Today's assignment produces Section 14.
- Practice Exam 3 opens now: fifteen questions across Sessions 1 to 14, closed book. Sit it before Session 15.
A benefit without an evidence state cannot be audited, only believed.
Key Terms
Demonstrated competence
Ability to perform a role's consequential task, shown by practical assessment, not attendance.
Task-specific readiness
Evidence a contractor has the competence, equipment, methods and mobilisation for this campaign, beyond prequalification.
Four clocks
When a condition was observed, assessed, decided and acted on, showing where a decision waits (Appendix B).
Benefit evidence state
Forecast, observed operational change, financially verified or withdrawn, kept separate from approval of the application.
Self-Check — Before You Submit
1. A fictional supervisor reports that the planned access method cannot reach the work area. The schedule shows work starting next week. What should leadership do?
- A. Keep the status green to avoid alarming the board
- B. Have an assistant rewrite the report more positively
- C. Ask for a truthful assessment and an alternative plan
- D. Replace the supervisor
C. Ask for a truthful assessment and an alternative plan — Demanding a green status teaches the team to conceal uncertainty. An assistant may help draft a report, but never by changing its meaning.
2. A pilot reduced report-preparation hours on a comparable sample, verification included. How does it enter the benefits ledger?
- A. As a financially verified saving
- B. As an observed operational change, not a cash saving unless an outflow or receipt really changes
- C. As an avoided failure
- D. In one combined savings total
B. As an observed operational change, not a cash saving unless an outflow or receipt really changes — Appendix B keeps the evidence states apart. Calling released hours cash is the tempting error; a cash saving needs a real change in an outflow or receipt.
3. A contractor is prequalified and presented its capability impressively. Your campaign starts next month. What does Chapter 16 say to check?
- A. Nothing more; prequalification settles it
- B. Only the price
- C. References from other owners
- D. Task-specific competence, equipment, approved methods and mobilisation dependencies
D. Task-specific competence, equipment, approved methods and mobilisation dependencies — A prequalified contractor may still be unready for a particular campaign, and a capability presentation is not a readiness record.
Today’s Assignment
Write Section 14 of your plan: the people account for your decision and the board measures your company will report.
- Whether the work is designed so people can succeed, and where AI would add verification work.
- The competence each role must demonstrate and how it is assessed through the work; the contractor's task-specific readiness, accepted completion and handover evidence.
- How a field warning reaches someone able to act without penalty, and what a handover must carry.
- The Appendix B measures you will report with asset-specific definitions; the four clocks for your decision; one benefits-ledger entry with baseline, comparator, owner and evidence state.
- Be candid about where your company today rewards a green status over a truthful one.
YOUR 90-DAY PLAN
Section 14 — People, contractors and the right to challenge, with the board measures we will report.
How this is marked — the rubric
| Diagnoses work design, role competence and contractor readiness with practical assessment and contract evidence | 30 |
| Shows how a field warning reaches an accountable person without penalty and what a handover carries | 25 |
| Selects Appendix B measures with asset-specific definitions and no universal targets, and sets the four clocks | 25 |
| One benefits-ledger entry with baseline, comparator and a correct evidence state | 20 |
| Total · pass mark 70 | 100 |
Submitting session 14
Write your answer in Word or Google Docs. The first line must read
Write your answer in Word or Google Docs. The first line must read
Session: 14, or the submission is refused. Upload the Word file (.docx), or switch to the Google Doc tab and paste your document’s share link.Marked against the rubric above and returned as a written letter, saved to your progress page. Do not submit until you have passed the self-check above.ASSESSMENT NOW OPEN
Practice Exam 3 — Full Syllabus opens now. Sit it before you start the next session — it is how you find out whether the doctrine landed, and the practice papers may be resat as often as you like.
Sit Practice Exam 3 →Closed book, no notes, and Coach Ritchie switches himself off while you sit it.Capabilities Covered
Work designDemonstrated competenceContractor readinessProtected escalationBoard measuresBenefits evidence
CapstoneSESSION 15
The Board Review: Your Signed 90-Day Intelligent Asset Plan
Why This Matters
Fourteen sessions ago you chose one unresolved asset decision: a condition, an anomaly or a lifecycle choice. Since then you have defined its function and ownership, recorded its evidence and uncertainty, placed it within criticality, inspection, failure modes, containment, access, repair and shutdown, carried it through life extension and retirement, governed the intelligence it relies on, priced it and put people behind it. Today you assemble those fourteen sections into one signed 90-Day Intelligent Asset Plan, with the Asset Decision Record at its heart, and set out the decision you will ask of your board or executive at day ninety. Then you sit the Final Exam.
25-Minute Module
0:00 – 5:00
Ninety days is a bounded proof
- Ninety days is a management period for a bounded proof, not a promise that every integrity or reliability application can be validated in one quarter. Rare failures, long inspection cycles and regulatory qualification can take longer.
- Give the pilot an operating owner and a decision that matters. Missing evidence in a critical work package, an unresolved anomaly queue or a reliability review burden may be a better start than a broad promise of predictive operations.
- Days 1 to 30: establish the decision and baseline. Write the charter with the users. Separate a data problem from a funding, competence or accountability problem. Measure the baseline from actual records or a time-bounded observation with stated limits; never build it from memory and call it measured. Choose the comparison method before results arrive.
- THE DOCTRINE
Agree the simpler alternative: better document control, a repaired sensor, a deterministic check or a more disciplined review. The objective is improved ownership, not compulsory use of AI.
A baseline you cannot show someone is not a baseline.
5:00 – 10:00
Build, rehearse, evaluate, decide
- Days 31 to 60: build and rehearse. Connect the minimum approved data and test against reviewed examples, including missing, stale or inconsistent evidence. Rehearse who receives the output, what they check and what happens when the reviewer is absent or the system fails. Internal pilot success is not external approval where qualification or regulatory process applies.
- Days 61 to 90: evaluate useful work. Use the application within approved limits once gates have passed. Record corrections and rejections; count review effort, data preparation and support. Compare with the baseline and the simpler method. If the pilot only establishes feasibility, say so.
- THE DOCTRINE
If no target failure occurs, do not claim the system prevented one. A reduction in preparation time is not automatically a headcount saving.
- The day-ninety decision review: continue within scope, extend evaluation, revise the design, expand under stated conditions or stop. A pilot that reveals unsuitability can protect the owner from a larger poor investment.
Stop is one of the five answers. Write it into the plan before you need it.
10:00 – 14:00
Capacity is not cash; keep the rhythm you already have
- The book's fictional capacity example: 80 evidence packs a month; an observed baseline of three staff-hours per pack falls to 1.5 in a comparable pilot, verification included. The difference is 240 − 120 = 120 hours a month. At an assumed USD 50 per hour, USD 6,000 of monthly capacity value; less USD 2,000 a month to operate, USD 4,000 of net monthly capacity value before other omitted costs.
- That is capacity, not cash profit. If payroll is unchanged, name the work the released hours will support. It also assumes equivalent pack quality and scope; the numbers are assumptions, not a target.
- Rhythm: the working level reviews output quality; the asset review connects outputs to dispositions and execution; the executive review examines exposure, funding, capability and authorised use. Use your existing meetings, not a parallel committee, with a route for urgent matters.
- Expand carefully: compare equipment, duty, data, staffing and workflow first. A document assistant may transfer more readily than a predictive model tied to one machine population.
Released hours are a promise until you can name the work they now do.
14:00 – 20:00
The Asset Decision Record at the heart of the plan
- Appendix A is a proposed management tool: one record per material condition or lifecycle choice, written in plain language first. The summary is an entry point to the evidence, not its replacement.
Field What belongs in the record Decision and boundary The question, equipment identity, system, service and period covered Required function What the asset must deliver and the conditions relevant to that duty Current evidence Source identifiers, dates, methods, revisions and accepted limitations Material uncertainty What is unknown and which choice it could change Current disposition The authorised position, conditions, validity and review trigger Eligible options Options accepted for comparison and their technical basis Field What belongs in the record Recommendation The preferred next commitment and the reasons for it Authority Technical acceptance, operating decision and spending approval as applicable Execution Scope owner, prerequisites, resources, window and acceptance evidence Reassessment Expiry, operating change, new finding or other trigger requiring review Closure Evidence of accepted completion and updates returned to controlled records - AT THE ASSET IN THE FIELD · THE WORKED CASE
- In the field. The book's fictional transfer-system anomaly: identity and location verified, but two reports describe the location differently and their measurements' comparability is unresolved. The integrity manager coordinates evidence and access; the technical authority determines the assessment.
IN THE BOARDROOM · THE PARALLEL- In the boardroom. The commitment requested is funding for a competent review and, if required, targeted examination, approved by finance within delegation. The existing disposition stays under its original authority. An email saying reports were sent does not close it.
- THE DOCTRINE
If specialists disagree, state the competing interpretations and the evidence that could resolve them. Do not average the opinions into an apparently agreed condition rating.
Record when the decision changes and why, and keep the previous basis.
20:00 – 25:00
Assemble, sign and set the day-ninety decision
- The Conclusion: an asset owner cannot transfer responsibility to an inspection report, a contractor or a model. That should make you a demanding user of AI, not a reluctant one: ask it to find missing evidence, expose a contradiction or shorten the route to an executable action, and test those claims. Value the engineer who explains uncertainty, the supervisor whose warning reaches someone able to act, the finance team that separates preference from affordability, and the board that accepts a justified restriction or retirement.
- THE DOCTRINE
The intelligent asset is not equipment that has acquired judgement. It is an asset whose owners exercise better judgement, supported by information they can trust and people they equip to act.
- Write the board paper in Appendix E's vocabulary, used precisely: disposition, technical authority, readiness, operating envelope, residual value and the rest. Fitness for service is an engineering assessment; a model prediction is not one.
- Assemble Sections 1 to 14 under a one-page summary. The charge: approve one bounded proof with an accountable owner and a day-ninety decision. Judge it by the work it improves and the evidence it produces.
- Sign and date the plan. Then sit the Final Exam: fifteen questions, three rebuilt from your own earlier mistakes, closed book, pass mark 70.
Sign only what you would be prepared to review honestly on day ninety. The decision still belongs to you.
Key Terms
90-Day Intelligent Asset Plan
Sections 1 to 14 assembled around one asset decision under a one-page summary, signed and dated.
Asset Decision Record
The author's proposed single record for a material condition or lifecycle choice (Appendix A).
Day-ninety decision review
Continue within scope, extend evaluation, revise the design, expand under stated conditions or stop.
Capacity value
Staff time released and valued on a stated basis; not cash profit unless an outflow or receipt changes.
Self-Check — Before You Submit
1. In the book's fictional capacity example, what is the USD 4,000 a month?
- A. Verified cash profit
- B. A headcount saving
- C. Net monthly capacity value before other omitted costs: USD 6,000 less USD 2,000 to operate
- D. A target every owner should meet
C. Net monthly capacity value before other omitted costs: USD 6,000 less USD 2,000 to operate — The book says do not label it cash profit: with unchanged payroll the business has released capacity. Its numbers are assumptions, not a target.
2. At day ninety no target failure occurred on the monitored equipment. The sponsor wants the paper to say the application prevented failures. What do you write?
- A. What was observed against the baseline and simpler method, without claiming a prevented failure
- B. That it prevented failures, since none occurred
- C. That it prevented failures, with an estimated value
- D. Nothing until another quarter has passed
A. What was observed against the baseline and simpler method, without claiming a prevented failure — Chapter 17 says if no target failure occurs, do not claim the system prevented one. Silence is also wrong: report what the pilot did establish.
3. Two specialists disagree: one reads an anomaly as active deterioration, the other as historic. A manager proposes a middle rating so the record looks agreed. What does Appendix A require?
- A. Record the middle rating
- B. Record the senior specialist's view
- C. Leave the field blank
- D. Record both interpretations, the evidence that could resolve them and the decision authority
D. Record both interpretations, the evidence that could resolve them and the decision authority — Appendix A forbids averaging opinions into an agreed rating. Deferring to seniority looks decisive but hides the disagreement the authority needs to see.
Today’s Assignment
Assemble Sections 1 to 14 into one signed 90-Day Intelligent Asset Plan for your company, with a one-page summary your board or executive can decide on.
- Assemble the fourteen sections in order, noting where a later session changed an earlier one.
- Complete the Asset Decision Record in every Appendix A field, recording disagreement rather than averaging it; name the operating owner and the technical, operating and spending authorities.
- Set out days 1 to 30, 31 to 60 and 61 to 90: baseline, simpler alternative, evidence per period and gates that apply.
- Describe the operating rhythm and the board measures from Section 14.
- State the day-ninety decision you will ask for, with permitted expenditure, conditions, owner and next review. If the honest answer is a simpler method, a funded repair or no model, say so.
- Sign and date it, and name the commitment you are least confident your company will keep.
YOUR 90-DAY PLAN
The signed 90-Day Intelligent Asset Plan.
How this is marked — the rubric
| Sections assembled, consistent, signed and dated under a one-page summary naming owner and authorities | 20 |
| Asset Decision Record complete in every field, disagreement recorded not averaged | 25 |
| Three periods with measured baseline, simpler alternative, evidence and gates | 25 |
| Operating rhythm and Section 14 measures in existing reviews | 10 |
| Day-ninety decision with permitted expenditure, conditions, owner and next review | 20 |
| Total · pass mark 70 | 100 |
Submitting session 15
Write your answer in Word or Google Docs. The first line must read
Write your answer in Word or Google Docs. The first line must read
Session: 15, or the submission is refused. Upload the Word file (.docx), or switch to the Google Doc tab and paste your document’s share link.Marked against the rubric above and returned as a written letter, saved to your progress page. Do not submit until you have passed the self-check above.FINAL EXAM
Your Final Exam is sat in this session. 15 questions, of which 3 are rebuilt from questions you got wrong on the practice papers. Sit it once your capstone is submitted.
Sit the Final Exam →Closed book, no notes, and Coach Ritchie switches himself off while you sit it.Capabilities Covered
Plan assemblyBounded proof designBaseline disciplineAsset Decision RecordBoard decision