The Doctrine Card

Everything this programme rests on, on one page. Print it and keep it in the folder you take to the operating review. These are the author’s management tools, not official standards: apply them to a real decision and judge them by what the team can explain, execute and verify.

The discipline

Define the decision. Establish dependable evidence. Assign authority. Measure the outcome. A refinery earns its result through connected decisions — feedstock, campaign, equipment, release, cash — and AI is useful only when it improves one of them and the improvement can be traced through the operating record.

The six frameworks

#	Framework	The executive test
1	The Operating-Intelligence Loop sense → decide → act → learn	Where does the loop break — late evidence, an unclear recommendation, an owner without authority, or no money? Each needs a different remedy.
2	The Data-Foundation Precondition	Does the evidence for this decision have an identity, unit, time basis, quality status and owner? When should the application withhold a recommendation?
3	The Autonomy Ladder inform → recommend → narrowly approved execution	What can it read, propose and write; who approves; what happens when it is wrong; how does a supervisor pause it? Accuracy is not permission.
4	The Value-versus-Risk Map	What is the consequence of error? A large expected benefit does not make an unacceptable safety exposure acceptable.
5	The Sequencing Discipline	Decision and data foundation, bounded proof, verified response and economics — then decide whether to expand. A second unit is a new assessment.
6	The Operations-Excellence Lens	Deliverable production, quality, availability, energy, working capital, verified compliance and collected contribution. Count a benefit once.

One decision record

Field	What it holds
Decision requested	The choice, in one sentence, with its deadline
Feasible alternatives	Including doing nothing, a temporary arrangement and a permanent change
Evidence and uncertainty	What is known, what is assumed, and the condition that would change the choice
Accountable owner	The person answerable for the business result
Operating and cash consequences	Physical effect, contribution and the payment dates
Next review	When the decision is revisited, and earlier versions preserved

Four accountable roles: executive sponsor · process owner · evidence owner · technical custodian. Know which method you are using: a material balance, linear programming, advanced process control or a safety instrumented system is not automatically AI.

Modular and integrated

Modular refinery	Integrated (mega) refinery
Concentrated constraints — a feedstock interruption, a product tank or a critical pump can affect much of the plan. Start with a campaign record linking crude receipt, laboratory evidence, production, tank movements and collection. Ask who reviews recommendations on the night shift.	More interfaces and alternative routes — a change at one unit can affect hydrogen, fuel gas, intermediates and the product pool. Make shared constraints explicit; keep local authority and escalate only where a choice changes another unit’s commitment or a financial exposure.

A common doctrine with different implementation. Nameplate capacity is not usable capacity for the decision period.

The first ninety days

Period	What happens
Days 1–15	Sponsor and business owner chosen; three recurring decisions compared on the Value-versus-Risk Map; baseline from actual records; one use case chosen at the first gate, or a conventional fix first
Days 16–30	Data contract and decision register; present rung on the Autonomy Ladder; access reviewed; investment boundary and benefit ledger; pilot charter with acceptance and stopping conditions before any supplier proposal is chosen
Days 31–60	The approved trial within scope; reasons for rejection preserved; weekly review of where the response chain is delayed; baseline maintained
Days 61–90	The trial closed against its charter; finance verifies the consequence; the sponsor decides whether the next commitment is justified — scale, revise or stop, with the unfavourable result kept on the record

The periods are indicative, not promised implementation times. The executive review pack keeps three accounts: operating, financial and programme.

The 20 sessions and the plan they build

#	Session	Plan section
1	Why AI Belongs in the Refinery's Decisions	Section 1 — Where we are today: the facility, the live refinery problem and the decision it needs.
2	The Reader's Toolkit for the Refinery	Section 2 — The decision record for our live problem, and its weakest link.
3	Leading the Intelligent Refinery	Section 3 — The operating objective, its boundary and the four accountable roles.
4	Modular and Integrated Operating Models	Section 4 — Our configuration sheet: fixed design facts versus what changes in the planning period.
5	Turning Operating Constraints into Decisions	Section 5 — The binding constraint, its register entry and two executable responses.
6	Margin, Yield and the Executive Operating Review	Section 6 — Our campaign bridge: contribution boundary, plan-to-actual and the cash dates.
7	Feedstock Quality and Crude Selection	Section 7 — Our crude-campaign comparison on a consistent delivered basis.
8	Blending and Feedstock Scheduling	Section 8 — One week's feedstock schedule traced to its critical dependency.
9	Planning Across Process Units	Section 9 — Our planning method, reference plan and the shared resource that decides it.
10	Yield and Product Quality	Section 10 — The quality chain for one product and where prediction may advise but not release.
11	Asset Health, Maintenance and Integrity Evidence	Section 11 — One asset family: the failure concern, the response chain and the integrity evidence.
12	Turnaround Planning and Execution	Section 12 — Our turnaround or major-work readiness test.
13	Process Safety and Management of Change	Section 13 — The management-of-change boundary for our AI application.
14	Utilities, Energy, Hydrogen and Fuel Gas	Section 14 — Our shared-utility constraint and how its value would be verified.
15	Laboratory Release, Inventory and Custody Transfer	Section 15 — The release-to-cash chain for one product, and where it loses time or quantity.
16	Critical Spares, Operating Data and Reconciliation	Section 16 — One critical spare decision and the data contract behind our first use case.
17	Cybersecurity, Controlled Access and Workforce Adoption	Section 17 — The access boundary, recovery route and role competence for our application.
18	Investment Cases and Value Attribution	Section 18 — The honest investment case and benefit ledger for our first application.
19	AI Governance, Pilots and Controlled Scaling	Section 19 — The use-case register entry and pilot charter, with acceptance agreed before results.
20	The First Ninety Days	Section 20 — The assembled, signed 90-Day Intelligent Refinery Plan with its executive summary.

ERW CommandRefinery · built on The Intelligent Refinery — An Operating Doctrine for AI in Nigerian Refinery Operations by Ritchie Wingo. Leadership training, not engineering, safety, legal or investment advice. Back to the programme home →

THE INTELLIGENT REFINERY LEADERSHIP PROGRAMME

AI earns its place in a refinery one decision at a time.