Turbine 7 — Kit 9

Theme: Offshore Amber (dark, industrial) · Audience: offshore wind control-room techs · Artifacts: simulator.html (2D) · 3d.html + turbine.js (3D nacelle) · shared case file scenarios.js

One-line: a restart/hold decision simulator where every safe answer has a price tag — scored on the decision PATTERN, not the outcome.

Why this kit exists

In this fictional brief, incident reviews tell the same story: cleared alarm + production pressure + restart = gearbox cascade. Training teaches thresholds; it doesn't practice resisting the economic argument under genuine ambiguity.

What the artifact does

A night shift at an offshore wind control desk. Each shift deals four alarm cases, one per decision pattern, from a bank of eight, in a random order. The learner checks the evidence (in 3D: on the turbine itself, across the string and on the met mast), chooses Restart, Restart derated, Hold for inspection or Escalate, names the pattern that decided it, and answers the production desk on the hold cases. The call stays locked until the part holding the deciding evidence has been opened. The shift ends with a handover note checked against a three-part rubric. Pass = 12/14 and no critical error (restarting into a rising trend or a recurring fault).

The 3D scene is the control room plus turbine T-07 at real scale among its string, with the weather, sea and light of each case; the consequence of the call plays out on the machine and the vessels. The 2D simulator runs the same cases, scoring and note.

The design decisions worth noticing

• The tempting wrong answer — every scenario's unsafe option is economically rational-looking; the temptation IS the lesson
• Pattern-based scoring — outcomes are noisy; rising-trend vs cleared-alarm is the transferable distinction
• The production-desk voice — pressure is a character in the scenarios, because it's a character on the real desk
• Evidence as environmental traces — people stay off-screen; the inspection card, the work order and the pitch mark they left are what the call rests on

Honest limitations

Simplified physics — real decision support belongs to OEM diagnostics. The trainer builds the judgment layer above them. Failure economics (€/day, vessel lead times) are illustrative composites, not field data. No wind O&M expert has reviewed the cases yet (the checklist is in the kit). No learners have used the build, so there are no results to report.

Part of I Wayan Usada's learning-design portfolio · all clients and scenarios are fictional composites built from realistic domain patterns · drafted with AI assistance and fully rewritten, fact-checked and owned by a human.

Process deliverables

• action-map.md — Action map — the behaviours the artifact must produce, traced to the business goal
• storyboard.md — Storyboard — 16 frames rendered from the build, the case bank and the scoring
• visual-mockups.md — Visual mock-ups & art direction — annotated renders and component decisions
• job-aid.md — Alarm-pattern card — one printable page for the control-room wall
• huddle-guide.md — Shift-huddle facilitator guide — a 30-minute monthly huddle for shift leads
• handover-rubric.md — Handover-note rubric (LO3) — criteria, the auto-check and its limits, worked examples
• scenario-authoring.md — Scenario authoring notes — how to add or change a case
• evaluation-plan.md — Evaluation plan (proposed) — pilot questions, measures and decision rules
• expert-review-checklist.md — Expert-review checklist — what a wind O&M SME should check before a pilot
• simulator.html — 2D simulator
• 3d.html — 3D control room and turbine, same cases and scoring as the simulator