Depot Morning Huddle — AI-Assisted Production Brief

Pick your tool based on data sensitivity

These files are NOTIONAL — no real CUI, PII, or operational data. That means any approved LLM (ChatGPT, Claude, Gemini) is fine for this exercise. If you were running this with real maintenance records, NALCOMIS exports, or BuNos: GenAI.mil (IL5) only, or depersonalize first.

Download the three CSV files

Click each file card below to download Depot_WIP_NOTIONAL.csv, Depot_Supply_AWP_NOTIONAL.csv, and Depot_Schedule_Throughput_NOTIONAL.csv. Keep them in a folder together — you will upload all three at once.

Upload all three files in a single session

Open GenAI.mil (for CUI/real data) or ChatGPT/Claude (for this notional exercise). Attach all three CSVs before sending the first message. The AI needs all three loaded to cross-reference across the datasets in later steps.

Run Step 1 — Orientation

"Summarize this WIP file. How many jobs, by status and by shop?" This proves the AI read the data correctly and gives you a baseline count before asking harder questions.

Run Steps 2–4 — Escalating complexity

Step 2: List every job past its ECD that is not yet RFI, sorted by days past ECD. Step 3: Cross-reference the AWP jobs against the supply file — which are stuck on a single backordered, sole-source OEM part? Step 4: Rank the EXREP jobs by likelihood of making ECD and identify which two need command attention.

Run Steps 5–6 — The brief writes itself

Step 5: "Draft a 5-bullet production status brief for the 0700 stand-up: throughput vs plan, top 3 schedule risks, AWP impact, and one recommended action." Step 6: "Based on actual vs target TAT trends by product line, which line is drifting and what is the likely impact on next quarter's output?" Point out to the room: that brief took 8 seconds and zero PowerPoint.