Repository Docs & Knowledge Base

id: 019cb9fd-149c-7b4b-8d15-2e83794c8e47 name: Forgotten Items Monitor kind: Anomaly cron: 0 0 * * * * analysis_window: 1h comparison_periods: [1d, 7d]

Detect items that are stuck on the KDS screen and never progressing through the kitchen workflow -- they were created but never prepared or completed.

Known Tables & Columns

curated.kitchen_orders

• guid (varchar, PK) - order identifier
• store_guid (varchar) - which store/restaurant
• items_count (int) - number of items in the order
• done (tinyint) - whether the order is complete
• is_deleted (tinyint) - soft delete flag
• create_time (int, unix timestamp) - when the order was created

curated.kitchen_order_items

• guid (varchar, PK) - item identifier
• order_guid (varchar, FK) - parent order
• name (varchar) - menu item name
• category (varchar) - Grill, Fryer, Salads, Drinks, Desserts
• device_id (int) - KDS station ID (1=Grill, 2=Fryer, 3=Salad, 4=Drinks, 5=Desserts)
• preparation_time (int, seconds, nullable) - expected prep time
• is_deleted (tinyint) - soft delete flag
• create_time (int, unix timestamp) - when the item was created

curated.kitchen_item_status_events

• guid (varchar, PK) - bump identifier
• item_guid (varchar, FK) - parent item
• status (tinyint) - 0=pending, 1=prepared, 2=done
• prepared_local_time (int, unix timestamp, nullable) - when item was prepared (0 if never prepared)
• done_local_time (int, unix timestamp, nullable) - when item was completed (0 if never completed)
• is_deleted (tinyint) - soft delete flag
• create_time (int, unix timestamp) - when the bump record was created

Analysis Approach

1. Query items created in the last {AnalysisWindow} from curated.kitchen_order_items (is_deleted=0), joined with curated.kitchen_item_status_events on item_bumps.item_guid = items.guid.
2. Verify that the analysis window contains at least 20 items total. If not, call emit_no_anomaly — the sample is too small.
3. Identify "stuck" items: those where item_bumps.status = 0 (still pending) AND the item was created more than 2x its expected preparation_time ago (or more than 15 minutes ago as a fallback).
4. Calculate the forgotten rate: the percentage of items in a time window that remain at status 0 long past their expected completion.
5. Compare against the historical baseline. Normally, almost all items should progress to status 1 (prepared) and then status 2 (done). A forgotten rate above 15% is unusual; above 30% is a clear anomaly.
6. Group by hour to find the window where the anomaly started.
7. Check if the stuck items cluster on a specific station (device_id) or are spread across all stations. A cluster on one station might indicate a station-specific issue; spread across all stations suggests a systemic problem.

Severity Guidance

• Low: 15-25% of items stuck beyond their expected prep time in a 1-hour window
• Medium: 25-40% of items stuck, or stuck items accumulating over multiple hours
• High: >40% of items stuck -- clear indication of items being forgotten and not prepared