Repository Docs & Knowledge Base

id: 019cb9fd-1498-7b73-8fc0-6861ae86b223 name: Slow Kitchen Monitor kind: Anomaly cron: 0 0 * * * * analysis_window: 1h comparison_periods: [1d, 7d]

Detect when the entire kitchen is running slow -- all preparation times are significantly elevated across every station.

Known Tables & Columns

curated.kitchen_orders

• guid (varchar, PK) - order identifier
• store_guid (varchar) - which store/restaurant
• preparation_time (int, seconds, nullable) - total order preparation time
• items_count (int) - number of items in the order
• order_type (varchar) - DINE_IN, TO_GO, DELIVERY
• destination (varchar) - Kitchen, Bar, Window, Expo
• done (tinyint) - whether the order is complete
• is_deleted (tinyint) - soft delete flag
• create_time (int, unix timestamp) - when the order was created
• update_time (int, unix timestamp) - last update

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) - how long the item took to prepare
• 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
• done_local_time (int, unix timestamp, nullable) - when item was completed
• is_deleted (tinyint) - soft delete flag

Analysis Approach

1. Query the last {AnalysisWindow} of completed orders (done=1, is_deleted=0, preparation_time IS NOT NULL).
2. Join with curated.kitchen_order_items and aggregate average item-level preparation_time by hour and by station.
3. Verify that each station has at least 15 items in the analysis window. Ignore stations with fewer data points.
4. Compare the per-hour averages against the historical baseline (same hour-of-day {ComparisonPeriods}).
5. A global slowdown affects ALL stations simultaneously -- check whether the elevated prep times span Grill, Fryer, Salad, Drinks, and Desserts uniformly rather than being isolated to one station.
6. Flag hours where the average prep time across all stations is >3 standard deviations above baseline AND the absolute elevation is at least 90 seconds above the baseline average.
7. To confirm this is a kitchen-wide issue (not a single-station bottleneck), verify that at least 4 out of 5 stations show elevated prep times in the same time window. If only 1-2 stations are elevated, call emit_no_anomaly (it may be a station bottleneck, not a slow kitchen).

Severity Guidance

• Low: 3-4 sigma deviation, at least 4/5 stations elevated by 90s+
• Medium: 4-5 sigma deviation, or prep times 3x above baseline across 4+ stations
• High: >5 sigma deviation, or prep times 5x+ above baseline across all stations