KPI Alarm System
Preset KPI limits per department with live severity, plus AI-recommended KPIs you can activate. Breaches raise warning or critical alarms, can be acknowledged with an audit entry, and feed the AI agents as operational context.
Active alarms
- OTIF (on time in full)CriticalExecutive · actual 91.4 % vs limit 92 % (min) · Board target 95%
- Gross marginPendingExecutive · actual 38.6 % vs limit 36 % (min) · Budget 39%
- Open critical risksCriticalExecutive · actual 17 vs limit 8 (max) · Enterprise risk policy
- Forecast accuracy (MAPE)PendingPlanning · actual 17.8 % vs limit 22 % (max) · S&OP policy ≤15%
- Days of inventory coverPendingPlanning · actual 118 days vs limit 130 days (max) · Working capital target 90–100 d
- POs at high delay riskCriticalProcurement · actual 8 vs limit 6 (max) · Supply continuity policy
- Qualified supplier coverageCriticalProcurement · actual 93.2 % vs limit 95 % (min) · EU GMP Ch.5 supplier control
- OEEPendingManufacturing · actual 74.6 % vs limit 70 % (min) · Site target 78%
- Batch yield variancePendingManufacturing · actual 2.1 % vs limit 3 % (max) · Master formula tolerance ±2%
- Right first time (EBR)PendingManufacturing · actual 88.5 % vs limit 85 % (min) · Data integrity KPI
- Batches awaiting QA releaseCriticalQuality · actual 13 vs limit 9 (max) · Release SLA 3 working days
- Open critical deviationsCriticalQuality · actual 9 vs limit 3 (max) · EU GMP Ch.1 quality system
- Overdue CAPAsCriticalQuality · actual 5 vs limit 5 (max) · CAPA SOP QA-014
- OOS ratePendingQuality · actual 1.4 % vs limit 2 % (max) · QC trending SOP
- Near-expiry stock exposureCriticalWarehouse · actual 71.6 EGP M vs limit 40 EGP M (max) · Write-off provision policy
- FEFO compliancePendingWarehouse · actual 96.1 % vs limit 95 % (min) · GDP Annex FEFO rule
- Cycle count accuracyPendingWarehouse · actual 98.7 % vs limit 97 % (min) · GxP inventory SOP
- Cold-chain excursionsPendingDistribution · actual 3 vs limit 5 (max) · GDP temperature control
- Average delivery lead timePendingDistribution · actual 41 hours vs limit 48 hours (max) · Service policy 36 h
- Orders on credit holdPendingCommercial · actual 5 vs limit 8 (max) · Credit policy
- DSOPendingFinance · actual 71 days vs limit 75 days (max) · Treasury target 60 d
- Cost variance (unfavourable)PendingFinance · actual 4.9 EGP M vs limit 6 EGP M (max) · Controlling tolerance
- Overdue receivables > 30 dPendingFinance · actual 22.8 % vs limit 25 % (max) · Credit policy
- Instruments overdue calibrationPendingEngineering · actual 2 vs limit 3 (max) · Calibration SOP ENG-007
- Critical asset uptimePendingEngineering · actual 96.4 % vs limit 94 % (min) · Maintenance plan
- Licences expiring < 90 daysPendingRegulatory · actual 4 vs limit 6 (max) · Regulatory calendar
- GxP training complianceCriticalRegulatory · actual 94.3 % vs limit 95 % (min) · Training SOP HR-002
- Interface failure ratePendingIT & Automation · actual 0.8 % vs limit 2 % (max) · Integration SLA
- Audit trail completenessPendingIT & Automation · actual 99.6 % vs limit 99 % (min) · 21 CFR Part 11
- Master data completenessCriticalMaster Data · actual 92.8 % vs limit 95 % (min) · Data governance policy
- Duplicate master recordsPendingMaster Data · actual 7 vs limit 15 (max) · Data governance policy
Preset KPI thresholds
| KPI | Department | Actual | Warning | Critical | State | Alarm |
|---|---|---|---|---|---|---|
OTIF (on time in full) Customer orders delivered complete and on date. | Executive | 91.4 % | Critical | |||
Gross margin Consolidated gross margin, month to date. | Executive | 38.6 % | Pending | |||
Open critical risks Critical deviations, recalls and supply risks open now. | Executive | 17 | Critical | |||
Forecast accuracy (MAPE) Mean absolute % error of the demand forecast. | Planning | 17.8 % | Pending | |||
Days of inventory cover Finished goods cover against the rolling plan. | Planning | 118 days | Pending | |||
POs at high delay risk Open purchase orders flagged high delay risk by the AI agent. | Procurement | 8 | Critical | |||
Qualified supplier coverage Spend covered by suppliers with valid GxP qualification. | Procurement | 93.2 % | Critical | |||
OEE Overall equipment effectiveness across packaging lines. | Manufacturing | 74.6 % | Pending | |||
Batch yield variance Deviation of actual yield from the master formula yield. | Manufacturing | 2.1 % | Pending | |||
Right first time (EBR) Batch records closed with no correction. | Manufacturing | 88.5 % | Pending | |||
Batches awaiting QA release Batches sitting in QA review beyond the release SLA. | Quality | 13 | Critical | |||
Open critical deviations Critical classified deviations not yet closed. | Quality | 9 | Critical | |||
Overdue CAPAs CAPA actions past their committed due date. | Quality | 5 | Critical | |||
OOS rate Out-of-specification results as % of tests performed. | Quality | 1.4 % | Pending | |||
Near-expiry stock exposure Value of lots expiring within 90 days. | Warehouse | 71.6 EGP M | Critical | |||
FEFO compliance Picks executed against the first-expiry lot. | Warehouse | 96.1 % | Pending | |||
Cycle count accuracy Bin-level inventory accuracy from cycle counting. | Warehouse | 98.7 % | Pending | |||
Cold-chain excursions Shipments with a temperature excursion in the last 30 days. | Distribution | 3 | Pending | |||
Average delivery lead time Order release to customer delivery. | Distribution | 41 hours | Pending | |||
Orders on credit hold Sales orders blocked on credit exposure. | Commercial | 5 | Pending | |||
Sales target attainment Net sales versus target, month to date. | Commercial | 104.9 % | Passed | |||
DSO Days sales outstanding on trade receivables. | Finance | 71 days | Pending | |||
Cost variance (unfavourable) Actual versus standard batch cost, month to date. | Finance | 4.9 EGP M | Pending | |||
Overdue receivables > 30 d Share of AR older than 30 days. | Finance | 22.8 % | Pending | |||
Instruments overdue calibration GxP instruments past their calibration due date. | Engineering | 2 | Pending | |||
Critical asset uptime Availability of qualified critical equipment. | Engineering | 96.4 % | Pending | |||
Licences expiring < 90 days Marketing authorisations and site licences near expiry. | Regulatory | 4 | Pending | |||
GxP training compliance Staff current on their assigned GxP curriculum. | Regulatory | 94.3 % | Critical | |||
Interface failure rate Failed messages on the MES / LIMS / SAP bus. | IT & Automation | 0.8 % | Pending | |||
Audit trail completeness Transactions carrying a complete ALCOA+ audit entry. | IT & Automation | 99.6 % | Pending | |||
Master data completeness Products, materials and partners with all mandatory GxP attributes. | Master Data | 92.8 % | Critical | |||
Duplicate master records Suspected duplicates detected by the data quality agent. | Master Data | 7 | Pending |
Thresholds are per-department presets derived from GxP policies and site targets; edits persist locally and are written to the audit trail on acknowledgement. Direction "max" alarms when the value rises above the limit, "min" when it drops below.
AI-recommended KPIs
DIO + DSO − DPO across the group.
Why: Inventory cover and DSO are both above target — a combined cycle KPI exposes the working-capital trend earlier.
Signed forecast error showing systematic over/under planning.
Why: MAPE alone hides direction; bias explains the rising inventory cover.
Spend on APIs with only one qualified supplier.
Why: Imported API delays are the top supply risk driver this quarter.
Average validated cleaning + changeover duration per line.
Why: Largest single contributor to the OEE gap on Line 3.
Batch completion to release decision.
Why: Counting the queue is lagging; cycle time detects the slowdown before the queue builds.
Deviations recurring on the same root cause within 12 months.
Why: Directly measures CAPA effectiveness — a common inspection finding.
Value written off for expiry versus stock value.
Why: Converts the near-expiry exposure alarm into realised financial loss tracking.
MKT of monitored lanes over 30 days.
Why: Excursion counts miss slow drift; MKT catches lane degradation.
Revenue at risk from accounts flagged by the churn agent.
Why: Turns the AI churn signals into a monitored, escalating alarm.
Actual versus planned landed cost on imported API.
Why: FX and freight drive most of the current unfavourable cost variance.
Mean time between failures for qualified critical equipment.
Why: Predictive maintenance alerts need a reliability baseline to be actionable.
Days to submit a response to an inspection observation.
Why: Late responses escalate observations; time-based alarming prevents it.
Age of the newest successfully synced SAP record.
Why: Stale master data silently corrupts planning and costing runs.
Request to approved change on GxP master records.
Why: Slow master data changes are the root cause of most planning exceptions.