Asem Alkamel

import pandas as pd
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split
from sklearn.metrics import f1_score

df = pd.read_csv('production_lines.csv')
X = df.drop(columns=['failure'])
y = df['failure']

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)
model = RandomForestClassifier(n_estimators=400, max_depth=12)
model.fit(X_train, y_train)

preds = model.predict(X_test)
print(f'F1: {f1_score(y_test, preds):.3f}')

-- SAP B1: daily output by line vs target
SELECT
  T0.U_LineCode      AS Line,
  CONVERT(date, T0.U_PostDate) AS Day,
  SUM(T0.U_Qty)      AS Produced,
  SUM(T0.U_Target)   AS Target,
  CAST(100.0 * SUM(T0.U_Qty) / NULLIF(SUM(T0.U_Target), 0) AS DECIMAL(5,2)) AS PctOfTarget
FROM [@MES_OUTPUT] T0
WHERE T0.U_PostDate >= DATEADD(day, -30, GETDATE())
GROUP BY T0.U_LineCode, CONVERT(date, T0.U_PostDate)
ORDER BY Day DESC, Line;

// Service Layer push: post production receipt
const payload = {
  ProductionOrderId: order.id,
  CompletedQuantity: line.produced,
  PostingDate: new Date().toISOString(),
  WarehouseCode: 'FG',
};

const res = await fetch(`${SL}/InventoryGenEntries`, {
  method: 'POST',
  headers: { 'Content-Type': 'application/json', Cookie: session },
  body: JSON.stringify(payload),
});
if (!res.ok) await reconcile(order, await res.json());

# IoT telemetry — anomaly detection on conveyor vibration
from river import anomaly

detector = anomaly.HalfSpaceTrees(n_trees=25, height=8, window_size=250)
for reading in stream:
    score = detector.score_one(reading)
    detector.learn_one(reading)
    if score > 0.8:
        emit_alert(line=reading['line'], severity='high', score=score)