Upload 3 files

data_loader.py

@@ -0,0 +1,14 @@
+import pandas as pd
+
+def load_data():
+    df_m = pd.read_csv("data/Hauler_Machine_IPR.csv")
+    df_p = pd.read_csv("data/Hauler_People_IPR.csv")
+    df_w = pd.read_csv("data/Hauler_Weather_IPR.csv")
+
+    # Gabungkan berdasarkan tanggal
+    df = df_m.merge(df_p, on="date", how="inner")
+    df = df.merge(df_w, on="date", how="inner")
+
+    # Pastikan kolom sesuai
+    df["cuaca_bin"] = df["cuaca"].map({"cerah": 0, "hujan": 1})
+    return df

tools.py

@@ -0,0 +1,35 @@
+import joblib
+import pandas as pd
+
+model = joblib.load("model_pty.pkl")
+
+def predict_pty(pa, wicope, backlog, cuaca, resign, bad_ethic, incompetent):
+    data = pd.DataFrame([{
+        "PA": pa,
+        "WICOPE": wicope,
+        "backlog": backlog,
+        "cuaca_bin": 1 if cuaca == "hujan" else 0,
+        "resign": resign,
+        "bad_ethic": bad_ethic,
+        "incompetent": incompetent
+    }])
+    result = model.predict(data)[0]
+    return int(result)
+
+def analyze_causes(pa, wicope, backlog, cuaca, resign, bad_ethic, incompetent):
+    alasan = []
+    if pa < 90: alasan.append("PA rendah (<90%)")
+    if wicope < 90: alasan.append("WICOPE <90%")
+    if backlog > 30: alasan.append("Backlog terlalu banyak")
+    if cuaca == "hujan": alasan.append("Cuaca buruk")
+    if resign > 1: alasan.append("Tingkat resign tinggi")
+    if bad_ethic > 0: alasan.append("Ada perilaku buruk")
+    if incompetent > 0: alasan.append("SDM tidak kompeten")
+    return alasan
+
+def give_recommendation(backlog, pa, resign):
+    rekom = []
+    if backlog > 30: rekom.append("Percepat eksekusi backlog")
+    if pa < 90: rekom.append("Tingkatkan ketersediaan alat")
+    if resign > 1: rekom.append("Cegah resign dengan retensi")
+    return rekom

train_model.py

@@ -0,0 +1,18 @@
+import pandas as pd
+from sklearn.ensemble import RandomForestRegressor
+import joblib
+from data_loader import load_data
+
+df = load_data()
+
+features = ['PA', 'WICOPE', 'backlog', 'cuaca_bin', 'resign', 'bad_ethic', 'incompetent']
+target = 'PTY'
+
+X = df[features]
+y = df[target]
+
+model = RandomForestRegressor()
+model.fit(X, y)
+
+joblib.dump(model, "model_pty.pkl")
+print("✅ Model trained and saved as model_pty.pkl")