Update app.py

app.py

@@ -23,18 +23,20 @@ IPLM 2025 — FINAL (NO UPLOAD) — FULL REWRITE (NO RINGKAS)
 ✅ Keseluruhan ringkasan = (final_sekolah+final_umum+final_khusus)/3 (missing=0, tetap ÷3)
 
 ✅ Detail entitas: Indeks_Final_0_100 menempel dari Agregat Wilayah (Keseluruhan) (bukan per-row)
-✅ Bell curve per JENIS berbasis indeks per entitas (row-level)
-✅ LLM analysis + Word
+✅ Bell curve per JENIS berbasis skor kinerja per entitas (row-level)
+
+✅ METODE PENILAIAN KINERJA (REKOMENDASI UTAMA):
+   - Tetap tampilkan skor absolut: Indeks_Final_... (disesuaikan target 33.88%)
+   - Tambahkan skor kinerja relatif yang stabil & audit-friendly:
+     1) Score_Kinerja_Percentile_0_100 (0–100)  ← utama
+     2) Score_Kinerja_RobustZ_0_100 (0–100; 50+10*z_robust) ← opsional, tahan outlier
+
 ✅ Download (tanpa upload box)
 ✅ Download Data Mentah (.xlsx) = RAW hasil filter (bukan agregat)
 
 FIX DISPLAY:
 ✅ “null/NaN” untuk target/pop/coverage jenis -> dibuat 0 agar tidak tampil null
 ✅ Verifikasi target 33.88% (tanpa koma untuk integer) -> target/pop/gap dibulatkan integer
-✅ TABEL faktor_wilayah:
-   - target_total_33_88 -> bilangan bulat
-   - pop_total -> bilangan bulat
-   - coverage_total_% -> decimal 2 digit
 ✅ TABEL "Agregat Wilayah × Jenis" (UI) hanya sampai kolom Indeks_Dasar_Agregat_0_100
 """
 
@@ -66,9 +68,13 @@ POP_KHUSUS  = os.getenv("POP_KHUSUS",  "Data_populasi_perp_khusus.xlsx")
 W_KEPATUHAN = float(os.getenv("W_KEPATUHAN", "0.30"))
 W_KINERJA   = float(os.getenv("W_KINERJA",   "0.70"))
 
-# ✅ UPDATE: target sampel 33.88% (bukan 68%)
+# ✅ target sampel 33.88%
 TARGET_RATIO = float(os.getenv("TARGET_RATIO", "0.3388"))
 
+# Kinerja relatif
+USE_PERCENTILE = True
+USE_ROBUST_Z   = True
+
 USE_LLM = True
 LLM_MODEL_NAME = os.getenv("LLM_MODEL_NAME", "meta-llama/Meta-Llama-3-8B-Instruct")
 HF_TOKEN = (
@@ -202,6 +208,69 @@ def faktor_penyesuaian_total(n_total: float, target_total: float) -> float:
         n_total = 0.0
     return float(min(float(n_total) / float(target_total), 1.0))
 
+def _clip01(x):
+    if pd.isna(x):
+        return 0.0
+    return float(min(max(float(x), 0.0), 1.0))
+
+def add_kinerja_scores(
+    df: pd.DataFrame,
+    score_col: str,
+    group_cols: list[str] | None,
+    prefix: str = "Score_Kinerja"
+) -> pd.DataFrame:
+    """
+    Tambah:
+    - {prefix}_Percentile_0_100
+    - {prefix}_RobustZ_0_100 (50+10*z_robust, clip 0..100)
+    Grouping untuk fairness: misal per Jenis.
+    """
+    if df is None or df.empty or score_col not in df.columns:
+        return df
+
+    out = df.copy()
+    x = pd.to_numeric(out[score_col], errors="coerce").astype(float)
+
+    # Percentile 0–100
+    if USE_PERCENTILE:
+        if group_cols:
+            out[f"{prefix}_Percentile_0_100"] = (
+                out.groupby(group_cols, dropna=False)[score_col]
+                   .rank(pct=True, method="average") * 100.0
+            )
+        else:
+            out[f"{prefix}_Percentile_0_100"] = out[score_col].rank(pct=True, method="average") * 100.0
+        out[f"{prefix}_Percentile_0_100"] = pd.to_numeric(out[f"{prefix}_Percentile_0_100"], errors="coerce").fillna(0.0).clip(0, 100).round(2)
+
+    # Robust Z to 0–100
+    if USE_ROBUST_Z:
+        def _robustz_to_0_100(s: pd.Series) -> pd.Series:
+            v = pd.to_numeric(s, errors="coerce").astype(float)
+            v = v.replace([np.inf, -np.inf], np.nan)
+            if v.dropna().shape[0] < 2:
+                return pd.Series(50.0, index=v.index)  # netral
+            med = float(np.nanmedian(v.values))
+            mad = float(np.nanmedian(np.abs(v.values - med)))
+            if not np.isfinite(mad) or mad <= 1e-12:
+                sd = float(np.nanstd(v.values, ddof=1))
+                if not np.isfinite(sd) or sd <= 1e-12:
+                    return pd.Series(50.0, index=v.index)
+                z = (v - med) / sd
+            else:
+                z = (v - med) / (1.4826 * mad)
+            score = 50.0 + 10.0 * z
+            score = score.clip(0, 100).fillna(50.0)
+            return score
+
+        if group_cols:
+            out[f"{prefix}_RobustZ_0_100"] = out.groupby(group_cols, dropna=False)[score_col].transform(_robustz_to_0_100)
+        else:
+            out[f"{prefix}_RobustZ_0_100"] = _robustz_to_0_100(out[score_col])
+
+        out[f"{prefix}_RobustZ_0_100"] = pd.to_numeric(out[f"{prefix}_RobustZ_0_100"], errors="coerce").fillna(50.0).clip(0, 100).round(2)
+
+    return out
+
 
 # ============================================================
 # 3) INDIKATOR IPLM
@@ -279,6 +348,7 @@ def prepare_global(df_src: pd.DataFrame) -> pd.DataFrame:
         return df_src
     df = df_src.copy()
 
+    # rename indikator
     rename_map = {}
     for col in df.columns:
         c = _canon(col)
@@ -296,6 +366,7 @@ def prepare_global(df_src: pd.DataFrame) -> pd.DataFrame:
     for c in available:
         df[c] = df[c].apply(coerce_num)
 
+    # YJ per indikator + MinMax global
     for c in available:
         x = pd.to_numeric(df[c], errors="coerce").astype(float).values
         mask = ~np.isnan(x)
@@ -343,7 +414,6 @@ def _parse_pop_khusus(path_xlsx: str) -> pd.DataFrame:
     if df is None or df.empty:
         return pd.DataFrame()
 
-    # file kamu: Propinsi/Kab/kota | POP_KHUSUS | SAMPEL_KHUSUS_68% (kolom target boleh ada, tapi kita akan hitung ulang 33.88%)
     c_mix = pick_col(df, [
         "Propinsi/Kab/kota", "Propinsi/Kab/Kota", "Provinsi/Kab/Kota",
         "Provinsi/Kab/kota", "Provinsi/Kabupaten/Kota",
@@ -367,11 +437,9 @@ def _parse_pop_khusus(path_xlsx: str) -> pd.DataFrame:
         if mm == "":
             continue
 
-        # === PROV row: dianggap TOTAL PROVINSI (punya nilai!) ===
         if mm.startswith("PROVINSI "):
             prov_name = mm.replace("PROVINSI", "").strip()
             current_prov = prov_name
-
             rows.append({
                 "LEVEL": "PROV",
                 "Provinsi_Label": f"PROVINSI {prov_name}",
@@ -380,7 +448,6 @@ def _parse_pop_khusus(path_xlsx: str) -> pd.DataFrame:
             })
             continue
 
-        # === KAB/KOTA row ===
         rows.append({
             "LEVEL": "KAB",
             "Provinsi_Label": f"PROVINSI {current_prov}" if current_prov else None,
@@ -392,18 +459,11 @@ def _parse_pop_khusus(path_xlsx: str) -> pd.DataFrame:
     if pop.empty:
         return pop
 
-    pop["Pop_Total_Jenis"] = pd.to_numeric(pop["Pop_Total_Jenis"], errors="coerce")
-
-    # fallback aman: kalau pop kosong, tetap 0
-    pop["Pop_Total_Jenis"] = pop["Pop_Total_Jenis"].fillna(0.0)
-
-    # keys
+    pop["Pop_Total_Jenis"] = pd.to_numeric(pop["Pop_Total_Jenis"], errors="coerce").fillna(0.0)
     pop["prov_key"] = pop["Provinsi_Label"].apply(norm_prov_label)
     pop["kab_key"] = pop["Kab_Kota_Label"].apply(norm_kab_label) if "Kab_Kota_Label" in pop.columns else None
-
     return pop
 
-
 def load_default_files(force=False):
     key = (
         DATA_FILE, POP_KAB, POP_PROV, POP_KHUSUS,
@@ -464,14 +524,10 @@ def load_default_files(force=False):
     df_raw = df_raw.drop_duplicates(subset=["_row_key"], keep="first").copy()
     after = len(df_raw)
 
-    # =========================
     # POP KAB
-    # =========================
     pk = pd.read_excel(POP_KAB)
-
     c_kab  = pick_col(pk, ["KABUPATEN_KOTA","Kab/Kota","Kabupaten/Kota","KAB/KOTA","Kabupaten_Kota","kab_kota","kabupaten_kota"])
     c_prov = pick_col(pk, ["PROVINSI","Provinsi","provinsi"])
-
     if c_kab is None:
         info = "❌ POP_KAB: wajib ada kolom Kab/Kota."
         _CACHE.update({"key": key, "df_all": None, "df_raw": None, "pop_kab": None, "pop_prov": None, "pop_khusus": None, "meta": {}, "info": info})
@@ -483,11 +539,8 @@ def load_default_files(force=False):
     pop_kab["kab_key"] = pop_kab["Kab_Kota_Label"].apply(norm_kab_label)
     pop_kab = pop_kab.groupby("kab_key", as_index=False).first()
 
-    # =========================
     # POP PROV
-    # =========================
     pp = pd.read_excel(POP_PROV)
-
     c_pr = pick_col(pp, ["Provinsi","PROVINSI","provinsi","Propinsi","PROPINSI","propinsi"])
     if c_pr is None:
         info = "❌ POP_PROV: wajib ada kolom Provinsi."
@@ -499,9 +552,7 @@ def load_default_files(force=False):
     pop_prov["prov_key"] = pop_prov["Provinsi_Label"].apply(norm_prov_label)
     pop_prov = pop_prov.groupby("prov_key", as_index=False).first()
 
-    # =========================
     # POP KHUSUS
-    # =========================
     try:
         pop_khusus = _parse_pop_khusus(POP_KHUSUS)
     except Exception as e:
@@ -510,7 +561,6 @@ def load_default_files(force=False):
         return None, None, None, None, None, {}, info
 
     df_all = prepare_global(df_raw)
-
     meta = dict(prov_col=prov_col, kab_col=kab_col, kew_col=kew_col, jenis_col=jenis_col, nama_col=nama_col)
 
     info = (
@@ -572,16 +622,13 @@ def build_faktor_wilayah_jenis(
             base_pop["kab_key"] = base_pop["Kab_Kota_Label"].apply(norm_kab_label) if "Kab_Kota_Label" in base_pop.columns else base_pop.iloc[:, 0].apply(norm_kab_label)
         base_pop = base_pop.set_index("kab_key") if (not base_pop.empty and "kab_key" in base_pop.columns) else pd.DataFrame().set_index(pd.Index([]))
 
-    # =========================================================
-    # ✅ GRID WAJIB: semua wilayah × 3 jenis (meski n=0)
-    # =========================================================
+    # GRID: semua wilayah × 3 jenis
     base_keys = df[[key_col, label_col]].drop_duplicates().rename(columns={key_col: "group_key", label_col: label_name})
     full = base_keys.assign(_tmp=1).merge(
         pd.DataFrame({"Jenis": jenis_list, "_tmp": 1}),
         on="_tmp"
     ).drop(columns="_tmp")
 
-    # hitung n per jenis dari DM (boleh 0)
     cnt = (
         df.groupby([key_col, label_col, "_dataset"], dropna=False)
           .size()
@@ -593,14 +640,10 @@ def build_faktor_wilayah_jenis(
     base_n = full.merge(cnt, on=["group_key", label_name, "Jenis"], how="left")
     base_n["n_jenis"] = pd.to_numeric(base_n["n_jenis"], errors="coerce").fillna(0).astype(int)
 
-    # kolom output faktor (target 33.88%)
     base_n["target_total_33_88_jenis"] = 0.0
     base_n["pop_total_jenis"] = 0.0
 
-    # =========================
     # SEKOLAH + UMUM dari POP_KAB / POP_PROV
-    # Target dihitung ulang: pop * TARGET_RATIO
-    # =========================
     if not base_pop.empty:
         if mode == "KAB":
             pop_sekolah = pd.to_numeric(base_pop.get("jumlah_populasi_sekolah", 0), errors="coerce").fillna(0.0)
@@ -627,10 +670,7 @@ def build_faktor_wilayah_jenis(
         base_n.loc[m, "pop_total_jenis"] = base_n.loc[m, "group_key"].map(pop_umum).fillna(0.0).values
         base_n.loc[m, "target_total_33_88_jenis"] = base_n.loc[m, "group_key"].map(tgt_umum).fillna(0.0).values
 
-    # =========================
     # KHUSUS dari POP_KHUSUS
-    # Target dihitung ulang: pop * TARGET_RATIO
-    # =========================
     if pop_khusus is not None and not pop_khusus.empty:
         pk = pop_khusus.copy()
         pk["Pop_Total_Jenis"] = pd.to_numeric(pk.get("Pop_Total_Jenis", 0), errors="coerce").fillna(0.0)
@@ -650,14 +690,12 @@ def build_faktor_wilayah_jenis(
         base_n.loc[m, "pop_total_jenis"] = base_n.loc[m, "group_key"].map(pop_series).fillna(0.0).values
         base_n.loc[m, "target_total_33_88_jenis"] = base_n.loc[m, "group_key"].map(tgt_series).fillna(0.0).values
 
-    # fallback pop dari target (jaga-jaga)
     base_n["target_total_33_88_jenis"] = pd.to_numeric(base_n["target_total_33_88_jenis"], errors="coerce").fillna(0.0)
     base_n["pop_total_jenis"] = pd.to_numeric(base_n["pop_total_jenis"], errors="coerce").fillna(0.0)
 
     m_need_pop = (base_n["pop_total_jenis"] <= 0) & (base_n["target_total_33_88_jenis"] > 0)
     base_n.loc[m_need_pop, "pop_total_jenis"] = base_n.loc[m_need_pop, "target_total_33_88_jenis"] / float(TARGET_RATIO)
 
-    # faktor / coverage / gap
     base_n["faktor_penyesuaian_jenis"] = [
         faktor_penyesuaian_total(n, t)
         for n, t in zip(
@@ -714,14 +752,14 @@ def build_agg_wilayah_jenis(df_filtered: pd.DataFrame, faktor_wilayah_jenis: pd.
 
     jenis_list = ["sekolah", "umum", "khusus"]
 
-    # GRID: semua wilayah × 3 jenis
+    # GRID semua wilayah × 3 jenis
     base_keys = df[[key_col, label_col]].drop_duplicates().rename(columns={key_col: "group_key", label_col: label_name})
     full = base_keys.assign(_tmp=1).merge(
         pd.DataFrame({"Jenis": jenis_list, "_tmp": 1}),
         on="_tmp"
     ).drop(columns="_tmp")
 
-    # agregat dari data yang ada
+    # agregat real
     agg_real = df.groupby([key_col, label_col, "_dataset"], dropna=False).agg(
         Jumlah=("Indeks_Dasar_0_100", "size"),
         Rata2_sub_koleksi=("sub_koleksi", "mean"),
@@ -735,7 +773,6 @@ def build_agg_wilayah_jenis(df_filtered: pd.DataFrame, faktor_wilayah_jenis: pd.
 
     agg_real["Jenis"] = agg_real["Jenis"].astype(str).str.lower().str.strip()
 
-    # tempel ke grid + fill 0
     agg = full.merge(agg_real, on=["group_key", label_name, "Jenis"], how="left")
     for c in ["Jumlah","Rata2_sub_koleksi","Rata2_sub_sdm","Rata2_sub_pelayanan","Rata2_sub_pengelolaan",
               "Rata2_dim_kepatuhan","Rata2_dim_kinerja","Indeks_Dasar_Agregat_0_100"]:
@@ -744,7 +781,7 @@ def build_agg_wilayah_jenis(df_filtered: pd.DataFrame, faktor_wilayah_jenis: pd.
 
     agg["Jumlah"] = agg["Jumlah"].round(0).astype(int)
 
-    # merge faktor PER JENIS
+    # merge faktor jenis
     if faktor_wilayah_jenis is None or faktor_wilayah_jenis.empty:
         agg["faktor_penyesuaian_jenis"] = 1.0
         agg["target_total_33_88_jenis"] = 0
@@ -771,13 +808,21 @@ def build_agg_wilayah_jenis(df_filtered: pd.DataFrame, faktor_wilayah_jenis: pd.
         if "coverage_jenis_%" in agg.columns:
             agg["coverage_jenis_%"] = pd.to_numeric(agg["coverage_jenis_%"], errors="coerce").fillna(0.0).round(2)
 
-    # Indeks FINAL PER JENIS
+    # Indeks FINAL per jenis
     agg["Indeks_Final_Agregat_0_100"] = (
         pd.to_numeric(agg["Indeks_Dasar_Agregat_0_100"], errors="coerce").fillna(0.0)
         * pd.to_numeric(agg["faktor_penyesuaian_jenis"], errors="coerce").fillna(1.0)
     )
 
-    # rounding tampilan
+    # Kinerja relatif per jenis (dibandingkan sesama jenis)
+    agg = add_kinerja_scores(
+        agg,
+        score_col="Indeks_Final_Agregat_0_100",
+        group_cols=["Jenis"],
+        prefix="Score_Kinerja_WilayahJenis"
+    )
+
+    # rounding
     for c in [
         "Rata2_sub_koleksi","Rata2_sub_sdm","Rata2_sub_pelayanan","Rata2_sub_pengelolaan",
         "Rata2_dim_kepatuhan","Rata2_dim_kinerja"
@@ -804,14 +849,12 @@ def build_agg_wilayah_total_from_jenis(agg_jenis: pd.DataFrame, faktor_wilayah_j
 
     kew_norm = str(kew_value or "").upper()
     label_name = "Provinsi" if "PROV" in kew_norm else "Kab/Kota"
-
     jenis_list = ["sekolah", "umum", "khusus"]
 
     a = agg_jenis.copy()
     a["Jenis"] = a["Jenis"].astype(str).str.lower().str.strip()
 
     base_keys = a[["group_key", label_name]].drop_duplicates()
-
     full = base_keys.assign(_tmp=1).merge(
         pd.DataFrame({"Jenis": jenis_list, "_tmp": 1}),
         on="_tmp"
@@ -832,7 +875,6 @@ def build_agg_wilayah_total_from_jenis(agg_jenis: pd.DataFrame, faktor_wilayah_j
         how="left"
     )
 
-    # missing=0 (avg3 tetap ÷3)
     for c in cols_present:
         full[c] = pd.to_numeric(full[c], errors="coerce").fillna(0.0)
 
@@ -848,7 +890,7 @@ def build_agg_wilayah_total_from_jenis(agg_jenis: pd.DataFrame, faktor_wilayah_j
         Indeks_Final_Wilayah_0_100=("Indeks_Final_Agregat_0_100", "mean"),
     )
 
-    # tempel Pop/Target/Terkumpul per jenis & total
+    # Tempel info Pop/Target/N per jenis + total
     if faktor_wilayah_jenis is not None and not faktor_wilayah_jenis.empty:
         fw = faktor_wilayah_jenis.copy()
         fw["Jenis"] = fw["Jenis"].astype(str).str.lower().str.strip()
@@ -859,24 +901,19 @@ def build_agg_wilayah_total_from_jenis(agg_jenis: pd.DataFrame, faktor_wilayah_j
             values=["pop_total_jenis", "target_total_33_88_jenis", "n_jenis", "gap_target33_88_jenis", "faktor_penyesuaian_jenis"],
             aggfunc="first"
         )
-
         piv.columns = [f"{v}_{k}" for v, k in piv.columns]
         piv = piv.reset_index()
-
         out = out.merge(piv, on=["group_key", label_name], how="left")
 
-        # NaN -> 0 / 1
         for j in ["sekolah", "umum", "khusus"]:
             for basecol in ["pop_total_jenis", "target_total_33_88_jenis", "n_jenis", "gap_target33_88_jenis"]:
                 c = f"{basecol}_{j}"
                 if c in out.columns:
                     out[c] = pd.to_numeric(out[c], errors="coerce").fillna(0).round(0).astype(int)
-
             cfac = f"faktor_penyesuaian_jenis_{j}"
             if cfac in out.columns:
                 out[cfac] = pd.to_numeric(out[cfac], errors="coerce").fillna(1.0).round(3)
 
-        # TOTAL (sum 3 jenis)
         out["pop_total_all"] = (
             out.get("pop_total_jenis_sekolah", 0)
             + out.get("pop_total_jenis_umum", 0)
@@ -902,6 +939,14 @@ def build_agg_wilayah_total_from_jenis(agg_jenis: pd.DataFrame, faktor_wilayah_j
         )
         out["coverage_target33_88_all_%"] = pd.to_numeric(out["coverage_target33_88_all_%"], errors="coerce").fillna(0.0).round(2)
 
+    # Tambah skor kinerja relatif untuk keseluruhan wilayah (dibandingkan seluruh wilayah)
+    out = add_kinerja_scores(
+        out,
+        score_col="Indeks_Final_Wilayah_0_100",
+        group_cols=None,
+        prefix="Score_Kinerja_WilayahTotal"
+    )
+
     # rounding index
     for c in [
         "Rata2_sub_koleksi","Rata2_sub_sdm","Rata2_sub_pelayanan","Rata2_sub_pengelolaan",
@@ -1035,6 +1080,7 @@ def build_summary_per_jenis(agg_jenis: pd.DataFrame, agg_total: pd.DataFrame):
 
 # ============================================================
 # 10) DETAIL ENTITAS: Final menempel dari agg_total (wilayah)
+#     + skor kinerja relatif per jenis (entitas-level)
 # ============================================================
 
 def attach_final_to_detail(df_filtered: pd.DataFrame, agg_total: pd.DataFrame, meta: dict, kew_value: str):
@@ -1078,6 +1124,14 @@ def attach_final_to_detail(df_filtered: pd.DataFrame, agg_total: pd.DataFrame, m
     out = df[keep].copy()
     out = out.rename(columns={label_cols[0]:"Provinsi", label_cols[1]:"Kab/Kota", "_dataset":"Jenis"})
 
+    # skor kinerja relatif per entitas (dibandingkan sesama jenis)
+    out = add_kinerja_scores(
+        out,
+        score_col="Indeks_Dasar_0_100",
+        group_cols=["Jenis"],
+        prefix="Score_Kinerja_Entitas"
+    )
+
     for c in ["sub_koleksi","sub_sdm","sub_pelayanan","sub_pengelolaan","dim_kepatuhan","dim_kinerja"]:
         if c in out.columns:
             out[c] = pd.to_numeric(out[c], errors="coerce").fillna(0.0).round(3)
@@ -1129,7 +1183,7 @@ def _make_bell_curve(dfp: pd.DataFrame, xcol: str, title: str, label_col: str |
     fig = go.Figure()
     fig.update_layout(
         title=title,
-        xaxis_title="Indeks (0–100)",
+        xaxis_title="Skor (0–100)",
         yaxis_title="Kepadatan",
         hovermode="x unified",
         margin=dict(l=40, r=20, t=60, b=40),
@@ -1157,7 +1211,7 @@ def _make_bell_curve(dfp: pd.DataFrame, xcol: str, title: str, label_col: str |
         fig.add_trace(go.Scatter(
             x=[x_single], y=[0], mode="markers", name="Data", marker=dict(size=10),
             hovertext=hovertext,
-            hovertemplate="%{hovertext}<extra></extra>" if hovertext is not None else "Indeks: %{x:.2f}<extra></extra>",
+            hovertemplate="%{hovertext}<extra></extra>" if hovertext is not None else "Skor: %{x:.2f}<extra></extra>",
             showlegend=False,
         ))
         fig.add_vline(x=x_single, line_width=1, line_dash="dash", annotation_text=f"Nilai: {x_single:.1f}", annotation_position="top")
@@ -1189,7 +1243,7 @@ def _make_bell_curve(dfp: pd.DataFrame, xcol: str, title: str, label_col: str |
     pdf = (1.0 / (sigma * np.sqrt(2 * np.pi))) * np.exp(-0.5 * ((xs - mu) / sigma) ** 2)
 
     fig.add_trace(go.Scatter(
-        x=xs, y=pdf, mode="lines", name="Kurva Normal",
+        x=xs, y=pdf, mode="lines", name="Kurva Normal (fit)",
         hovertemplate="x=%{x:.2f}<br>pdf=%{y:.4f}<extra></extra>"
     ))
 
@@ -1219,7 +1273,7 @@ def _make_bell_curve(dfp: pd.DataFrame, xcol: str, title: str, label_col: str |
     fig.add_trace(go.Scatter(
         x=x, y=np.zeros_like(x), mode="markers", name="Data", marker=dict(size=8),
         hovertext=hovertext,
-        hovertemplate="%{hovertext}<extra></extra>" if hovertext is not None else "Indeks: %{x:.2f}<extra></extra>",
+        hovertemplate="%{hovertext}<extra></extra>" if hovertext is not None else "Skor: %{x:.2f}<extra></extra>",
         showlegend=False
     ))
 
@@ -1233,7 +1287,7 @@ def _make_bell_curve(dfp: pd.DataFrame, xcol: str, title: str, label_col: str |
 
 
 # ============================================================
-# 13) KPI DASHBOARD (FINAL: hanya Final & Dasar)
+# 13) KPI DASHBOARD (FINAL: skor absolut)
 # ============================================================
 
 def compute_dashboard_kpis(summary_jenis: pd.DataFrame):
@@ -1245,7 +1299,6 @@ def compute_dashboard_kpis(summary_jenis: pd.DataFrame):
 
     final_all = _get("keseluruhan", "Indeks_Final_Disesuaikan_0_100")
     dasar_all = _get("keseluruhan", "Indeks_Dasar_0_100")
-
     return {"final_all": final_all, "dasar_all": dasar_all}
 
 def build_kpi_markdown(summary_jenis: pd.DataFrame) -> str:
@@ -1259,15 +1312,15 @@ def build_kpi_markdown(summary_jenis: pd.DataFrame) -> str:
     return f"""
 <div style="display:flex; gap:12px; flex-wrap:wrap;">
   <div style="border:1px solid #333; border-radius:10px; padding:10px 12px; min-width:260px;">
-    <div style="opacity:0.8;">Indeks IPLM FINAL (Disesuaikan)</div>
+    <div style="opacity:0.8;">Indeks IPLM FINAL (Disesuaikan 33.88%)</div>
     <div style="font-size:26px; font-weight:700;">{fmt(k["final_all"],2)}</div>
-    <div style="opacity:0.7;">Sumber: Ringkasan baris “keseluruhan”</div>
+    <div style="opacity:0.7;">Skor absolut (untuk akuntabilitas)</div>
   </div>
 
   <div style="border:1px solid #333; border-radius:10px; padding:10px 12px; min-width:260px;">
     <div style="opacity:0.8;">Indeks Dasar (Tanpa Penyesuaian)</div>
     <div style="font-size:26px; font-weight:700;">{fmt(k["dasar_all"],2)}</div>
-    <div style="opacity:0.7;">Sumber: Ringkasan baris “keseluruhan”</div>
+    <div style="opacity:0.7;">Sebelum faktor kecukupan sampel</div>
   </div>
 </div>
 """.strip()
@@ -1313,6 +1366,15 @@ def build_context(summary_jenis: pd.DataFrame, agg_total: pd.DataFrame, verif_to
             wl = r.get(label_col, "(wilayah)") if label_col else "(wilayah)"
             lines.append(f"- {wl}: Final={float(r['Indeks_Final_Wilayah_0_100']):.2f}")
 
+    # kinerja relatif (percentile) jika ada
+    if agg_total is not None and not agg_total.empty and "Score_Kinerja_WilayahTotal_Percentile_0_100" in agg_total.columns:
+        label_col = "Kab/Kota" if "Kab/Kota" in agg_total.columns else ("Provinsi" if "Provinsi" in agg_total.columns else None)
+        lines.append("\nTop 5 wilayah (Percentile kinerja tertinggi):")
+        top = agg_total.sort_values("Score_Kinerja_WilayahTotal_Percentile_0_100", ascending=False).head(5)
+        for _, r in top.iterrows():
+            wl = r.get(label_col, "(wilayah)") if label_col else "(wilayah)"
+            lines.append(f"- {wl}: Pctl={float(r['Score_Kinerja_WilayahTotal_Percentile_0_100']):.2f}")
+
     return "\n".join(lines)
 
 def generate_llm_analysis(summary_jenis, agg_total, verif_total, wilayah, kew):
@@ -1328,8 +1390,8 @@ DATA IPLM (RINGKAS):
 {ctx}
 
 Buat analisis 3 paragraf:
-1) Gambaran umum.
-2) Per jenis (sekolah/umum/khusus) + keseluruhan.
+1) Gambaran umum (skor absolut).
+2) Kinerja relatif (percentile) + per jenis.
 3) Rekomendasi singkat.
 Catatan: target sampel yang digunakan adalah {TARGET_RATIO*100:.2f}% (bukan 68%).
 """
@@ -1350,6 +1412,7 @@ def generate_word_report(wilayah, summary_jenis, analysis_text):
     doc = Document()
     doc.add_heading(f"Laporan IPLM — {wilayah}", level=1)
     doc.add_paragraph(f"Target sampel per jenis: {TARGET_RATIO*100:.2f}%")
+    doc.add_paragraph("Catatan: Skor kinerja relatif menggunakan Percentile (0–100) yang stabil terhadap bentuk distribusi.")
 
     doc.add_heading("Ringkasan (Jenis + Keseluruhan)", level=2)
 
@@ -1410,7 +1473,7 @@ def run_calc(prov_value, kab_value, kew_value, df_all, df_raw, pop_kab, pop_prov
         if df_all is None or df_all.empty or df_raw is None or df_raw.empty:
             return _empty_outputs("⚠️ Data belum ter-load. Pastikan file tersedia di repo/server.")
 
-        # FILTER (df_all)
+        # FILTER df_all
         df = df_all.copy()
         if prov_value and prov_value != "(Semua)":
             df = df[df["PROV_DISP"] == prov_value]
@@ -1431,7 +1494,7 @@ def run_calc(prov_value, kab_value, kew_value, df_all, df_raw, pop_kab, pop_prov
         verif_total = build_verif_jenis(faktor_wilayah_jenis, kew_value or "(Semua)")
         detail_view = attach_final_to_detail(df, agg_total, meta, kew_value or "(Semua)")
 
-        # view agg_jenis (UI cuma sampai indeks dasar)
+        # agg_jenis view (UI hanya sampai indeks dasar)
         if agg_jenis_full is None or agg_jenis_full.empty:
             agg_jenis_view = agg_jenis_full
         else:
@@ -1449,7 +1512,7 @@ def run_calc(prov_value, kab_value, kew_value, df_all, df_raw, pop_kab, pop_prov
             cols_upto = [c for c in cols_upto if c in agg_jenis_full.columns]
             agg_jenis_view = agg_jenis_full[cols_upto].copy()
 
-        # FILTER RAW DOWNLOAD (df_raw)
+        # FILTER RAW DOWNLOAD
         raw = df_raw.copy()
         if prov_value and prov_value != "(Semua)":
             raw = raw[raw["PROV_DISP"] == prov_value]
@@ -1458,27 +1521,28 @@ def run_calc(prov_value, kab_value, kew_value, df_all, df_raw, pop_kab, pop_prov
         if kew_value and kew_value != "(Semua)":
             raw = raw[raw["KEW_NORM"] == kew_value]
 
-        # bell curve per jenis (entitas)
+        # bell curve per jenis (entitas) -> gunakan Percentile (utama) kalau ada
         if detail_view is None or detail_view.empty:
-            fig_sekolah = _make_bell_curve(pd.DataFrame(), "Indeks_Dasar_0_100", "Bell Curve — Jenis: Sekolah", min_points=2)
-            fig_umum = _make_bell_curve(pd.DataFrame(), "Indeks_Dasar_0_100", "Bell Curve — Jenis: Umum", min_points=2)
-            fig_khusus = _make_bell_curve(pd.DataFrame(), "Indeks_Dasar_0_100", "Bell Curve — Jenis: Khusus", min_points=2)
+            fig_sekolah = _make_bell_curve(pd.DataFrame(), "Score_Kinerja_Entitas_Percentile_0_100", "Bell Curve — Jenis: Sekolah", min_points=2)
+            fig_umum = _make_bell_curve(pd.DataFrame(), "Score_Kinerja_Entitas_Percentile_0_100", "Bell Curve — Jenis: Umum", min_points=2)
+            fig_khusus = _make_bell_curve(pd.DataFrame(), "Score_Kinerja_Entitas_Percentile_0_100", "Bell Curve — Jenis: Khusus", min_points=2)
         else:
-            xcol_ent = "Indeks_Dasar_0_100" if "Indeks_Dasar_0_100" in detail_view.columns else "Indeks_Final_0_100"
+            xcol_ent = "Score_Kinerja_Entitas_Percentile_0_100" if "Score_Kinerja_Entitas_Percentile_0_100" in detail_view.columns else "Indeks_Dasar_0_100"
             label_col_e = "nm_perpustakaan" if "nm_perpustakaan" in detail_view.columns else None
-            hover_cols_e = [c for c in ["Provinsi", "Kab/Kota", "KEW_NORM", "Jenis", "Indeks_Dasar_0_100", "Indeks_Final_0_100"] if c in detail_view.columns]
+            hover_cols_e = [c for c in ["Provinsi", "Kab/Kota", "KEW_NORM", "Jenis", "Indeks_Dasar_0_100", "Indeks_Final_0_100", xcol_ent] if c in detail_view.columns]
 
             def _fig_jenis_ent(jenis_key: str, judul: str):
                 d = detail_view[detail_view["Jenis"].astype(str).str.lower() == jenis_key].copy()
                 return _make_bell_curve(d, xcol=xcol_ent, title=judul, label_col=label_col_e, hover_cols=hover_cols_e, min_points=2)
 
-            fig_sekolah = _fig_jenis_ent("sekolah", "Bell Curve — Jenis: Sekolah (Indeks per Entitas)")
-            fig_umum = _fig_jenis_ent("umum", "Bell Curve — Jenis: Umum (Indeks per Entitas)")
-            fig_khusus = _fig_jenis_ent("khusus", "Bell Curve — Jenis: Khusus (Indeks per Entitas)")
+            fig_sekolah = _fig_jenis_ent("sekolah", f"Bell Curve — Jenis: Sekolah (Skor: {xcol_ent})")
+            fig_umum = _fig_jenis_ent("umum", f"Bell Curve — Jenis: Umum (Skor: {xcol_ent})")
+            fig_khusus = _fig_jenis_ent("khusus", f"Bell Curve — Jenis: Khusus (Skor: {xcol_ent})")
 
         # KPI
         kpi_md = build_kpi_markdown(summary_jenis)
 
+        # export
         tmpdir = tempfile.mkdtemp()
         prov_slug = (_canon(prov_value or "SEMUA").upper() or "SEMUA")
         kab_slug = (_canon(kab_value or "SEMUA").upper() or "SEMUA")
@@ -1560,7 +1624,7 @@ def on_prov_change(prov_value):
 
 with gr.Blocks() as demo:
     gr.Markdown(f"""
-# IPLM 2025 — Final (Target Sampel **33.88%** per Jenis)
+# IPLM 2025 — Final (Target Sampel **33.88%** per Jenis) + Penilaian Kinerja Relatif (Percentile)
 **Mode NO UPLOAD (cache aktif).** File dibaca dari repo/server:
 - `DATA_FILE`   = **{DATA_FILE}**
 - `POP_KAB`     = **{POP_KAB}**
@@ -1569,10 +1633,12 @@ with gr.Blocks() as demo:
 
 **TARGET RATIO (per jenis): {TARGET_RATIO*100:.2f}%**
 
-**FIX UTAMA:**
-- Ringkasan tampil Pop/Target33.88/Terkumpul/Coverage untuk **sekolah, umum, khusus, keseluruhan**
-- Target dihitung ulang dari Pop: **target = pop × {TARGET_RATIO:.4f}**
-- Pop khusus dari POP_KHUSUS (Propinsi/Kab/kota | POP_KHUSUS | ...)
+**Kinerja Relatif (untuk evaluasi kinerja):**
+- `Score_Kinerja_*_Percentile_0_100` (utama, stabil tanpa asumsi normal)
+- `Score_Kinerja_*_RobustZ_0_100` (opsional, tahan outlier)
+
+**Skor Absolut (untuk akuntabilitas):**
+- `Indeks_Final_*` (sudah disesuaikan target 33.88%)
 """)
 
     state_df = gr.State(None)
@@ -1599,19 +1665,19 @@ with gr.Blocks() as demo:
     gr.Markdown("## Ringkasan (Jenis + Keseluruhan) — Pop/Target33.88/Terkumpul/Coverage + Penyesuaian")
     out_summary = gr.DataFrame(interactive=False)
 
-    gr.Markdown("## Agregat Wilayah (Keseluruhan) — FIX: avg3 dari 3 jenis")
+    gr.Markdown("## Agregat Wilayah (Keseluruhan) — FIX: avg3 dari 3 jenis + Skor Kinerja Relatif")
     out_agg_total = gr.DataFrame(interactive=False)
 
-    gr.Markdown("## Agregat Wilayah × Jenis (Sekolah, Umum, Khusus) — (ditampilkan sampai Indeks_Dasar_Agregat_0_100)")
+    gr.Markdown("## Agregat Wilayah × Jenis — (ditampilkan sampai Indeks_Dasar_Agregat_0_100)")
     out_agg_jenis = gr.DataFrame(interactive=False)
 
-    gr.Markdown("## Detail Entitas (Final menempel dari wilayah)")
+    gr.Markdown("## Detail Entitas (Final menempel dari wilayah + Skor Kinerja Relatif per Jenis)")
     out_detail = gr.DataFrame(interactive=False)
 
     gr.Markdown("## Kecukupan Sampel 33.88% (tanpa angka koma untuk integer)")
     out_verif = gr.DataFrame(interactive=False)
 
-    gr.Markdown("## Bell Curve — per Jenis Perpustakaan (Indeks per Entitas)")
+    gr.Markdown("## Bell Curve — per Jenis (berbasis Score_Kinerja_Entitas_Percentile_0_100 jika tersedia)")
     gr.Markdown("### Perpustakaan Umum")
     bell_umum = gr.Plot(scale=1)