Update app.py

app.py

@@ -1987,16 +1987,31 @@ else:
     st.info("No data available for non-positive issue categories with 100% coverage and positive trend.")
 
 # =================== OBJECTIVE 7 — Insight and Recommendation ===================
-st.markdown("<h3 class='section-title'>OBJECTIVE 7 — Insight and Recommendation</h3>", unsafe_allow_html=True)
+import streamlit as st
+import pandas as pd
+import requests
+
+# =========================
+#  UNIVERSAL LLM CALL
+# =========================
+def llm_call(prompt: str):
+    """Universal call untuk LLM (HF Docker / LM Studio / OpenAI)."""
+    url = "http://localhost:1234/v1/chat/completions"    # UBAH jika perlu
+    payload = {
+        "model": "Qwen2.5-7B-Instruct",                  # UBAH sesuai model
+        "messages": [{"role": "user", "content": prompt}],
+        "temperature": 0.3,
+        "max_tokens": 700
+    }
+    r = requests.post(url, json=payload)
+    r.raise_for_status()
+    return r.json()["choices"][0]["message"]["content"]
 
-# ✅ Pastikan df_filtered tersedia
-if 'df_filtered' not in st.session_state:
-    st.error("⚠️ `df_filtered` not found in session state. Please ensure filters are applied.")
-    st.stop()
-df_filtered = st.session_state.df_filtered
 
-# ✅ Definisi fungsi — dipastikan di global scope
-def extract_agentic_insights_v5(df: pd.DataFrame):
+# =========================
+#  EXTRACT INSIGHTS
+# =========================
+def extract_agentic_insights(df: pd.DataFrame):
     dev = {
         "lowest_ratio_9_locs": [],
         "obj3a_lowest_div": None,
@@ -2011,218 +2026,182 @@ def extract_agentic_insights_v5(df: pd.DataFrame):
         "obj6_top2_categories": [],
     }
 
-    # === 1. 9 locations with lowest finding-to-reporter ratio ===
+    # === 1. 9 lowest location ratios ===
     if {'nama_lokasi_full', 'creator_nid', 'created_at', 'kode_temuan'}.issubset(df.columns):
-        calc = df[['nama_lokasi_full', 'creator_nid', 'created_at', 'kode_temuan']].copy()
+        calc = df.copy()
         calc['created_at'] = pd.to_datetime(calc['created_at'], errors='coerce')
-        calc = calc.dropna(subset=['created_at', 'nama_lokasi_full', 'creator_nid'])
+        calc = calc.dropna(subset=['created_at', 'creator_nid'])
         calc['bulan'] = calc['created_at'].dt.to_period('M')
+
         monthly = calc.groupby(['nama_lokasi_full', 'bulan']).agg(
             findings=('kode_temuan', 'size'),
             reporters=('creator_nid', 'nunique')
         ).reset_index()
+
         monthly = monthly[monthly['reporters'] > 0]
         monthly['ratio'] = monthly['findings'] / monthly['reporters']
+
         loc_avg = monthly.groupby('nama_lokasi_full')['ratio'].mean()
-        lowest_9 = loc_avg.nsmallest(9)
-        dev["lowest_ratio_9_locs"] = [(loc, round(ratio, 2)) for loc, ratio in lowest_9.items()]
+        lowest9 = loc_avg.nsmallest(9)
+
+        dev["lowest_ratio_9_locs"] = [(k, round(v, 3)) for k, v in lowest9.items()]
 
-    # === 2a: Division — lowest ratio ===
+    # === 2. Divisions & reporters ===
     if {'nama', 'creator_nid', 'created_at', 'kode_temuan'}.issubset(df.columns):
-        calc = df[['nama', 'creator_nid', 'created_at', 'kode_temuan']].copy()
+        calc = df.copy()
         calc['bulan'] = pd.to_datetime(calc['created_at']).dt.to_period('M')
+
         agg = calc.groupby(['nama', 'bulan']).agg(
             findings=('kode_temuan', 'size'),
             reporters=('creator_nid', 'nunique')
-        )
-        agg = agg[agg['reporters'] > 0].reset_index()
+        ).reset_index()
+
+        agg = agg[agg['reporters'] > 0]
         agg['ratio'] = agg['findings'] / agg['reporters']
+
         div_ratio = agg.groupby('nama')['ratio'].mean()
         if not div_ratio.empty:
-            name = div_ratio.idxmin()
-            val = round(div_ratio.min(), 2)
-            dev["obj3a_lowest_div"] = (name, val)
+            dev["obj3a_lowest_div"] = (div_ratio.idxmin(), round(div_ratio.min(), 2))
 
-    # === 2b: Executor — slowest resolution ===
+    # Slowest executor
     if 'days_to_close' in df.columns:
-        valid = df[df['days_to_close'].notna() & (df['days_to_close'] >= 0)]
-        exec_col = 'nama_pic' if 'nama_pic' in valid.columns else 'creator_name'
+        valid = df[df['days_to_close'] >= 0]
+        exec_col = 'nama_pic' if 'nama_pic' in df.columns else 'creator_name'
+
         if exec_col in valid.columns:
             lead = valid.groupby(exec_col)['days_to_close'].mean()
-            if not lead.empty:
-                name = lead.idxmax()
-                val = round(lead.max(), 2)
-                dev["obj3b_slowest_executor"] = (name, val)
+            dev["obj3b_slowest_executor"] = (lead.idxmax(), round(lead.max(), 1))
 
-    # === 2c: Reporter — lowest frequency ===
+    # Lowest reporter
     if {'creator_name', 'created_at'}.issubset(df.columns):
-        calc = df[['creator_name', 'created_at']].copy()
+        calc = df.copy()
         calc['bulan'] = pd.to_datetime(calc['created_at']).dt.to_period('M')
-        monthly = calc.groupby(['creator_name', 'bulan']).size().reset_index(name='count')
-        avg = monthly.groupby('creator_name')['count'].mean()
-        avg = avg[avg > 0]
-        if not avg.empty:
-            name = avg.idxmin()
-            val = round(avg.min(), 2)
-            dev["obj3c_lowest_reporter"] = (name, val)
-
-    # === 2d: Division — slowest resolution ===
-    if 'days_to_close' in df.columns and 'nama' in df.columns:
-        valid = df[df['days_to_close'].notna() & (df['days_to_close'] >= 0)]
-        if not valid.empty:
-            lead = valid.groupby('nama')['days_to_close'].mean()
-            if not lead.empty:
-                name = lead.idxmax()
-                val = round(lead.max(), 2)
-                dev["obj3d_slowest_div"] = (name, val)
-
-    # === 3. Non-Positive composition ===
+        monthly = calc.groupby(['creator_name', 'bulan']).size()
+        avg = monthly.groupby('creator_name').mean()
+        dev["obj3c_lowest_reporter"] = (avg.idxmin(), round(avg.min(), 2))
+
+    # Slowest division (lead time)
+    if {'nama', 'days_to_close'}.issubset(df.columns):
+        lead = df.groupby('nama')['days_to_close'].mean()
+        dev["obj3d_slowest_div"] = (lead.idxmax(), round(lead.max(), 1))
+
+    # === 3. Non positive %
     if 'temuan_kategori' in df.columns:
         cnt = df['temuan_kategori'].value_counts(normalize=True) * 100
-        dev["obj4_unsafe_condition_pct"] = round(cnt.get("Unsafe Condition", 0), 2)
-        dev["obj4_unsafe_action_pct"] = round(cnt.get("Unsafe Action", 0), 2)
-        dev["obj4_near_miss_pct"] = round(cnt.get("Near Miss", 0), 2)
+        dev["obj4_unsafe_condition_pct"] = round(cnt.get("Unsafe Condition", 0), 1)
+        dev["obj4_unsafe_action_pct"] = round(cnt.get("Unsafe Action", 0), 1)
+        dev["obj4_near_miss_pct"] = round(cnt.get("Near Miss", 0), 1)
 
-    # === 4. Risk Quadrants ===
-    X_LIMIT, Y_LIMIT = 20, 3
+    # === 4. Quadrants ===
     if {'nama', 'created_at', 'days_to_close', 'kode_temuan'}.issubset(df.columns):
         calc = df.copy()
-        calc['created_at'] = pd.to_datetime(calc['created_at'], errors='coerce')
-        calc = calc.assign(month=calc['created_at'].dt.to_period('M').astype(str))
-        monthly_counts = calc.groupby(['nama', 'month'])['kode_temuan'].nunique().reset_index()
-        avg_count = monthly_counts.groupby('nama')['kode_temuan'].mean().reset_index(name='Finding Count')
-        leadtime = calc.groupby('nama')['days_to_close'].mean().reset_index(name='Avg Lead Time')
-        mat = avg_count.merge(leadtime, on='nama', how='left').fillna(0)
-        for _, r in mat.iterrows():
-            if r['Finding Count'] >= X_LIMIT and r['Avg Lead Time'] >= Y_LIMIT:
+        calc['created_at'] = pd.to_datetime(calc['created_at'])
+        calc['month'] = calc['created_at'].dt.to_period('M').astype(str)
+
+        monthly_count = calc.groupby(['nama', 'month'])['kode_temuan'].size().reset_index(name='count')
+        avg_count = monthly_count.groupby('nama')['count'].mean().reset_index(name='Finding Count')
+        avg_lead = calc.groupby('nama')['days_to_close'].mean().reset_index(name='Avg Lead Time')
+
+        m = avg_count.merge(avg_lead, on='nama')
+        X, Y = 20, 3
+
+        for _, r in m.iterrows():
+            if r['Finding Count'] >= X and r['Avg Lead Time'] >= Y:
                 dev["obj5_q1_divs"].append(r['nama'])
-            elif r['Finding Count'] < X_LIMIT and r['Avg Lead Time'] >= Y_LIMIT:
+            elif r['Finding Count'] < X and r['Avg Lead Time'] >= Y:
                 dev["obj5_q2_divs"].append(r['nama'])
 
-    # === 5. Top 2 non-Positive categories ===
+    # === 5. top 2 category ===
     if {'kategori', 'temuan_kategori', 'created_at'}.issubset(df.columns):
-        nonpos = df[df['temuan_kategori'] != 'Positive']
-        if not nonpos.empty:
-            start = nonpos['created_at'].min().to_period('M')
-            end = nonpos['created_at'].max().to_period('M')
-            n_months = len(pd.period_range(start=start, end=end, freq='M'))
-            cat_avg = (nonpos.groupby('kategori').size() / n_months).sort_values(ascending=False).head(2)
-            dev["obj6_top2_categories"] = [(cat, round(val, 2)) for cat, val in cat_avg.items()]
+        nonpos = df[df['temuan_kategori'] != "Positive"]
+        start = nonpos['created_at'].min().to_period('M')
+        end = nonpos['created_at'].max().to_period('M')
+        n_months = len(pd.period_range(start, end, freq='M'))
+        cat_avg = nonpos.groupby('kategori').size() / n_months
+        cat_avg = cat_avg.sort_values(ascending=False).head(2)
+        dev["obj6_top2_categories"] = [(k, round(v, 1)) for k, v in cat_avg.items()]
 
     return dev
 
-# === Jalankan ekstraksi ===
-dev = extract_agentic_insights_v5(df_filtered)
 
-# === Siapkan entri tabel ===
-entries = []
+# =========================
+#  RENDER + LLM GENERATION
+# =========================
 
-# 1. Low-ratio locations
-if dev["lowest_ratio_9_locs"]:
-    loc_list = ", ".join([f"{loc} ({ratio:.2f})" for loc, ratio in dev["lowest_ratio_9_locs"]])
-    entries.append({
-        "Risk Category": "Reporting Coverage Risk",
-        "Insight": f"Nine locations with the lowest finding-to-reporter ratio: {loc_list}.",
-        "Recommendation": "Launch <em>Agency Activation Sprint</em>: assign Safety Champions to conduct ≥1 spot inspection/week per site.",
-        "Mitigation": "Deploy QR-code checklists + automated WhatsApp reminders. Target: ratio ≥0.5 within 45 days."
-    })
+dev = extract_agentic_insights(df_filtered)
 
-# 2. Capacity imbalance
-parts = []
-if dev["obj3a_lowest_div"]:
-    name, val = dev["obj3a_lowest_div"]
-    parts.append(f"division <strong>{name}</strong> (ratio: {val:.2f})")
-if dev["obj3c_lowest_reporter"]:
-    name, val = dev["obj3c_lowest_reporter"]
-    parts.append(f"reporter <strong>{name}</strong> ({val:.2f} findings/month)")
-if dev["obj3d_slowest_div"]:
-    name, val = dev["obj3d_slowest_div"]
-    parts.append(f"division <strong>{name}</strong> (avg. resolution: {val:.2f} days)")
-if dev["obj3b_slowest_executor"]:
-    name, val = dev["obj3b_slowest_executor"]
-    parts.append(f"executor <strong>{name}</strong> (avg. resolution: {val:.2f} days)")
-
-if parts:
-    insight = f"Uneven operational capacity detected: {'; '.join(parts)}."
-    entries.append({
-        "Risk Category": "Capacity Imbalance Risk",
-        "Insight": insight,
-        "Recommendation": "Activate <em>Agentic Capacity Dashboard</em> for real-time monitoring of reporting & resolution KPIs.",
-        "Mitigation": "Auto-trigger coaching alerts to Area PICs if deviation >20% from baseline, with peer benchmarking."
-    })
+# ======== BUILD TEXT FOR LLM ========
+prompt = f"""
+You are an advanced Safety Analytics LLM.
 
-# 3. Non-Positive composition
-uc, ua, nm = dev["obj4_unsafe_condition_pct"], dev["obj4_unsafe_action_pct"], dev["obj4_near_miss_pct"]
-if uc + ua + nm > 0:
-    insight = f"Non-Positive finding composition: Unsafe Condition ({uc:.2f}%), Unsafe Action ({ua:.2f}%), Near Miss ({nm:.2f}%)."
-    entries.append({
-        "Risk Category": "Data Quality & Categorization Risk",
-        "Insight": insight,
-        "Recommendation": "Enforce photo-based validation for all Unsafe Condition/Action/Near Miss submissions.",
-        "Mitigation": "System blocks submission if photo evidence or justification is missing."
-    })
+Given the following structured insights from real safety operational data:
 
-# 4. Risk Quadrants
-if dev["obj5_q1_divs"] or dev["obj5_q2_divs"]:
-    q1 = ", ".join([f"{d}" for d in dev["obj5_q1_divs"][:3]]) or "—"
-    q2 = ", ".join([f"{d}" for d in dev["obj5_q2_divs"][:3]]) or "—"
-    insight = f"High-risk divisions (Q1): {q1}; Hidden-risk divisions (Q2): {q2}."
-    entries.append({
-        "Risk Category": "SLA & Backlog Risk",
-        "Insight": insight,
-        "Recommendation": "Assign dedicated safety crews to QI divisions; enforce <em>One Finding, One Day</em> closure for QII.",
-        "Mitigation": "Auto-generate executive escalation reports to VP Ops if any division remains in QI/QII ≥2 months."
-    })
+{dev}
 
-# 5. Top categories
-if dev["obj6_top2_categories"]:
-    c1, c2 = dev["obj6_top2_categories"]
-    insight = f"Top recurring non-Positive categories: <strong>{c1[0]}</strong> ({c1[1]:.2f}/month) and <strong>{c2[0]}</strong> ({c2[1]:.2f}/month)."
-    entries.append({
-        "Risk Category": "Recurring Hazard Risk",
-        "Insight": insight,
-        "Recommendation": f"Form cross-functional <em>RCA Task Force</em> (Civil, Electrical, HSE, Contractors) for <strong>{c1[0]}</strong> and <strong>{c2[0]}</strong>.",
-        "Mitigation": "Update tender templates: all bids must include historical mitigations for these categories."
-    })
+Your tasks:
 
-# === RENDER TABEL TERPADU ===
-if entries:
-    rows = []
-    for e in entries:
-        rows.append(f"""
-        <tr>
-            <td style="padding:12px; font-weight:bold; vertical-align:top; background-color:#f8f9fa;">{e['Risk Category']}</td>
-            <td style="padding:12px; vertical-align:top; line-height:1.5;">{e['Insight']}</td>
-            <td style="padding:12px; vertical-align:top; line-height:1.5; color:#2E7D32;"><strong>▶</strong> {e['Recommendation']}</td>
-            <td style="padding:12px; vertical-align:top; line-height:1.5; color:#1976D2;"><strong>✓</strong> {e['Mitigation']}</td>
-        </tr>
-        """)
-
-    table_html = f"""
-    <div class="card" style="
-        background-color: white;
-        border-left: 5px solid #003DA5;
-        padding: 20px;
-        margin-bottom: 24px;
-        border-radius: 6px;
-        box-shadow: 0 3px 8px rgba(0,0,0,0.08);
-    ">
-        <h4 style="margin-top: 0; color: #003DA5; text-align: center;">🔍 Objective 7 — Risk-Based Insight, Recommendation & Mitigation</h4>
-        <table style="width:100%; border-collapse:collapse; font-size:0.95em; margin-top:16px;">
+1. Write a concise **Insight Summary** (max 6 bullets). Use corporate tone and highlight anomalies.
+2. Generate **5 Recommended Actions**, each 1–2 sentences.
+3. Generate **5 Risk Mitigation Strategies**, each paired to each recommendation.
+
+Return output in this JSON structure ONLY:
+
+{{
+"insight_summary": "...",
+"recommendations": ["...", "...", "...", "...", "..."],
+"mitigations": ["...", "...", "...", "...", "..."]
+}}
+"""
+
+llm_output = llm_call(prompt)
+
+import json
+out = json.loads(llm_output)
+
+# ----------------------------
+#  STREAMLIT RENDERING
+# ----------------------------
+st.markdown("<h3 class='section-title'>OBJECTIVE 7 — Insight and Recommendation</h3>", unsafe_allow_html=True)
+
+# Insight card
+st.markdown(
+    f"""
+    <div style="background:#f8f9fa; border-left:4px solid #003DA5; padding:16px; border-radius:4px;">
+        <h4 style="margin:0;color:#003DA5;">Insight Summary (LLM Generated)</h4>
+        <p style="line-height:1.6;">{out['insight_summary'].replace("\n", "<br>")}</p>
+    </div>
+    """,
+    unsafe_allow_html=True
+)
+
+# Recommendations + Mitigations table
+rows = ""
+for i in range(5):
+    rows += f"""
+    <tr>
+        <td style='text-align:center; font-weight:bold;'>{i+1}</td>
+        <td style='padding:8px;'>{out['recommendations'][i]}</td>
+        <td style='padding:8px;'>{out['mitigations'][i]}</td>
+    </tr>
+    """
+
+st.markdown(
+    f"""
+    <div style="background:#e8f5e9; border-left:4px solid #4CAF50; padding:16px; border-radius:4px;">
+        <h4 style="margin:0;color:#2E7D32;">Recommended Actions & Agentic Risk Mitigation (LLM)</h4>
+        <table style="width:100%; border-collapse:collapse; margin-top:12px;">
             <thead>
-                <tr style="background-color:#e3f2fd;">
-                    <th style="padding:12px; text-align:center; border:1px solid #ccc; font-weight:600; color:#0D47A1;">Risk Category</th>
-                    <th style="padding:12px; text-align:left;   border:1px solid #ccc; font-weight:600; color:#0D47A1;">Insight</th>
-                    <th style="padding:12px; text-align:left;   border:1px solid #ccc; font-weight:600; color:#2E7D32;">Recommendation</th>
-                    <th style="padding:12px; text-align:left;   border:1px solid #ccc; font-weight:600; color:#0D47A1;">Risk Mitigation Strategy</th>
+                <tr style="background:#d4efdf;">
+                    <th>#</th>
+                    <th>Recommended Action</th>
+                    <th>Risk Mitigation</th>
                 </tr>
             </thead>
             <tbody>
-                {"".join(rows)}
+                {rows}
             </tbody>
         </table>
     </div>
-    """
-    st.markdown(table_html, unsafe_allow_html=True)
-else:
-    st.info("ℹ️ No actionable insights generated. Ensure required columns exist.")
+    """,
+    unsafe_allow_html=True
+)