Analisis Defect & Root Cause Analysis

# ================================================================
# S11409 - Sesi 6: Analisis Defect & Visualisasi Pareto
# Dosen: Riadi Marta Dinata, S.Ti., M.Kom. | ISTN Jakarta
# ================================================================

import pandas as pd
import matplotlib.pyplot as plt
import matplotlib.ticker as mticker
import numpy as np
from collections import Counter

# ── DATASET SIMULASI DEFECT ──
data_defect = {
    'id': [f'BUG-{i:03d}' for i in range(1, 121)],
    'sprint': np.random.choice(['S1','S2','S3','S4','S5','S6'], 120,
                p=[0.10,0.15,0.22,0.20,0.18,0.15]),
    'kategori': np.random.choice([
        'Validasi Input', 'Error Handling', 'Race Condition',
        'UI Tampilan', 'Logika Bisnis', 'Lain-lain'
    ], 120, p=[0.38,0.25,0.16,0.11,0.07,0.03]),
    'severity': np.random.choice(['Critical','High','Medium','Low'],
                  120, p=[0.10,0.25,0.40,0.25]),
    'waktu_fix': np.random.exponential(scale=3, size=120).round(1)  # jam
}
df = pd.DataFrame(data_defect)

# ── ANALISIS DASAR ──
print("="*55)
print("    LAPORAN ANALISIS DEFECT — PROYEK SIAKAD")
print("="*55)
print(f"Total defect        : {len(df)}")
print(f"Rata-rata waktu fix : {df['waktu_fix'].mean():.1f} jam")
print(f"Defect Critical     : {(df['severity']=='Critical').sum()}")
print()

# ── PARETO ANALYSIS ──
kategori_count = df['kategori'].value_counts().sort_values(ascending=False)
total = kategori_count.sum()
persen = (kategori_count / total * 100).round(1)
kumulatif = persen.cumsum()

print("--- Pareto Analysis ---")
for kat, cnt in kategori_count.items():
    print(f"  {kat:<22} : {cnt:3d} ({cnt/total*100:5.1f}%)")
print()

# ── DEFECT ARRIVAL/CLOSURE CURVE ──
arrival_by_sprint = df.groupby('sprint').size()

# Simulasi closure (biasanya lebih lambat dari arrival)
closure_by_sprint = (arrival_by_sprint * np.array([0.7,0.8,0.85,0.9,0.95,1.0])).astype(int)

# ── VISUALISASI (3 plot) ──
fig = plt.figure(figsize=(16, 5))
fig.patch.set_facecolor('#0A0608')

# Plot 1: Pareto Chart
ax1 = fig.add_subplot(1, 3, 1)
ax1.set_facecolor('#160C0E')
colors = ['#C8102E','#F97316','#FCD34D','#22C55E','#38BDF8','#9B7A80']
bars = ax1.bar(kategori_count.index, kategori_count.values,
               color=colors[:len(kategori_count)], alpha=0.85)
ax1.set_title('Pareto Defect by Kategori', color='white', pad=10)
ax1.tick_params(axis='x', rotation=45, colors='#9B7A80', labelsize=8)
ax1.tick_params(axis='y', colors='#9B7A80')
ax2_twin = ax1.twinx()
ax2_twin.plot(range(len(kumulatif)), kumulatif.values,
              color='#D4A017', marker='o', linewidth=2, markersize=5)
ax2_twin.axhline(80, color='#D4A017', linestyle='--', alpha=0.5)
ax2_twin.set_ylabel('Kumulatif %', color='#D4A017')
ax2_twin.tick_params(colors='#D4A017')
ax2_twin.yaxis.set_major_formatter(mticker.PercentFormatter())

# Plot 2: Defect per Sprint (Arrival vs Closure)
ax3 = fig.add_subplot(1, 3, 2)
ax3.set_facecolor('#160C0E')
x = np.arange(len(arrival_by_sprint))
ax3.plot(x, arrival_by_sprint.values, color='#C8102E', marker='o',
         lw=2, label='Defect Masuk')
ax3.plot(x, closure_by_sprint.values, color='#22C55E', marker='s',
         lw=2, linestyle='--', label='Defect Ditutup')
ax3.fill_between(x, arrival_by_sprint.values, closure_by_sprint.values,
                 alpha=0.15, color='#C8102E', label='Backlog')
ax3.set_xticks(x); ax3.set_xticklabels(arrival_by_sprint.index)
ax3.set_title('Defect Arrival vs Closure Curve', color='white', pad=10)
ax3.legend(fontsize=8); ax3.grid(alpha=0.15)
ax3.tick_params(colors='#9B7A80')

# Plot 3: Severity Distribution
ax4 = fig.add_subplot(1, 3, 3)
ax4.set_facecolor('#160C0E')
sev_count = df['severity'].value_counts()
sev_colors = ['#F87171','#F97316','#FCD34D','#22C55E']
wedges, texts, autotexts = ax4.pie(sev_count.values,
    labels=sev_count.index, colors=sev_colors,
    autopct='%1.1f%%', startangle=90, pctdistance=0.85)
for t in texts: t.set_color('#9B7A80'); t.set_fontsize(9)
for at in autotexts: at.set_color('white'); at.set_fontsize(9)
ax4.set_title('Distribusi Severity Defect', color='white', pad=10)

plt.tight_layout()
plt.savefig('analisis_defect.png', dpi=150, facecolor='#0A0608', bbox_inches='tight')
print("✅ Grafik tersimpan: analisis_defect.png")
print("\n💡 Rekomendasi perbaikan berdasarkan Pareto:")
top3 = kategori_count.head(3)
for kat, cnt in top3.items():
    print(f"  → Fokus pada '{kat}' ({cnt} bug, {cnt/total*100:.0f}%)")