┌─────────────────────────────────────────────────────────────────────────────┐
│                    COMPLAINT MANAGEMENT PROCESS                           │
├─────────────────────────────────────────────────────────────────────────────┤
│                                                                             │
│  ┌──────────────────────────────────────────────────────────────────────┐   │
│  │                    COMPLAINT RECEIPT                               │   │
│  │  (Channel: phone, email, app, branch, regulator)                   │   │
│  └──────────────────────────────────────────────────────────────────────┘   │
│                                    │                                        │
│                                    v                                        │
│  ┌──────────────────────────────────────────────────────────────────────┐   │
│  │                    COMPLAINT REGISTRATION                           │   │
│  │  (Log, categorise, assign)                                         │   │
│  └──────────────────────────────────────────────────────────────────────┘   │
│                                    │                                        │
│                                    v                                        │
│  ┌──────────────────────────────────────────────────────────────────────┐   │
│  │                    INVESTIGATION                                    │   │
│  │  (Review, gather evidence, assess)                                 │   │
│  └──────────────────────────────────────────────────────────────────────┘   │
│                                    │                                        │
│                                    v                                        │
│  ┌──────────────────────────────────────────────────────────────────────┐   │
│  │                    RESOLUTION                                       │   │
│  │  (Decision, remedy, communication)                                 │   │
│  └──────────────────────────────────────────────────────────────────────┘   │
│                                    │                                        │
│                                    v                                        │
│  ┌──────────────────────────────────────────────────────────────────────┐   │
│  │                    CLOSURE & REPORTING                              │   │
│  │  (Close complaint, report, learnings)                              │   │
│  └──────────────────────────────────────────────────────────────────────┘   │
│                                                                             │
└─────────────────────────────────────────────────────────────────────────────┘

# ===================================================================
# MODULE 6, LESSON 4: CONSUMER PROTECTION AND FAIR LENDING
# ===================================================================

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from datetime import datetime, timedelta
import warnings
warnings.filterwarnings('ignore')

print("="*70)
print("CONSUMER PROTECTION AND FAIR LENDING IN DIGITAL BANKING")
print("="*70)

# ----------------------------------------------------------------
# PART A: FAIR LENDING TESTING SIMULATION
# ----------------------------------------------------------------

print("\n" + "-"*60)
print("PART A: Fair Lending Testing Simulation")
print("-"*60)

# Generate lending data
np.random.seed(42)
n_applications = 2000

# Define protected groups
groups = ['White', 'Black', 'Hispanic', 'Asian', 'Other']
group_weights = [0.55, 0.15, 0.15, 0.10, 0.05]

# Generate application data
lending_data = pd.DataFrame({
    'application_id': range(1, n_applications + 1),
    'group': np.random.choice(groups, n_applications, p=group_weights),
    'income': np.random.gamma(5, 20, n_applications) + 20,
    'credit_score': np.random.normal(700, 50, n_applications).clip(550, 850).astype(int),
    'dti': np.random.beta(2, 5, n_applications) * 60,
    'loan_amount': np.random.gamma(4, 50, n_applications) + 30
})

# Generate approval decisions (with slight bias)
def generate_approval(row):
    # Base approval probability
    log_odds = -4.5 + 0.04 * row['dti'] - 0.005 * row['credit_score'] + 0.01 * (row['loan_amount']/1000)
    base_prob = 1 / (1 + np.exp(-log_odds))
    
    # Add slight bias for certain groups
    if row['group'] == 'Black':
        base_prob *= 0.92
    elif row['group'] == 'Hispanic':
        base_prob *= 0.95
    elif row['group'] == 'Asian':
        base_prob *= 1.02
    
    return np.random.binomial(1, base_prob)

lending_data['approved'] = lending_data.apply(generate_approval, axis=1)

# Calculate approval rates by group
approval_rates = lending_data.groupby('group').agg({
    'approved': ['mean', 'count'],
    'credit_score': 'mean',
    'income': 'mean'
}).round(3)
approval_rates.columns = ['Approval Rate', 'Count', 'Avg Credit Score', 'Avg Income']

print("Approval Rates by Group:")
print(approval_rates)

# Calculate disparate impact
max_rate = approval_rates['Approval Rate'].max()
approval_rates['Impact Ratio'] = approval_rates['Approval Rate'] / max_rate
approval_rates['Pass 4/5 Rule'] = approval_rates['Impact Ratio'] >= 0.80

print("\nDisparate Impact Analysis:")
print(approval_rates[['Approval Rate', 'Impact Ratio', 'Pass 4/5 Rule']])

# Visualise
fig, axes = plt.subplots(1, 2, figsize=(14, 5))

# Approval Rates
ax = axes[0]
colors = ['green' if rate >= max_rate * 0.80 else 'red' for rate in approval_rates['Approval Rate']]
bars = ax.bar(approval_rates.index, approval_rates['Approval Rate'] * 100, color=colors, alpha=0.7)
ax.axhline(y=max_rate * 100 * 0.80, color='red', linestyle='--', label='4/5 Rule (80%)')
ax.set_ylabel('Approval Rate (%)')
ax.set_title('Approval Rates by Group')
ax.legend()
ax.grid(True, alpha=0.3)

for bar, rate in zip(bars, approval_rates['Approval Rate']):
    ax.text(bar.get_x() + bar.get_width()/2, bar.get_height() + 1, 
            f'{rate*100:.1f}%', ha='center', va='bottom')

# Impact Ratios
ax = axes[1]
colors = ['green' if ratio >= 0.80 else 'red' for ratio in approval_rates['Impact Ratio']]
bars = ax.bar(approval_rates.index, approval_rates['Impact Ratio'], color=colors, alpha=0.7)
ax.axhline(y=0.80, color='red', linestyle='--', label='4/5 Rule Threshold')
ax.set_ylabel('Impact Ratio')
ax.set_title('Disparate Impact Ratios')
ax.legend()
ax.grid(True, alpha=0.3)

for bar, ratio in zip(bars, approval_rates['Impact Ratio']):
    ax.text(bar.get_x() + bar.get_width()/2, bar.get_height() + 0.01, 
            f'{ratio:.3f}', ha='center', va='bottom')

plt.tight_layout()
plt.savefig('fair_lending_analysis.png', dpi=300, bbox_inches='tight')
plt.show()
print("Fair lending analysis visualisation saved as 'fair_lending_analysis.png'")

# ----------------------------------------------------------------
# PART B: COMPLAINT MANAGEMENT SIMULATION
# ----------------------------------------------------------------

print("\n" + "-"*60)
print("PART B: Complaint Management Simulation")
print("-"*60)

# Generate complaint data
np.random.seed(42)
n_complaints = 500

complaint_data = pd.DataFrame({
    'complaint_id': range(1, n_complaints + 1),
    'date': [datetime.now() - timedelta(days=np.random.randint(0, 365)) for _ in range(n_complaints)],
    'channel': np.random.choice(['App', 'Phone', 'Email', 'Branch', 'Regulator'], n_complaints,
                                p=[0.3, 0.25, 0.2, 0.15, 0.1]),
    'category': np.random.choice(['Product', 'Service', 'Transaction', 'Communication', 'Technology'], n_complaints,
                                 p=[0.25, 0.25, 0.2, 0.15, 0.15]),
    'priority': np.random.choice(['Low', 'Medium', 'High', 'Critical'], n_complaints, p=[0.2, 0.3, 0.3, 0.2]),
    'status': np.random.choice(['Open', 'In Progress', 'Resolved', 'Closed'], n_complaints, p=[0.15, 0.25, 0.35, 0.25]),
    'resolution_time': np.random.gamma(3, 2, n_complaints).clip(0.5, 20).round(1)
})

print("Complaint Data Sample:")
print(complaint_data.head())

# Complaint summary
complaint_summary = complaint_data.groupby('category').agg({
    'complaint_id': 'count',
    'resolution_time': 'mean',
    'priority': lambda x: x.value_counts().index[0]
}).round(1)
complaint_summary.columns = ['Count', 'Avg Resolution Time (days)', 'Most Common Priority']

print("\nComplaint Summary by Category:")
print(complaint_summary)

# Visualise
fig, axes = plt.subplots(2, 2, figsize=(14, 10))

# Complaint Categories
ax = axes[0, 0]
category_counts = complaint_data['category'].value_counts()
ax.bar(category_counts.index, category_counts.values, color='teal', alpha=0.7)
ax.set_xlabel('Category')
ax.set_ylabel('Count')
ax.set_title('Complaints by Category')
ax.tick_params(axis='x', rotation=45)
ax.grid(True, alpha=0.3)

# Complaint Channels
ax = axes[0, 1]
channel_counts = complaint_data['channel'].value_counts()
ax.pie(channel_counts.values, labels=channel_counts.index, autopct='%1.1f%%')
ax.set_title('Complaints by Channel')

# Priority Distribution
ax = axes[1, 0]
priority_counts = complaint_data['priority'].value_counts()
colors = {'Critical': 'red', 'High': 'orange', 'Medium': 'yellow', 'Low': 'green'}
ax.bar(priority_counts.index, priority_counts.values, 
       color=[colors.get(p, 'gray') for p in priority_counts.index], alpha=0.7)
ax.set_xlabel('Priority')
ax.set_ylabel('Count')
ax.set_title('Complaint Priority')
ax.grid(True, alpha=0.3)

# Resolution Time by Category
ax = axes[1, 1]
resolution_by_category = complaint_data.groupby('category')['resolution_time'].mean().sort_values()
ax.barh(resolution_by_category.index, resolution_by_category.values, color='blue', alpha=0.7)
ax.set_xlabel('Average Resolution Time (days)')
ax.set_title('Resolution Time by Category')
ax.grid(True, alpha=0.3)

plt.tight_layout()
plt.savefig('complaint_analysis.png', dpi=300, bbox_inches='tight')
plt.show()
print("Complaint analysis visualisation saved as 'complaint_analysis.png'")

# ----------------------------------------------------------------
# PART C: CONSUMER PROTECTION METRICS
# ----------------------------------------------------------------

print("\n" + "-"*60)
print("PART C: Consumer Protection Metrics")
print("-"*60)

consumer_metrics = pd.DataFrame({
    'Metric': [
        'Disparate Impact Ratio',
        'Fair Lending Compliance',
        'Complaint Resolution Rate',
        'Average Resolution Time',
        'Customer Complaint Rate',
        'Dispute Resolution Rate',
        'Transparency Score',
        'Customer Satisfaction (CSAT)'
    ],
    'Current Value': [
        '0.76',
        '85%',
        '75%',
        '5.2 days',
        '2.5%',
        '72%',
        '78/100',
        '72%'
    ],
    'Target Value': [
        '> 0.80',
        '100%',
        '> 95%',
        '< 3 days',
        '< 1%',
        '> 90%',
        '> 90/100',
        '> 80%'
    ],
    'Status': ['🔴', '🟡', '🟡', '🔴', '🟡', '🟡', '🟡', '🟡']
})

print("Consumer Protection Metrics:")
print(consumer_metrics.to_string(index=False))

# ----------------------------------------------------------------
# PART D: CONSUMER PROTECTION REGULATORY REQUIREMENTS
# ----------------------------------------------------------------

print("\n" + "-"*60)
print("PART D: Consumer Protection Regulatory Requirements")
print("-"*60)

regulatory_requirements = pd.DataFrame({
    'Requirement': [
        'Fair Lending Compliance',
        'Transparent Disclosures',
        'Complaint Handling',
        'Dispute Resolution',
        'Data Privacy',
        'Accessibility',
        'Financial Education',
        'Marketing Compliance'
    ],
    'Status': ['✅', '🟡', '✅', '🟡', '✅', '🟡', '🟡', '✅'],
    'Regulation': ['ECOA', 'UDAAP', 'CFPB', 'FCRA', 'GDPR', 'ADA', 'Various', 'UDAAP'],
    'Owner': ['Compliance', 'Compliance', 'Compliance', 'Compliance', 'Privacy', 'Compliance', 'Compliance', 'Marketing']
})

print("Consumer Protection Regulatory Requirements:")
print(regulatory_requirements.to_string(index=False))

# ----------------------------------------------------------------
# PART E: CONSUMER PROTECTION ROADMAP
# ----------------------------------------------------------------

print("\n" + "-"*60)
print("PART E: Consumer Protection Roadmap")
print("-"*60)

roadmap = {
    "Phase 1 (0-6 months) – Foundation": {
        "Focus": "Establish consumer protection foundation.",
        "Activities": [
            "Implement fair lending testing framework.",
            "Establish complaint management process.",
            "Develop transparent disclosures.",
            "Train employees on consumer protection."
        ],
        "Success Metrics": ["Disparate impact ratio > 0.80", "Complaint resolution rate > 80%"]
    },
    "Phase 2 (6-12 months) – Scale": {
        "Focus": "Scale consumer protection capabilities.",
        "Activities": [
            "Implement automated fair lending monitoring.",
            "Enhance complaint management system.",
            "Improve dispute resolution process.",
            "Implement accessibility standards."
        ],
        "Success Metrics": ["Disparate impact ratio > 0.85", "Complaint resolution rate > 90%"]
    },
    "Phase 3 (12-24 months) – Advanced": {
        "Focus": "Advanced consumer protection.",
        "Activities": [
            "Implement AI-powered fair lending monitoring.",
            "Deploy predictive complaint analytics.",
            "Build consumer protection dashboards.",
            "Establish consumer protection culture."
        ],
        "Success Metrics": ["Disparate impact ratio > 0.90", "Complaint resolution rate > 95%"]
    },
    "Phase 4 (24+ months) – Leadership": {
        "Focus": "Industry-leading consumer protection.",
        "Activities": [
            "Implement autonomous consumer protection.",
            "Build predictive consumer insights.",
            "Achieve industry leadership.",
            "Establish consumer-centric culture."
        ],
        "Success Metrics": ["Industry-leading consumer protection", "Continuous improvement"]
    }
}

for phase, details in roadmap.items():
    print(f"\n{phase}:")
    print(f"  Focus: {details['Focus']}")
    print("  Activities:")
    for activity in details['Activities']:
        print(f"    • {activity}")
    print("  Success Metrics:")
    for metric in details['Success Metrics']:
        print(f"    • {metric}")

# ----------------------------------------------------------------
# PART F: CONSUMER PROTECTION CHECKLIST
# ----------------------------------------------------------------

print("\n" + "-"*60)
print("PART F: Consumer Protection Checklist")
print("-"*60)

consumer_checklist = [
    "✅ Fair lending policy in place.",
    "✅ Disparate impact testing conducted.",
    "✅ Complaint management process documented.",
    "✅ Dispute resolution process in place.",
    "✅ Transparent disclosures for all products.",
    "✅ Data privacy and consent management.",
    "✅ Accessibility standards implemented.",
    "✅ Customer education materials available.",
    "✅ Marketing compliance reviewed.",
    "✅ Consumer protection training provided."
]

print("Consumer Protection Checklist:")
for item in consumer_checklist:
    print(f"  {item}")

# ----------------------------------------------------------------
# PART G: SUMMARY AND RECOMMENDATIONS
# ----------------------------------------------------------------

print("\n" + "="*70)
print("PART G: Summary and Recommendations")
print("="*70)

print("""
Consumer Protection and Fair Lending – Key Takeaways:

1. Consumer protection is critical for trust and regulatory compliance.
2. Key regulations: ECOA, FCRA, UDAAP, CFPB, Consumer Credit Directive.
3. Fair lending: ECOA compliance, disparate impact testing, 4/5 rule.
4. Complaint management: receipt → registration → investigation → resolution → closure.
5. Fair lending testing: identify policy, collect data, calculate impact, apply 4/5 rule.
6. Key metrics: disparate impact ratio, complaint resolution rate, customer satisfaction.
7. Roadmap: foundation → scale → advanced → leadership.

Recommendations:
  - Implement fair lending testing framework.
  - Establish complaint management process.
  - Provide transparent disclosures.
  - Train employees on consumer protection.
  - Monitor consumer protection metrics.
  - Continuously improve consumer protection practices.
""")

print("="*70)
print("END OF LESSON 4 – MODULE 6")
print("="*70)