import numpy as np
import pandas as pd

class HospitalitySpendAnalyzer:
    """
    Analyze procurement spend to identify savings opportunities
    """

    def __init__(self, spend_data):
        self.spend_data = spend_data  # DataFrame with transactions

    def perform_spend_analysis(self):
        """
        Comprehensive spend analysis

        Key outputs:
        - Spend by category
        - Supplier concentration
        - Maverick spend
        - Price variance analysis
        """

        # Spend by category
        category_spend = self.spend_data.groupby('category').agg({
            'amount': 'sum',
            'supplier': 'nunique',
            'transaction_id': 'count'
        }).reset_index()

        category_spend.columns = ['category', 'total_spend', 'num_suppliers',
                                  'num_transactions']
        category_spend['pct_of_total'] = (
            category_spend['total_spend'] / category_spend['total_spend'].sum() * 100
        )

        # Supplier concentration (80/20 rule)
        supplier_spend = self.spend_data.groupby('supplier')['amount'].sum().sort_values(
            ascending=False
        )

        cumulative_pct = supplier_spend.cumsum() / supplier_spend.sum() * 100
        top_suppliers = cumulative_pct[cumulative_pct <= 80].index

        # Pareto analysis
        pareto = {
            'top_20_suppliers': len(top_suppliers),
            'top_20_spend_pct': supplier_spend.loc[top_suppliers].sum() / supplier_spend.sum() * 100,
            'total_suppliers': len(supplier_spend),
            'concentration_ratio': len(top_suppliers) / len(supplier_spend)
        }

        # Price variance (for commodities)
        price_variance = self.calculate_price_variance()

        return {
            'category_spend': category_spend,
            'pareto_analysis': pareto,
            'price_variance': price_variance,
            'total_spend': self.spend_data['amount'].sum()
        }

    def calculate_price_variance(self):
        """
        Identify price discrepancies across suppliers and locations

        Opportunities for standardization and negotiation
        """

        # Calculate unit prices where applicable
        items_with_prices = self.spend_data[
            self.spend_data['quantity'].notna() & (self.spend_data['quantity'] > 0)
        ].copy()

        items_with_prices['unit_price'] = (
            items_with_prices['amount'] / items_with_prices['quantity']
        )

        # Variance by item
        variance_analysis = items_with_prices.groupby('item_description').agg({
            'unit_price': ['mean', 'std', 'min', 'max', 'count']
        }).reset_index()

        variance_analysis.columns = ['item', 'avg_price', 'std_price',
                                     'min_price', 'max_price', 'transactions']

        # Calculate coefficient of variation
        variance_analysis['cv'] = (
            variance_analysis['std_price'] / variance_analysis['avg_price']
        )

        # Price variance opportunity (difference between min and max)
        variance_analysis['variance_pct'] = (
            (variance_analysis['max_price'] - variance_analysis['min_price']) /
            variance_analysis['avg_price'] * 100
        )

        # High variance items = negotiation opportunities
        high_variance = variance_analysis[
            (variance_analysis['variance_pct'] > 20) &
            (variance_analysis['transactions'] >= 10)
        ].sort_values('variance_pct', ascending=False)

        return high_variance

    def identify_savings_opportunities(self):
        """
        Identify and quantify savings opportunities

        Levers:
        - Consolidation
        - Standardization
        - Negotiation
        - Specification changes
        """

        analysis = self.perform_spend_analysis()

        opportunities = []

        # Supplier consolidation
        category_spend = analysis['category_spend']

        for _, cat in category_spend.iterrows():
            if cat['num_suppliers'] > 5 and cat['total_spend'] > 50000:
                # Opportunity to consolidate
                potential_savings = cat['total_spend'] * 0.08  # 8% savings estimate

                opportunities.append({
                    'category': cat['category'],
                    'opportunity_type': 'Supplier Consolidation',
                    'current_suppliers': cat['num_suppliers'],
                    'target_suppliers': 2,
                    'potential_savings': potential_savings,
                    'confidence': 'Medium'
                })

        # Price standardization
        price_variance = analysis['price_variance']

        for _, item in price_variance.head(20).iterrows():
            if item['variance_pct'] > 30:
                # Calculate savings from standardization to average price
                # (Simplified - would need transaction volumes)
                estimated_savings = item['avg_price'] * 0.15 * item['transactions']

                opportunities.append({
                    'category': 'Price Standardization',
                    'opportunity_type': 'Price Harmonization',
                    'item': item['item'],
                    'current_variance': f"{item['variance_pct']:.1f}%",
                    'potential_savings': estimated_savings,
                    'confidence': 'High'
                })

        return pd.DataFrame(opportunities)

---

class SupplierPerformanceManager:
    """
    Track and manage supplier performance across key metrics
    """

    def __init__(self, suppliers):
        self.suppliers = suppliers

    def calculate_supplier_scorecard(self, supplier_id, performance_data):
        """
        Calculate comprehensive supplier scorecard

        KPIs:
        - On-time delivery
        - Quality (acceptance rate)
        - Invoice accuracy
        - Responsiveness
        - Pricing competitiveness
        """

        metrics = {}

        # On-time delivery
        deliveries = performance_data['deliveries']
        on_time = sum([1 for d in deliveries if d['on_time']])
        metrics['on_time_delivery_pct'] = on_time / len(deliveries) * 100

        # Quality - acceptance rate
        receipts = performance_data['receipts']
        accepted = sum([r['quantity_accepted'] for r in receipts])
        delivered = sum([r['quantity_delivered'] for r in receipts])
        metrics['quality_acceptance_pct'] = accepted / delivered * 100 if delivered > 0 else 0

        # Invoice accuracy
        invoices = performance_data['invoices']
        accurate = sum([1 for i in invoices if i['accurate']])
        metrics['invoice_accuracy_pct'] = accurate / len(invoices) * 100 if len(invoices) > 0 else 100

        # Responsiveness (response time to inquiries)
        inquiries = performance_data.get('inquiries', [])
        if inquiries:
            avg_response_hours = np.mean([i['response_time_hours'] for i in inquiries])
            metrics['avg_response_hours'] = avg_response_hours
            # Score: < 4 hours = 100, 4-24 = 80, > 24 = 50
            if avg_response_hours < 4:
                metrics['responsiveness_score'] = 100
            elif avg_response_hours < 24:
                metrics['responsiveness_score'] = 80
            else:
                metrics['responsiveness_score'] = 50
        else:
            metrics['responsiveness_score'] = 100

        # Pricing competitiveness
        price_index = performance_data.get('price_vs_market', 1.0)
        # 1.0 = at market, < 1.0 = below market (better)
        metrics['price_index'] = price_index
        if price_index < 0.95:
            metrics['price_score'] = 100
        elif price_index < 1.05:
            metrics['price_score'] = 90
        else:
            metrics['price_score'] = 70

        # Overall score (weighted)
        weights = {
            'on_time_delivery_pct': 0.25,
            'quality_acceptance_pct': 0.30,
            'invoice_accuracy_pct': 0.15,
            'responsiveness_score': 0.10,
            'price_score': 0.20
        }

        overall_score = sum([
            metrics.get(key, 100) * weight
            for key, weight in weights.items()
        ])

        metrics['overall_score'] = overall_score

        # Performance tier
        if overall_score >= 90:
            tier = 'Preferred'
        elif overall_score >= 75:
            tier = 'Approved'
        elif overall_score >= 60:
            tier = 'Conditional'
        else:
            tier = 'Review Required'

        metrics['performance_tier'] = tier

        return metrics

    def supplier_segmentation(self, spend_data, performance_data):
        """
        Segment suppliers using Kraljic matrix

        Dimensions:
        - Spend/value (high/low)
        - Supply risk (high/low)

        Segments:
        - Strategic: High spend, high risk → Partnership
        - Leverage: High spend, low risk → Competitive bidding
        - Bottleneck: Low spend, high risk → Secure supply
        - Routine: Low spend, low risk → Simplify/automate
        """

        supplier_segments = {}

        for supplier_id, spend in spend_data.items():
            risk_score = performance_data.get(supplier_id, {}).get('supply_risk', 50)

            # Determine segment
            high_spend = spend > 100000
            high_risk = risk_score > 60

            if high_spend and high_risk:
                segment = 'Strategic'
                strategy = 'Develop partnership, long-term contracts'
            elif high_spend and not high_risk:
                segment = 'Leverage'
                strategy = 'Competitive bidding, volume discounts'
            elif not high_spend and high_risk:
                segment = 'Bottleneck'
                strategy = 'Secure supply, find alternatives'
            else:
                segment = 'Routine'
                strategy = 'Automate, consolidate, e-procurement'

            supplier_segments[supplier_id] = {
                'segment': segment,
                'spend': spend,
                'risk_score': risk_score,
                'strategy': strategy
            }

        return supplier_segments

---

def evaluate_gpo_membership(current_spend, gpo_contracts, admin_fee_pct=0.03):
    """
    Evaluate value of GPO membership vs. direct negotiation

    Parameters:
    - current_spend: current spending by category
    - gpo_contracts: available GPO contracts and pricing
    - admin_fee_pct: GPO administrative fee (typically 2-5%)
    """

    results = []

    for category, spend in current_spend.items():
        # Current situation
        current_price_index = 1.0  # baseline

        # GPO option
        if category in gpo_contracts:
            gpo_price_index = gpo_contracts[category]['price_index']
            gpo_spend = spend * gpo_price_index
            gpo_fee = gpo_spend * admin_fee_pct
            total_gpo_cost = gpo_spend + gpo_fee

            savings = spend - total_gpo_cost
            savings_pct = savings / spend * 100

            results.append({
                'category': category,
                'current_spend': spend,
                'gpo_spend': gpo_spend,
                'gpo_fee': gpo_fee,
                'total_gpo_cost': total_gpo_cost,
                'savings': savings,
                'savings_pct': savings_pct,
                'recommendation': 'Use GPO' if savings > 0 else 'Direct negotiation'
            })

    results_df = pd.DataFrame(results)

    return {
        'total_current_spend': sum(current_spend.values()),
        'total_gpo_spend': results_df['total_gpo_cost'].sum(),
        'total_savings': results_df['savings'].sum(),
        'savings_pct': results_df['savings'].sum() / sum(current_spend.values()) * 100,
        'category_analysis': results_df
    }

---

class SustainableProcurementManager:
    """
    Manage sustainability in procurement decisions
    """

    def __init__(self, sustainability_goals):
        self.goals = sustainability_goals

    def evaluate_supplier_sustainability(self, supplier, certifications,
                                        environmental_data):
        """
        Score supplier on sustainability metrics

        Criteria:
        - Certifications (organic, Fair Trade, sustainable seafood, etc.)
        - Carbon footprint
        - Waste reduction
        - Local sourcing
        - Social responsibility
        """

        score = {}

        # Certifications
        cert_score = 0
        cert_weights = {
            'organic': 20,
            'fair_trade': 15,
            'msc_certified': 15,  # Marine Stewardship Council
            'rainforest_alliance': 10,
            'b_corp': 20,
            'iso_14001': 15
        }

        for cert, points in cert_weights.items():
            if cert in certifications:
                cert_score += points

        score['certification_score'] = min(cert_score, 100)

        # Carbon footprint
        carbon_emissions = environmental_data.get('carbon_kg_per_unit', 0)
        industry_avg = environmental_data.get('industry_avg_carbon', 10)

        if carbon_emissions < industry_avg * 0.7:
            score['carbon_score'] = 100
        elif carbon_emissions < industry_avg:
            score['carbon_score'] = 80
        elif carbon_emissions < industry_avg * 1.3:
            score['carbon_score'] = 60
        else:
            score['carbon_score'] = 40

        # Local sourcing (miles from property)
        distance = environmental_data.get('distance_miles', 1000)

        if distance < 50:
            score['local_score'] = 100
        elif distance < 150:
            score['local_score'] = 80
        elif distance < 500:
            score['local_score'] = 60
        else:
            score['local_score'] = 40

        # Social responsibility
        social_score = environmental_data.get('social_responsibility_score', 70)
        score['social_score'] = social_score

        # Waste reduction practices
        waste_diversion_pct = environmental_data.get('waste_diversion_pct', 50)
        score['waste_score'] = min(waste_diversion_pct, 100)

        # Overall sustainability score (weighted)
        overall = (
            score['certification_score'] * 0.25 +
            score['carbon_score'] * 0.25 +
            score['local_score'] * 0.15 +
            score['social_score'] * 0.20 +
            score['waste_score'] * 0.15
        )

        score['overall_sustainability_score'] = overall

        # Tier
        if overall >= 80:
            tier = 'Sustainability Leader'
        elif overall >= 65:
            tier = 'Sustainable'
        elif overall >= 50:
            tier = 'Developing'
        else:
            tier = 'Needs Improvement'

        score['sustainability_tier'] = tier

        return score

    def calculate_sustainable_procurement_pct(self, spend_by_supplier,
                                             supplier_sustainability):
        """
        Calculate percentage of spend with sustainable suppliers
        """

        total_spend = sum(spend_by_supplier.values())
        sustainable_spend = 0

        for supplier, spend in spend_by_supplier.items():
            sustainability_score = supplier_sustainability.get(supplier, {})

            if sustainability_score.get('overall_sustainability_score', 0) >= 65:
                sustainable_spend += spend

        sustainable_pct = sustainable_spend / total_spend * 100 if total_spend > 0 else 0

        return {
            'total_spend': total_spend,
            'sustainable_spend': sustainable_spend,
            'sustainable_pct': sustainable_pct,
            'target_pct': self.goals.get('sustainable_spend_target', 50),
            'gap_to_target': self.goals.get('sustainable_spend_target', 50) - sustainable_pct
        }

def calculate_p2p_automation_roi(current_process_metrics, automation_costs,
                                transaction_volumes):
    """
    Calculate ROI of procurement automation

    Benefits:
    - Reduced manual processing
    - Fewer errors
    - Better compliance
    - Spend visibility
    - Faster cycle times
    """

    # Current state costs
    current_costs = {
        'manual_po_processing': (
            transaction_volumes['po_count'] *
            current_process_metrics['minutes_per_po'] / 60 *
            current_process_metrics['hourly_cost']
        ),
        'invoice_processing': (
            transaction_volumes['invoice_count'] *
            current_process_metrics['minutes_per_invoice'] / 60 *
            current_process_metrics['hourly_cost']
        ),
        'supplier_inquiries': (
            transaction_volumes['supplier_inquiries'] *
            current_process_metrics['minutes_per_inquiry'] / 60 *
            current_process_metrics['hourly_cost']
        ),
        'maverick_spend_cost': (
            current_process_metrics['maverick_spend_pct'] *
            transaction_volumes['total_spend'] *
            0.15  # 15% premium on maverick spend
        )
    }

    total_current_cost = sum(current_costs.values())

    # Future state with automation
    automation_efficiency = {
        'manual_po_processing': 0.70,  # 70% reduction
        'invoice_processing': 0.80,    # 80% reduction (3-way match)
        'supplier_inquiries': 0.60,    # 60% reduction (self-service)
        'maverick_spend_cost': 0.50    # 50% reduction (controlled procurement)
    }

    future_costs = {
        category: cost * (1 - automation_efficiency[category])
        for category, cost in current_costs.items()
    }

    total_future_cost = sum(future_costs.values())

    # Annual savings
    annual_savings = total_current_cost - total_future_cost

    # Implementation costs
    implementation_cost = automation_costs['software_license'] + \
                         automation_costs['implementation_fee'] + \
                         automation_costs['training']

    # Ongoing costs
    annual_ongoing_cost = automation_costs['annual_support'] + \
                         automation_costs['hosting']

    # Net annual benefit
    net_annual_benefit = annual_savings - annual_ongoing_cost

    # Payback period
    payback_months = implementation_cost / (net_annual_benefit / 12)

    # 3-year ROI
    three_year_benefit = net_annual_benefit * 3 - implementation_cost
    three_year_roi = three_year_benefit / implementation_cost * 100

    return {
        'current_annual_cost': total_current_cost,
        'future_annual_cost': total_future_cost,
        'annual_savings': annual_savings,
        'implementation_cost': implementation_cost,
        'annual_ongoing_cost': annual_ongoing_cost,
        'net_annual_benefit': net_annual_benefit,
        'payback_months': payback_months,
        'three_year_roi_pct': three_year_roi,
        'savings_breakdown': {
            category: current_costs[category] - future_costs[category]
            for category in current_costs
        }
    }

---