Upload 9 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+static/images/sensitivity_analysis_tornado.png filter=lfs diff=lfs merge=lfs -text

app.py

@@ -0,0 +1,155 @@
+from flask import Flask, render_template, request, redirect, url_for, send_file, flash, jsonify
+import os
+import traceback
+from pathlib import Path
+import json
+import pandas as pd
+
+app = Flask(__name__)
+app.secret_key = os.urandom(24)
+
+try:
+    import main as project_main
+except Exception as e:
+    project_main = None
+    import_error = traceback.format_exc()
+else:
+    import_error = None
+
+RESULTS_CSV = Path("results_flask.csv")
+
+def safe_update_dict_from_json(orig_dict, json_text):
+    if not json_text or not json_text.strip():
+        return orig_dict
+    try:
+        new = json.loads(json_text)
+        if not isinstance(new, dict):
+            return orig_dict
+        d = orig_dict.copy()
+        d.update(new)
+        return d
+    except Exception:
+        return orig_dict
+
+@app.route("/", methods=["GET"])
+def index():
+    if project_main is None:
+        return render_template("index.html", import_error=import_error, project=None)
+    techs = list(project_main.TECHNOLOGY_DATA.keys())
+    defaults = {
+        "inflation": project_main.INFLATION_RATE,
+        "tax": project_main.TAX_RATE,
+        "years": project_main.PROJECT_YEARS,
+        "cap_min": project_main.OPTIMIZATION_SPACE['capacity_kta'][0],
+        "cap_max": project_main.OPTIMIZATION_SPACE['capacity_kta'][1],
+        "export_mix": project_main.OPTIMIZATION_SPACE['export_market_mix'][0],
+    }
+    return render_template("index.html", import_error=None, project=project_main, techs=techs, defaults=defaults)
+
+@app.route("/run", methods=["POST"])
+def run():
+    if project_main is None:
+        flash("خطا: فایل main.py بارگزاری نشد. لطفاً لاگ را بررسی کنید.", "danger")
+        return redirect(url_for("index"))
+    try:
+        inflation_rate = float(request.form.get("inflation", project_main.INFLATION_RATE))
+        tax_rate = float(request.form.get("tax", project_main.TAX_RATE))
+        project_years = int(request.form.get("years", project_main.PROJECT_YEARS))
+        capacity_min = float(request.form.get("cap_min", project_main.OPTIMIZATION_SPACE['capacity_kta'][0]))
+        capacity_max = float(request.form.get("cap_max", project_main.OPTIMIZATION_SPACE['capacity_kta'][1]))
+        technology = request.form.get("technology", list(project_main.TECHNOLOGY_DATA.keys())[0])
+        export_mix = float(request.form.get("export_mix", project_main.OPTIMIZATION_SPACE['export_market_mix'][0]))
+        sell_byproducts = request.form.get("sell_byproducts") == "on"
+        tech_json = request.form.get("tech_json", "")
+        prices_json = request.form.get("prices_json", "")
+    except Exception as e:
+        flash("خطا در خواندن ورودی‌ها: " + str(e), "danger")
+        return redirect(url_for("index"))
+
+    try:
+        project_main.INFLATION_RATE = inflation_rate
+        project_main.TAX_RATE = tax_rate
+        project_main.PROJECT_YEARS = project_years
+
+        project_main.OPTIMIZATION_SPACE['capacity_kta'] = (capacity_min, capacity_max)
+        if technology not in project_main.OPTIMIZATION_SPACE.get('technology', []):
+            project_main.OPTIMIZATION_SPACE['technology'] = [technology]
+        low = max(0.0, export_mix - 0.05)
+        high = min(1.0, export_mix + 0.05)
+        if low == high:
+            high = min(1.0, high + 0.01)
+        project_main.OPTIMIZATION_SPACE['export_market_mix'] = (low, high)
+        project_main.OPTIMIZATION_SPACE['sell_byproducts'] = [bool(sell_byproducts)]
+
+        project_main.TECHNOLOGY_DATA = safe_update_dict_from_json(project_main.TECHNOLOGY_DATA, tech_json)
+        project_main.PRODUCT_PRICES_USD_PER_TON = safe_update_dict_from_json(project_main.PRODUCT_PRICES_USD_PER_TON, prices_json)
+    except Exception as e:
+        flash("خطا در اعمال پارامترها: " + str(e), "danger")
+        return redirect(url_for("index"))
+
+    try:
+        flash("محاسبات شروع شد — صبر کنید تا عملیات به پایان برسد...", "info")
+        results = project_main.run_optimizations_without_ml()
+        df_results = pd.DataFrame(results).sort_values(by="irr", ascending=False).reset_index(drop=True)
+        df_results = df_results.round(2)
+
+        df_results.to_csv(RESULTS_CSV, index=False, encoding='utf-8-sig')
+
+        try:
+            project_main.display_and_save_results(df_results)
+            project_main.create_kpi_comparison_dashboard(df_results)
+        except Exception:
+            pass
+
+        if not df_results.empty:
+            best = df_results.iloc[0]
+            top_kpis = {
+                "irr": round(float(best['irr']), 2),
+                "annual_profit_M": round(float(best['annual_profit'])/1_000_000, 2),
+                "capex_M": round(float(best['total_capex'])/1_000_000, 2),
+                "payback": round(float(best['payback_period']), 2)
+            }
+        else:
+            top_kpis = None
+
+        charts_data = df_results.to_dict(orient="records")
+
+        flash("محاسبات با موفقیت تکمیل شد.", "success")
+     
+        kpi_img = Path("static/images/kpi_dashboard.png") if Path("static/images/kpi_dashboard.png").exists() else None
+        tornado_img = Path("static/images/sensitivity_analysis_tornado.png") if Path("static/images/sensitivity_analysis_tornado.png").exists() else None
+
+        return render_template(
+            "index.html",
+            project=project_main,
+            techs=list(project_main.TECHNOLOGY_DATA.keys()),
+            defaults={"inflation": project_main.INFLATION_RATE, "tax": project_main.TAX_RATE},
+            table_html = df_results.to_html(
+                classes='table table-striped table-dark',
+                index=False,
+                justify='center',
+                float_format=lambda x: f"{x:.2f}"
+                ),
+            top_kpis=top_kpis,
+            charts_json=json.dumps(charts_data, default=str),
+            kpi_img="images/kpi_dashboard.png" if kpi_img else None,
+            tornado_img="images/sensitivity_analysis_tornado.png" if tornado_img else None
+        )
+
+    except Exception as e:
+        tb = traceback.format_exc()
+        flash("خطا در اجرای الگوریتم‌ها. لاگ: " + str(e), "danger")
+        return render_template("index.html", import_error=tb, project=project_main)
+
+@app.route("/download")
+def download_results():
+    if RESULTS_CSV.exists():
+        return send_file(str(RESULTS_CSV), as_attachment=True)
+    flash("فایل نتایج آماده نیست.", "warning")
+    return redirect(url_for("index"))
+
+if __name__ == "__main__":
+    app.run(debug=True)
+    
+    
+    

main.py

@@ -0,0 +1,313 @@
+import pandas as pd
+import numpy as np
+import random
+import copy
+import optuna
+import matplotlib.pyplot as plt
+import seaborn as sns
+from skopt import gp_minimize
+from skopt.space import Real, Categorical
+from skopt.utils import use_named_args
+from statsmodels.tsa.arima.model import ARIMA
+import numpy_financial as npf
+
+optuna.logging.set_verbosity(optuna.logging.WARNING)  
+
+PROJECT_YEARS = 15
+BASE_CAPACITY_KTA = 300
+INFLATION_RATE = 0.015
+TAX_RATE = 0.10
+DEPRECIATION_YEARS = 15
+
+TECHNOLOGY_DATA = {
+    "JNC": {"capex_base_M": 180.0, "opex_base_cents_kg": 150.0},
+    "Hoechst_AG": {"capex_base_M": 230.0, "opex_base_cents_kg": 155.0},
+    "BF_Goodrich": {"capex_base_M": 280.0, "opex_base_cents_kg": 160.0},
+    "Shin_Etsu_1991": {"capex_base_M": 260.0, "opex_base_cents_kg": 155.0},
+    "Shin_Etsu_2004": {"capex_base_M": 1500.0, "opex_base_cents_kg": 150.0},
+    "Vinnolit": {"capex_base_M": 240.0, "opex_base_cents_kg": 155.0},
+    "QVC_Qatar": {"capex_base_M": 200.0, "opex_base_cents_kg": 145.0},
+    "SP_Chemicals": {"capex_base_M": 250.0, "opex_base_cents_kg": 145.0},
+    "Engro_Pakistan": {"capex_base_M": 1400.0, "opex_base_cents_kg": 155.0},
+    "Formosa_BR_USA": {"capex_base_M": 380.0, "opex_base_cents_kg": 150.0},
+    "Shintech_USA_Exp": {"capex_base_M": 1700.0, "opex_base_cents_kg": 160.0},
+    "Zhongtai_China": {"capex_base_M": 1100.0, "opex_base_cents_kg": 155.0},
+    "Shintech_USA_2021": {"capex_base_M": 1700.0, "opex_base_cents_kg": 160.0},
+    "Reliance_India_2024": {"capex_base_M": 2200.0, "opex_base_cents_kg": 155.0},
+    "Orbia_Germany_2023": {"capex_base_M": 180.0, "opex_base_cents_kg": 155.0},
+    "Westlake_USA_2022": {"capex_base_M": 850.0, "opex_base_cents_kg": 160.0}
+}
+
+STRATEGY_DATA = {
+    'Integrated_Production': {'sourcing_cost_per_ton_pvc': 450.0, 'byproducts': {'caustic_soda_ton': 1.1, 'surplus_edc_ton': 0.523}},
+    'Purchase_VCM': {'sourcing_cost_per_ton_pvc': 650.0, 'byproducts': {'caustic_soda_ton': 0, 'surplus_edc_ton': 0}}
+}
+
+PRODUCT_PRICES_USD_PER_TON = {
+    'pvc_s65_export': 1100, 'pvc_s70_domestic': 950,
+    'byproduct_caustic_soda': 450, 'byproduct_surplus_edc': 170
+}
+
+OPTIMIZATION_SPACE = {
+    'capacity_kta': (500, 600),
+    'technology': ["Engro_Pakistan"], #, "Shin_Etsu_2004"],
+    'sourcing_strategy': ['Integrated_Production'],
+    'export_market_mix': (0.6, 0.8),
+    'sell_byproducts': [True]
+}
+
+def forecast_prices(base_price, years=PROJECT_YEARS, cagr=0.04):
+    historical = [base_price * (1 + cagr + random.uniform(-0.03, 0.03)) ** i for i in range(-5, 0)]
+    model = ARIMA(historical, order=(1, 1, 0))
+    fit = model.fit()
+    forecast = fit.forecast(steps=years)
+    return [p * (1 + INFLATION_RATE) ** i for i, p in enumerate(forecast)]
+
+def calculate_project_kpis(**params):
+    tech_data = TECHNOLOGY_DATA[params['technology']]
+    scaling_factor = (params['capacity_kta'] / BASE_CAPACITY_KTA) ** 0.65
+    total_capex = tech_data['capex_base_M'] * scaling_factor * 1_000_000
+    capacity_tons = params['capacity_kta'] * 1000
+
+    price_s65_forecast = forecast_prices(PRODUCT_PRICES_USD_PER_TON['pvc_s65_export'])
+    price_s70_forecast = forecast_prices(PRODUCT_PRICES_USD_PER_TON['pvc_s70_domestic'])
+    byproduct_caustic_forecast = forecast_prices(PRODUCT_PRICES_USD_PER_TON['byproduct_caustic_soda'])
+    byproduct_edc_forecast = forecast_prices(PRODUCT_PRICES_USD_PER_TON['byproduct_surplus_edc'])
+
+    cash_flows = [-total_capex]
+    depreciation_annual = total_capex / DEPRECIATION_YEARS
+
+    for year in range(1, PROJECT_YEARS + 1):
+        infl_factor = (1 + INFLATION_RATE) ** (year - 1)
+        price_s65 = price_s65_forecast[year - 1]
+        price_s70 = price_s70_forecast[year - 1] * 0.95 if params['capacity_kta'] >= 550 else price_s70_forecast[year - 1]
+        revenue_export = (capacity_tons * params['export_market_mix']) * price_s65
+        revenue_domestic = (capacity_tons * (1 - params['export_market_mix'])) * price_s70
+        pvc_revenue = revenue_export + revenue_domestic
+
+        byproduct_revenue = 0
+        if params['sourcing_strategy'] == 'Integrated_Production' and params['sell_byproducts']:
+            byproducts = STRATEGY_DATA['Integrated_Production']['byproducts']
+            byproduct_revenue += (byproducts['caustic_soda_ton'] * capacity_tons * byproduct_caustic_forecast[year - 1])
+            byproduct_revenue += (byproducts['surplus_edc_ton'] * capacity_tons * byproduct_edc_forecast[year - 1])
+
+        total_revenue = pvc_revenue + byproduct_revenue
+
+        opex_sourcing = STRATEGY_DATA['Integrated_Production']['sourcing_cost_per_ton_pvc'] * capacity_tons * infl_factor
+        opex_base = (tech_data['opex_base_cents_kg'] / 100) * capacity_tons * infl_factor
+        total_opex = opex_sourcing + opex_base
+
+        ebitda = total_revenue - total_opex
+        taxable_income = ebitda - depreciation_annual
+        taxes = max(taxable_income * TAX_RATE, 0)
+        net_income = taxable_income - taxes
+        free_cash_flow = net_income + depreciation_annual
+        cash_flows.append(free_cash_flow)
+
+    irr = npf.irr(cash_flows) * 100 if npf.irr(cash_flows) > 0 else -1.0
+    cumulative_cash_flow = np.cumsum(cash_flows)
+    payback_period_years = np.where(cumulative_cash_flow > 0)[0]
+    payback_period = payback_period_years[0] + 1 if len(payback_period_years) > 0 else float('inf')
+    annual_profit = np.mean([cf for cf in cash_flows[1:] if cf > 0])
+
+    return {"irr": irr, "annual_profit": annual_profit, "total_capex": total_capex, "payback_period": payback_period}
+
+def run_optimizations_without_ml():
+    print("\n--- Running Optimization Algorithms ---")
+    results = []
+    results.append(run_bayesian_optimization())
+    results.append(run_genetic_algorithm())
+    results.append(run_optuna_direct())
+    return results
+
+def run_genetic_algorithm():
+    print("Running Genetic Algorithm (GA)...")
+    population = [{k: random.choice(v) if isinstance(v, list) else random.uniform(*v) for k,v in OPTIMIZATION_SPACE.items()} for _ in range(50)]
+    best_overall_individual = None
+    best_overall_fitness = -float('inf')
+    for _ in range(100):
+        fitnesses = [calculate_project_kpis(**ind)['irr'] for ind in population]
+        if max(fitnesses) > best_overall_fitness:
+            best_overall_fitness = max(fitnesses)
+            best_overall_individual = population[np.argmax(fitnesses)]
+        selected = [max(random.sample(list(zip(population, fitnesses)), 5), key=lambda i: i[1])[0] for _ in range(50)]
+        next_gen = []
+        for i in range(0, 50, 2):
+            p1, p2 = selected[i], selected[i+1]
+            c1, c2 = copy.deepcopy(p1), copy.deepcopy(p2)
+            if random.random() < 0.9: c1['technology'], c2['technology'] = p2['technology'], c1['technology']
+            if random.random() < 0.2: c1['export_market_mix'] = random.uniform(*OPTIMIZATION_SPACE['export_market_mix'])
+            next_gen.extend([c1, c2])
+        population = next_gen
+    kpis = calculate_project_kpis(**best_overall_individual)
+    return {"Method": "Genetic Algorithm", **kpis, "Params": best_overall_individual}
+
+def run_bayesian_optimization():
+    print("Running Bayesian Optimization...")
+    skopt_space = [
+        Real(OPTIMIZATION_SPACE['capacity_kta'][0], OPTIMIZATION_SPACE['capacity_kta'][1], name='capacity_kta'),
+        Categorical(OPTIMIZATION_SPACE['technology'], name='technology'),
+        Categorical(OPTIMIZATION_SPACE['sourcing_strategy'], name='sourcing_strategy'),
+        Real(OPTIMIZATION_SPACE['export_market_mix'][0], OPTIMIZATION_SPACE['export_market_mix'][1], name='export_market_mix'),
+        Categorical(OPTIMIZATION_SPACE['sell_byproducts'], name='sell_byproducts')
+    ]
+    @use_named_args(skopt_space)
+    def objective(**params):
+        return -calculate_project_kpis(**params)['irr']
+    res = gp_minimize(objective, skopt_space, n_calls=100, random_state=42, n_initial_points=20)
+    best_params = {space.name: val for space, val in zip(skopt_space, res.x)}
+    kpis = calculate_project_kpis(**best_params)
+    return {"Method": "Bayesian Opt", **kpis, "Params": best_params}
+
+def run_optuna_direct():
+    print("Running Optuna (TPE)...")
+    def objective(trial):
+        params = {
+            "capacity_kta": trial.suggest_float("capacity_kta", *OPTIMIZATION_SPACE['capacity_kta']),
+            "technology": trial.suggest_categorical("technology", OPTIMIZATION_SPACE['technology']),
+            "sourcing_strategy": trial.suggest_categorical("sourcing_strategy", OPTIMIZATION_SPACE['sourcing_strategy']),
+            "export_market_mix": trial.suggest_float("export_market_mix", *OPTIMIZATION_SPACE['export_market_mix']),
+            "sell_byproducts": trial.suggest_categorical("sell_byproducts", OPTIMIZATION_SPACE['sell_byproducts'])
+        }
+        return calculate_project_kpis(**params)['irr']
+    study = optuna.create_study(direction="maximize")
+    study.optimize(objective, n_trials=200, n_jobs=-1)
+    kpis = calculate_project_kpis(**study.best_params)
+    return {"Method": "Optuna (TPE - Direct)", **kpis, "Params": study.best_params}
+
+def display_and_save_results(df_results):
+    print("\n--- Final Results and Comparison ---")
+    df_display = pd.DataFrame()
+    df_display['Method'] = df_results['Method']
+    df_display['Optimal IRR (%)'] = df_results['irr'].map('{:,.2f}%'.format)
+    df_display['Annual Profit ($M)'] = (df_results['annual_profit'] / 1_000_000).map('{:,.1f}'.format)
+    df_display['CAPEX ($M)'] = (df_results['total_capex'] / 1_000_000).map('{:,.1f}'.format)
+    df_display['Payback (Yrs)'] = df_results['payback_period'].map('{:,.1f}'.format)
+    param_df = pd.DataFrame(df_results['Params'].tolist())
+    param_df['capacity_kta'] = param_df['capacity_kta'].round(1)
+    param_df['export_market_mix'] = (param_df['export_market_mix'] * 100).round(1).astype(str) + '%'
+    df_display = pd.concat([df_display, param_df.rename(columns={
+        'capacity_kta': 'Capacity (KTA)', 'technology': 'Technology', 'sourcing_strategy': 'Sourcing',
+        'export_market_mix': 'Export Mix', 'sell_byproducts': 'Sell Byproducts'
+    })], axis=1)
+
+    print("\n✅ **Final Comparison of Optimal Scenarios (Sorted by Best IRR)**")
+    print("="*120)
+    print(df_display.to_string(index=False))
+    print("="*120)
+    df_display.to_csv("results.csv", index=False, encoding='utf-8-sig')
+
+def create_kpi_comparison_dashboard(df_results):
+    print("\n--- Generating KPI Comparison Dashboard ---")
+    df_plot = df_results.sort_values(by='irr', ascending=False)
+    df_plot['annual_profit_M'] = df_plot['annual_profit'] / 1_000_000
+    df_plot['total_capex_M'] = df_plot['total_capex'] / 1_000_000
+
+    fig, axes = plt.subplots(2, 2, figsize=(20, 14))
+    fig.suptitle('Dashboard: Comprehensive Comparison of Optimization Methods', fontsize=22, weight='bold')
+    palettes = ['viridis', 'plasma', 'magma', 'cividis']
+    metrics = [
+        ('irr', 'Optimal IRR (%)', axes[0, 0]),
+        ('annual_profit_M', 'Annual Profit ($M)', axes[0, 1]),
+        ('total_capex_M', 'Total CAPEX ($M)', axes[1, 0]),
+        ('payback_period', 'Payback Period (Years)', axes[1, 1])
+    ]
+
+    for i, (metric, title, ax) in enumerate(metrics):
+        sns.barplot(x=metric, y='Method', data=df_plot, ax=ax, palette=palettes[i])
+        ax.set_title(title, fontsize=16, weight='bold')
+        ax.set_xlabel('')
+        ax.set_ylabel('')
+        for p in ax.patches:
+            width = p.get_width()
+            ax.text(width * 1.01, p.get_y() + p.get_height() / 2,
+                    f'{width:,.2f}',
+                    va='center', fontsize=11)
+
+    plt.tight_layout(rect=[0, 0.03, 1, 0.95])
+    plt.savefig("static/images/kpi_dashboard.png", bbox_inches='tight')
+    print("\n✅ A KPI dashboard graph has been saved as 'kpi_dashboard.png'")
+
+def run_sensitivity_analysis(best_params, base_irr):
+    global PRODUCT_PRICES_USD_PER_TON, STRATEGY_DATA
+    print("\n--- Sensitivity Analysis on Best Scenario ---")
+    print(f"Analyzing sensitivity around the base IRR of {base_irr:,.2f}%")
+
+    sensitivity_vars = {
+        'Byproduct Caustic Soda Price': ('price', 'byproduct_caustic_soda'),
+        'Sourcing Cost Integrated': ('strategy', 'Integrated_Production', 'sourcing_cost_per_ton_pvc'),
+        'Domestic PVC S70 Price': ('price', 'pvc_s70_domestic')
+    }
+    variations = [-0.20, -0.10, 0.10, 0.20]
+    results = []
+    original_prices = copy.deepcopy(PRODUCT_PRICES_USD_PER_TON)
+    original_strategies = copy.deepcopy(STRATEGY_DATA)
+
+    for key, path in sensitivity_vars.items():
+        for var in variations:
+            PRODUCT_PRICES_USD_PER_TON = copy.deepcopy(original_prices)
+            STRATEGY_DATA = copy.deepcopy(original_strategies)
+            if path[0] == 'price':
+                base_value = PRODUCT_PRICES_USD_PER_TON[path[1]]
+                PRODUCT_PRICES_USD_PER_TON[path[1]] = base_value * (1 + var)
+            else:
+                base_value = STRATEGY_DATA[path[1]][path[2]]
+                STRATEGY_DATA[path[1]][path[2]] = base_value * (1 + var)
+
+            kpis = calculate_project_kpis(**best_params)
+            results.append({
+                'Variable': key, 'Change': f'{var:+.0%}',
+                'New IRR (%)': kpis['irr'], 'IRR Delta (%)': kpis['irr'] - base_irr
+            })
+
+    PRODUCT_PRICES_USD_PER_TON = original_prices
+    STRATEGY_DATA = original_strategies
+
+    df_sensitivity = pd.DataFrame(results)
+    print("\n✅ **Sensitivity Analysis Results**")
+    print("="*60)
+    print(df_sensitivity.to_string(index=False))
+    print("="*60)
+
+    tornado_data = []
+    for var_name in df_sensitivity['Variable'].unique():
+        subset = df_sensitivity[df_sensitivity['Variable'] == var_name]
+        min_delta = subset['IRR Delta (%)'].min()
+        max_delta = subset['IRR Delta (%)'].max()
+        tornado_data.append({
+            'Variable': var_name,
+            'Min_Delta': min_delta,
+            'Max_Delta': max_delta,
+            'Range': max_delta - min_delta
+        })
+    df_tornado = pd.DataFrame(tornado_data).sort_values('Range', ascending=True)
+
+    fig, ax = plt.subplots(figsize=(12, 8))
+    y = np.arange(len(df_tornado))
+    ax.barh(y, df_tornado['Max_Delta'], color='mediumseagreen', label='Positive Impact')
+    ax.barh(y, df_tornado['Min_Delta'], color='lightcoral', label='Negative Impact')
+    ax.set_yticks(y)
+    ax.set_yticklabels(df_tornado['Variable'], fontsize=12)
+    ax.axvline(0, color='black', linewidth=0.8, linestyle='--')
+    ax.set_title('Tornado Chart: IRR Sensitivity to Key Variables', fontsize=18, pad=20, weight='bold')
+    ax.set_xlabel(f'Change in IRR (%) from Base IRR ({base_irr:.2f}%)', fontsize=14)
+    ax.set_ylabel('Variable', fontsize=14)
+    ax.legend()
+    ax.grid(axis='x', linestyle='--', alpha=0.7)
+    for i, (p, n) in enumerate(zip(df_tornado['Max_Delta'], df_tornado['Min_Delta'])):
+        ax.text(p, i, f' +{p:.2f}%', va='center', ha='left', color='darkgreen')
+        ax.text(n, i, f' {n:.2f}%', va='center', ha='right', color='darkred')
+    plt.tight_layout()
+    plt.savefig("static/images/sensitivity_analysis_tornado.png", bbox_inches='tight')
+    print("\n✅ A sensitivity analysis Tornado chart has been saved as 'sensitivity_analysis_tornado.png'")
+
+if __name__ == "__main__":
+    optimization_results = run_optimizations_without_ml()
+    df_results = pd.DataFrame(optimization_results).sort_values(by="irr", ascending=False).reset_index(drop=True)
+    df_results.round(2)
+    display_and_save_results(df_results)
+    create_kpi_comparison_dashboard(df_results)
+    
+    if not df_results.empty:
+        best_scenario = df_results.iloc[0]
+        run_sensitivity_analysis(best_scenario['Params'], best_scenario['irr'])

static/css/style.css

@@ -0,0 +1,204 @@
+:root {
+  --bg-main: #121212;
+  --bg-surface: #1E1E1E;
+  --border: #2C2C2C;
+  --text-primary: #E0E0E0;
+  --text-secondary: #A0A0A0;
+  --accent: #2DECBF;
+  --accent-hover: #58FFCF;
+}
+
+html, body {
+  background-color: var(--bg-main);
+  color: var(--text-primary);
+  margin: 0;
+  padding: 0;
+  font-family: "Vazirmatn", sans-serif;
+  height: 100%;
+}
+
+body, .container, .container-fluid, .row, .col, main {
+  background-color: var(--bg-main) !important;
+  color: var(--text-primary) !important;
+}
+
+.navbar {
+  background-color: var(--bg-surface) !important;
+  border-bottom: 1px solid var(--border);
+}
+.navbar-brand {
+  color: var(--accent) !important;
+  font-weight: bold;
+}
+
+.card {
+  background-color: var(--bg-surface) !important;
+  border: 1px solid var(--border);
+  border-radius: 10px;
+  box-shadow: 0 4px 12px rgba(0,0,0,0.6);
+  color: var(--text-primary);
+}
+.card-header {
+  background-color: var(--bg-surface) !important;
+  border-bottom: 1px solid var(--border);
+  font-weight: 600;
+  color: var(--text-primary);
+}
+
+.btn-primary {
+  background-color: var(--accent);
+  border: none;
+  color: #000000;
+  font-weight: 600;
+  border-radius: 6px;
+  transition: background-color 0.2s, transform 0.1s;
+}
+.btn-primary:hover {
+  background-color: var(--accent-hover);
+  transform: translateY(-1px);
+}
+
+.btn-outline-light {
+  background-color: transparent;
+  border: 1px solid var(--accent);
+  color: var(--accent);
+}
+.btn-outline-light:hover {
+  background-color: var(--accent);
+  color: #000000;
+}
+
+
+.form-control, .form-select, textarea {
+  background-color: var(--bg-main) !important;
+  color: var(--text-primary) !important;
+  border: 1px solid var(--border);
+  border-radius: 6px;
+}
+.form-control:focus, .form-select:focus {
+  border-color: var(--accent);
+  box-shadow: 0 0 0 0.2rem rgba(45, 236, 191, 0.25);
+  background-color: var(--bg-surface);
+}
+
+
+.table {
+  background-color: var(--bg-surface) !important;
+  color: var(--text-primary);
+  border-color: var(--border);
+}
+.table-striped>tbody>tr:nth-of-type(odd)>* {
+  background-color: var(--bg-main);
+}
+.table-striped>tbody>tr:hover>* {
+  background-color: #2A2A2A;
+}
+
+.table td, .table th {
+  color: #FFFFFF !important;
+}
+
+.card-body, .alert, .form-label, .form-check-label, .card-header {
+  color: #FFFFFF !important;
+}
+
+.alert, .custom-alert {
+  background-color: #1E1E1E !important;
+  color: #FFFFFF !important;
+}
+
+.custom-alert, .alert {
+  background-color: var(--bg-surface) !important;
+  border-left: 4px solid var(--accent);
+  color: var(--text-primary);
+}
+
+canvas, img {
+  background-color: var(--bg-surface);
+  border: 1px solid var(--border);
+  border-radius: 8px;
+  padding: 8px;
+}
+
+footer {
+  background-color: var(--bg-surface);
+  border-top: 1px solid var(--border);
+  color: var(--text-secondary);
+  padding-top: 10px;
+  padding-bottom: 10px;
+}
+
+::-webkit-scrollbar {
+  width: 10px;
+}
+::-webkit-scrollbar-track {
+  background: var(--bg-main);
+}
+::-webkit-scrollbar-thumb {
+  background: var(--border);
+  border-radius: 8px;
+}
+::-webkit-scrollbar-thumb:hover {
+  background: var(--accent);
+}
+
+.table>:not(caption)>*>* {
+  background-color: transparent !important;
+}
+.modal-content, .dropdown-menu, .popover, .tooltip {
+  background-color: var(--bg-surface) !important;
+  color: var(--text-primary) !important;
+  border: 1px solid var(--border);
+}
+
+::placeholder {
+  color: var(--text-secondary);
+  opacity: 1;
+}
+::-webkit-input-placeholder {
+  color: var(--text-secondary);
+  opacity: 1;
+}
+:-ms-input-placeholder {
+  color: var(--text-secondary);
+  opacity: 1;
+}
+
+#loading-overlay {
+  position: fixed;
+  top: 0;
+  left: 0;
+  width: 100%;
+  height: 100%;
+  background-color: rgba(18,18,18,0.95);
+  display: flex;
+  flex-direction: column;
+  align-items: center;
+  justify-content: center;
+  z-index: 9999;
+}
+
+.loading-spinner {
+  width: 60px;
+  height: 60px;
+  border: 6px solid #2C2C2C;
+  border-top-color: #2DECBF;
+  border-radius: 50%;
+  animation: spin 1s linear infinite;
+}
+
+.text-accent {
+  color: var(--accent);
+}
+
+.loading-text {
+  margin-top: 15px;
+  font-size: 1.1rem;
+  color: #E0E0E0;
+  text-align: center;
+  letter-spacing: 0.5px;
+}
+
+@keyframes spin {
+  to { transform: rotate(360deg); }
+}

static/js/charts.js

@@ -0,0 +1,72 @@
+function renderKPIChart(labels, data) {
+  const ctx = document.getElementById('kpiChart').getContext('2d');
+  if (window.kpiChart) window.kpiChart.destroy();
+
+  window.kpiChart = new Chart(ctx, {
+    type: 'bar',
+    data: {
+      labels: labels,
+      datasets: [{
+        label: 'IRR (%)',
+        data: data,
+        backgroundColor: '#00C3A3',
+        borderColor: '#2DECBF',
+        borderWidth: 1,
+        borderRadius: 6,
+      }]
+    },
+    options: {
+      responsive: true,
+      plugins: {
+        legend: { display: false },
+        tooltip: { mode: 'index', intersect: false, bodyColor: '#FFFFFF', titleColor: '#FFFFFF' }
+      },
+      scales: {
+        x: {
+          ticks: { color: '#FFFFFF' },
+          grid: { color: 'rgba(255,255,255,0.1)' }
+        },
+        y: {
+          ticks: { color: '#FFFFFF' },
+          grid: { color: 'rgba(255,255,255,0.1)' }
+        }
+      }
+    }
+  });
+}
+
+function renderTornadoChart(labels, data) {
+  const ctx = document.getElementById('tornadoChart').getContext('2d');
+  if (window.tornadoChart) window.tornadoChart.destroy();
+
+  window.tornadoChart = new Chart(ctx, {
+    type: 'bar',
+    data: {
+      labels: labels,
+      datasets: [{
+        label: 'ΔIRR',
+        data: data,
+        backgroundColor: '#2DECBF',
+        borderRadius: 5
+      }]
+    },
+    options: {
+      indexAxis: 'y',
+      responsive: true,
+      plugins: {
+        legend: { display: false },
+        tooltip: { bodyColor: '#FFFFFF', titleColor: '#FFFFFF' }
+      },
+      scales: {
+        x: {
+          ticks: { color: '#FFFFFF' },
+          grid: { color: 'rgba(255,255,255,0.1)' }
+        },
+        y: {
+          ticks: { color: '#FFFFFF' },
+          grid: { color: 'rgba(255,255,255,0.1)' }
+        }
+      }
+    }
+  });
+}

templates/index.html

@@ -0,0 +1,156 @@
+{% extends "layout.html" %}
+{% block content %}
+
+<div class="row">
+  <div class="col-lg-4">
+    <div class="card bg-secondary mb-3">
+      <div class="card-header">پارامترهای مدل</div>
+      <div class="card-body">
+        {% if import_error %}
+          <div class="alert alert-danger">خطا در بارگذاری main.py — لطفاً فایل و وابستگی‌ها را بررسی کنید.<pre style="white-space:pre-wrap">{{ import_error }}</pre></div>
+        {% endif %}
+        <form method="post" action="{{ url_for('run') }}">
+          <div class="mb-2">
+            <label class="form-label">نرخ تورم (مثلاً 0.015)</label>
+            <input name="inflation" type="text" class="form-control" value="{{ defaults.inflation if defaults else '0.015' }}">
+          </div>
+          <div class="mb-2">
+            <label class="form-label">نرخ مالیات (مثلاً 0.10)</label>
+            <input name="tax" type="text" class="form-control" value="{{ defaults.tax if defaults else '0.10' }}">
+          </div>
+          <div class="mb-2">
+            <label class="form-label">طول پروژه (سال)</label>
+            <input name="years" type="number" class="form-control" value="{{ defaults.years if defaults else 15 }}">
+          </div>
+          <div class="mb-2">
+            <label class="form-label">ظرفیت (KTA) - حداقل</label>
+            <input name="cap_min" type="number" step="1" class="form-control" value="{{ defaults.cap_min if defaults else 500 }}">
+          </div>
+          <div class="mb-2">
+            <label class="form-label">ظرفیت (KTA) - حداکثر</label>
+            <input name="cap_max" type="number" step="1" class="form-control" value="{{ defaults.cap_max if defaults else 600 }}">
+          </div>
+          <div class="mb-2">
+            <label class="form-label">Technology</label>
+            <select name="technology" class="form-select">
+              {% for t in techs %}
+                <option {{ 'selected' if loop.first }}>{{ t }}</option>
+              {% endfor %}
+            </select>
+          </div>
+          <div class="mb-2">
+            <label class="form-label">Export market mix (0-1)</label>
+            <input name="export_mix" type="number" step="0.01" class="form-control" value="{{ defaults.export_mix if defaults else 0.7 }}">
+          </div>
+          <div class="form-check mb-2">
+            <input class="form-check-input" type="checkbox" id="sell_byproducts" name="sell_byproducts" checked>
+            <label class="form-check-label" for="sell_byproducts">فروش محصولات جانبی (Sell byproducts)</label>
+          </div>
+
+          <hr class="my-3">
+          <label class="form-label">اعمال تغییرات پیشرفته (JSON)</label>
+          <small class="text-muted d-block mb-2">برای ویرایش / اضافه کردن تکنولوژی یا قیمت‌ها از JSON استفاده کنید (اختیاری)</small>
+          <div class="mb-2">
+            <textarea name="tech_json" rows="6" class="form-control" placeholder='مثال: {"MyTech": {"capex_base_M": 200, "opex_base_cents_kg": 150}}'></textarea>
+          </div>
+          <div class="mb-2">
+            <textarea name="prices_json" rows="4" class="form-control" placeholder='مثال: {"pvc_s65_export": 1200}'></textarea>
+          </div>
+
+          <button class="btn btn-primary w-100" type="submit">اجرا و محاسبه</button>
+        </form>
+      </div>
+    </div>
+
+    <div class="card bg-secondary">
+      <div class="card-header">فایل خروجی</div>
+      <div class="card-body">
+        <a class="btn btn-outline-light w-100" href="{{ url_for('download_results') }}">دانلود CSV نتایج</a>
+      </div>
+    </div>
+  </div>
+
+  <div class="col-lg-8">
+    {% if top_kpis %}
+    <div class="row mb-3">
+      <div class="col-md-3">
+        <div class="card text-dark bg-light mb-2">
+          <div class="card-body text-center">
+            <h6>بهترین IRR (%)</h6>
+            <h4>{{ top_kpis.irr }}</h4>
+          </div>
+        </div>
+      </div>
+      <div class="col-md-3">
+        <div class="card text-dark bg-light mb-2">
+          <div class="card-body text-center">
+            <h6>سود سالانه ($M)</h6>
+            <h4>{{ top_kpis.annual_profit_M }}</h4>
+          </div>
+        </div>
+      </div>
+      <div class="col-md-3">
+        <div class="card text-dark bg-light mb-2">
+          <div class="card-body text-center">
+            <h6>CAPEX ($M)</h6>
+            <h4>{{ top_kpis.capex_M }}</h4>
+          </div>
+        </div>
+      </div>
+      <div class="col-md-3">
+        <div class="card text-dark bg-light mb-2">
+          <div class="card-body text-center">
+            <h6>Payback (yrs)</h6>
+            <h4>{{ top_kpis.payback }}</h4>
+          </div>
+        </div>
+      </div>
+    </div>
+    {% endif %}
+
+    <!-- Charts -->
+    <div class="card bg-secondary mb-3">
+      <div class="card-header">نمودارها و تحلیل حساسیت</div>
+      <div class="card-body text-center">
+        
+        {% if kpi_img %}
+          <h6 class="text-light mb-2">KPI Dashboard</h6>
+          <img src="{{ url_for('static', filename=kpi_img) }}" alt="KPI Dashboard" class="img-fluid rounded shadow border border-secondary mb-3">
+        {% else %}
+          <div class="row">
+            <div class="col-md-6">
+              <canvas id="kpiChart" height="200"></canvas>
+            </div>
+            <div class="col-md-6">
+              <canvas id="tornadoChart" height="200"></canvas>
+            </div>
+          </div>
+        {% endif %}
+
+        {% if tornado_img %}
+          <h6 class="text-light mt-4 mb-2">Sensitivity Analysis</h6>
+          <img src="{{ url_for('static', filename=tornado_img) }}" alt="Tornado Chart" class="img-fluid rounded shadow border border-secondary">
+        {% endif %}
+      </div>
+    </div>
+
+
+  </div>
+</div>
+
+<script>
+  const chartsPayload = {{ charts_json|safe if charts_json else 'null' }};
+  if (chartsPayload) {
+    const labels = chartsPayload.map((r, i) => r.Method ? r.Method : `M${i+1}`);
+    const irrs = chartsPayload.map(r => Number(r.irr) || 0);
+
+    renderKPIChart(labels, irrs);
+
+    const ranges = chartsPayload.map(r => {
+      const low = r.irr - 5; const high = r.irr + 5; return high - low;
+    });
+    renderTornadoChart(labels, ranges);
+  }
+</script>
+
+{% endblock %}

templates/layout.html

@@ -0,0 +1,59 @@
+<!doctype html>
+<html lang="fa" dir="rtl">
+  <head>
+    <meta charset="utf-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1">
+    <title>داشبورد بهینه‌سازی پروژه PVC</title>
+
+    <link href="https://cdn.jsdelivr.net/npm/bootstrap@5.3.0/dist/css/bootstrap.rtl.min.css" rel="stylesheet">
+
+    <script src="https://cdn.jsdelivr.net/npm/chart.js"></script>
+
+    <link rel="stylesheet" href="{{ url_for('static', filename='css/style.css') }}">
+  </head>
+
+  <body>
+    <nav class="navbar navbar-dark bg-dark border-bottom border-secondary shadow-sm">
+    <div class="container-fluid d-flex align-items-center">
+        <img src="{{ url_for('static', filename='images/logo.png') }}" alt="Logo" height="120" class="me-2 rounded">
+        <span class="navbar-brand fw-bold text-accent">🏭 Petro Optima</span>
+    </div>
+    </nav>
+
+    <main class="px-4 py-3">
+      {% with messages = get_flashed_messages(with_categories=true) %}
+        {% if messages %}
+          {% for category, msg in messages %}
+            <div class="alert alert-{{ 'info' if category=='info' else ('danger' if category=='danger' else ('success' if category=='success' else 'warning')) }} alert-dismissible fade show custom-alert" role="alert">
+              {{ msg }}
+              <button type="button" class="btn-close" data-bs-dismiss="alert" aria-label="Close"></button>
+            </div>
+          {% endfor %}
+        {% endif %}
+      {% endwith %}
+      {% block content %}{% endblock %}
+    </main>
+
+    <footer class="text-center text-muted py-3 border-top border-secondary">
+    </footer>
+
+    <script src="https://cdn.jsdelivr.net/npm/bootstrap@5.3.0/dist/js/bootstrap.bundle.min.js"></script>
+    <script src="{{ url_for('static', filename='js/charts.js') }}"></script>
+    <div id="loading-overlay" class="d-none">
+    <div class="loading-spinner"></div>
+    <div class="loading-text">در حال انجام محاسبات، لطفاً صبر کنید...</div>
+    </div>
+
+    <script>
+    document.addEventListener("DOMContentLoaded", function() {
+        const forms = document.querySelectorAll("form");
+        forms.forEach(form => {
+        form.addEventListener("submit", function() {
+            const overlay = document.getElementById("loading-overlay");
+            overlay.classList.remove("d-none");
+        });
+        });
+    });
+    </script>
+  </body>
+</html>