diff --git a/__pycache__/generate_employee_data.cpython-312.pyc b/__pycache__/generate_employee_data.cpython-312.pyc
new file mode 100644
index 0000000..db0b100
Binary files /dev/null and b/__pycache__/generate_employee_data.cpython-312.pyc differ
diff --git a/__pycache__/test_generate_employee_data.cpython-312.pyc b/__pycache__/test_generate_employee_data.cpython-312.pyc
new file mode 100644
index 0000000..8ae1e74
Binary files /dev/null and b/__pycache__/test_generate_employee_data.cpython-312.pyc differ
diff --git a/employee_data.xlsx b/employee_data.xlsx
new file mode 100644
index 0000000..8d6a38f
Binary files /dev/null and b/employee_data.xlsx differ
diff --git a/generate_employee_data.py b/generate_employee_data.py
new file mode 100644
index 0000000..7f69658
--- /dev/null
+++ b/generate_employee_data.py
@@ -0,0 +1,126 @@
+import pandas as pd
+from faker import Faker
+import random
+from datetime import datetime, timedelta
+
+# Initialize Faker
+fake = Faker('ja_JP')  # Use Japanese locale for realism if desired, or standard. Given the user prompt was in Japanese, let's try 'ja_JP'.
+
+def generate_data(num_employees=1000):
+    """
+    Generates sample data for departments, employees, and salaries.
+
+    Args:
+        num_employees (int): The number of employees to generate.
+
+    Returns:
+        tuple: A tuple containing three pandas DataFrames (departments, employees, salaries).
+    """
+    # 1. Departments
+    departments_data = [
+        {'dept_id': 1, 'dept_name': 'Human Resources', 'base_salary': 4000000},
+        {'dept_id': 2, 'dept_name': 'Engineering', 'base_salary': 6000000},
+        {'dept_id': 3, 'dept_name': 'Sales', 'base_salary': 5000000},
+        {'dept_id': 4, 'dept_name': 'Marketing', 'base_salary': 4500000},
+        {'dept_id': 5, 'dept_name': 'Finance', 'base_salary': 5500000},
+        {'dept_id': 6, 'dept_name': 'Operations', 'base_salary': 4200000},
+        {'dept_id': 7, 'dept_name': 'Legal', 'base_salary': 7000000},
+        {'dept_id': 8, 'dept_name': 'Product', 'base_salary': 5800000},
+    ]
+    df_departments = pd.DataFrame(departments_data)
+
+    # 2. Employees
+    employees = []
+    salaries = []
+
+    start_date = datetime(2015, 1, 1)
+    end_date = datetime(2023, 12, 31)
+
+    for i in range(1, num_employees + 1):
+        emp_id = i
+        dept = random.choice(departments_data)
+        dept_id = dept['dept_id']
+
+        # Employee Data
+        first_name = fake.first_name()
+        last_name = fake.last_name()
+        email = f"{last_name.lower()}.{first_name.lower()}@example.com" # Simple email generation
+        hire_date = fake.date_between(start_date=start_date, end_date=end_date)
+
+        employees.append({
+            'emp_id': emp_id,
+            'first_name': first_name,
+            'last_name': last_name,
+            'email': email,
+            'hire_date': hire_date,
+            'dept_id': dept_id
+        })
+
+        # Salary Data (Simplified: current annual salary)
+        # Random variation around base salary
+        base = dept['base_salary']
+        variation = random.uniform(0.8, 1.5) # -20% to +50%
+        amount = int(base * variation)
+
+        salaries.append({
+            'salary_id': i, # Simple 1-to-1 mapping for this example
+            'emp_id': emp_id,
+            'amount': amount,
+            'effective_date': datetime(2024, 1, 1).date()
+        })
+
+    df_employees = pd.DataFrame(employees)
+    df_salaries = pd.DataFrame(salaries)
+
+    return df_departments, df_employees, df_salaries
+
+def analyze_data(df_departments, df_employees, df_salaries):
+    """
+    Performs aggregation on the employee data.
+
+    Args:
+        df_departments (pd.DataFrame): Departments data.
+        df_employees (pd.DataFrame): Employees data.
+        df_salaries (pd.DataFrame): Salaries data.
+
+    Returns:
+        pd.DataFrame: A summary DataFrame with aggregation results.
+    """
+    # Merge
+    merged = df_employees.merge(df_salaries, on='emp_id').merge(df_departments, on='dept_id')
+
+    # Aggregation
+    summary = merged.groupby('dept_name').agg(
+        employee_count=('emp_id', 'count'),
+        average_salary=('amount', 'mean'),
+        total_salary=('amount', 'sum')
+    ).reset_index()
+
+    # Format
+    summary['average_salary'] = summary['average_salary'].astype(int)
+
+    return summary
+
+def main():
+    """
+    Main execution function.
+    """
+    print("Generating data...")
+    df_departments, df_employees, df_salaries = generate_data(1000)
+
+    print("Analyzing data...")
+    df_summary = analyze_data(df_departments, df_employees, df_salaries)
+
+    output_file = 'employee_data.xlsx'
+    print(f"Saving to {output_file}...")
+
+    with pd.ExcelWriter(output_file, engine='openpyxl') as writer:
+        df_departments.to_excel(writer, sheet_name='Departments', index=False)
+        df_employees.to_excel(writer, sheet_name='Employees', index=False)
+        df_salaries.to_excel(writer, sheet_name='Salaries', index=False)
+        df_summary.to_excel(writer, sheet_name='Summary_Aggregation', index=False)
+
+    print("Done!")
+
+if __name__ == "__main__":
+    main()