mesa/测试数据 产业 数据.py

import pandas as pd
import numpy as np

# 设置随机种子
np.random.seed(42)

num_companies = 170  # 企业ID范围

# 生成企业和设备数据
num_rows = 220  # 每个表的行数

company_ids = np.arange(num_companies)

# 第二步：生成剩余的随机企业ID
remaining_ids = np.random.randint(0, num_companies, size=num_rows - num_companies)

# 合并两部分的企业ID
all_company_ids = np.concatenate([company_ids, remaining_ids])

# 第三步：对企业ID进行升序排序
all_company_ids.sort()

device_ids = np.random.randint(51, 107, size=num_rows)
material_ids = np.random.randint(0, 51, size=num_rows)
product_ids = np.random.randint(0, 107, size=num_rows)

device_quantities = np.random.randint(50, 200, size=num_rows)
material_quantities = np.random.randint(100,200, size=num_rows)
product_quantities = np.random.randint(20, 100, size=num_rows)

# 创建三个表格的数据框
df_devices = pd.DataFrame({
    'Firm_Code': all_company_ids,
    '设备id': device_ids,
    '设备数量': device_quantities
})

df_materials = pd.DataFrame({
    'Firm_Code': all_company_ids,
    '材料id': material_ids,
    '材料数量': material_quantities
})

df_products = pd.DataFrame({
    'Firm_Code': all_company_ids,
    '产品id': product_ids,
    '产品数量': product_quantities
})

# 保存为CSV文件
df_devices.to_csv('测试数据 companies_devices.csv', index=False)
df_materials.to_csv('测试数据 companies_materials.csv', index=False)
df_products.to_csv('测试数据 companies_products.csv', index=False)