综述Overview

The concept of the "Global 500" originates from international business rankings that sort companies by operating revenue. However, this index does not rely on the rankings of any existing list; instead, it returns to the underlying operating data of enterprises, builds its own evaluation model, and computes the rankings itself. The list covers about 40 leading enterprises worldwide, spanning major industries such as energy, finance, technology, automobiles, retail, pharmaceuticals, infrastructure, telecommunications, and food, striving to capture the true weight of each enterprise from multiple dimensions.

"全球企业500强"这一概念源自以营业收入排序的国际商业榜单，但本指数不以任何既有榜单的名次为依据，而是回归企业经营的底层数据，自建评价模型、自算排名。本榜单覆盖全球约40家头部企业，横跨能源、金融、科技、汽车、零售、医药、基建、通信、食品等主要行业，力求立体刻画企业的真实分量。

This index follows the CPPPS four-dimensional indicator system (Scale + Quality + Innovation + Contribution). The Scale dimension measures corporate size through operating revenue, total assets, and number of employees; the Quality dimension measures profitability through net profit and profit margin; the Innovation dimension measures technological accumulation through R&D investment; and the Contribution dimension measures how enterprises give back to society and people's livelihoods through taxes, employment capacity, and livelihood protection. Among these, "livelihood protection" is a distinctive CPPPS dimension that especially affirms foundational enterprises in energy, electric power, railways, food, and telecommunications that bear on the national economy and people's livelihood—a perspective generally absent from purely scale-based international rankings.

本指数遵循 CPPPS 四维指标体系（规模 + 质量 + 创新 + 贡献）：规模维度以营业收入、总资产、员工人数衡量企业体量；质量维度以净利润、利润率衡量盈利能力；创新维度以研发投入衡量技术积淀；贡献维度以税收、就业承载、民生保障衡量企业对社会与民生的反哺——其中"民生保障"为 CPPPS 特色维度，重点肯定能源、电力、铁路、粮食、通信等关系国计民生的基础性企业，这是国际纯规模榜单普遍缺失的视角。

For large-span scale indicators (revenue, total assets, R&D), the model first takes logarithms and then applies range standardization to prevent top enterprises from monopolizing the scores. Each indicator is min-max normalized to [0,1] and then combined by weighting, with all weights publicly disclosed and reproducible. The evaluation results show that Saudi Aramco, Amazon, State Grid, PetroChina, and Apple rank in the top five, with foundational energy and power enterprises forming the first tier together with global technology and retail giants.

模型对规模类大跨度指标（营收、总资产、研发）先取对数再做极差标准化，避免头部企业垄断分值；各项指标经 min-max 归一化至 [0,1] 后加权合成，权重全部公开、可复现。评价结果显示：沙特阿美、亚马逊、国家电网、中国石油、苹果位列前五，能源电力等基础性企业与全球科技、零售巨头共同构成第一梯队。

This ranking uses approximate values from publicly available common knowledge, all marked "subject to official figures," with data drawn from annual reports of listed companies, national statistics, and disclosures by state-owned asset regulators. It is permanently free and public, for research and reference only.

本排名数据采用公开常识近似值，一律标注"以官方为准"，数据来源为上市公司年报、各国统计与国资监管披露等；永久免费公开，仅供研究与参考。

全球企业500强 · 综合实力排名（约40家）Global 500 · Overall Strength Ranking (about 40)

排名	企业	国别	行业	综合指数	营收(亿$)	净利(亿$)	总资产(亿$)	员工(万)	研发(亿$)	税收(亿$)	民生	梯队
1	沙特阿美Saudi Aramco	沙特阿拉伯	石油天然气	61.61	4,940	1,213	6,600	7	12	950	88	A++
2	亚马逊Amazon	美国	电商·云计算	56.65	5,750	305	5,280	155	856	95	75	A++
3	国家电网State Grid	中国	电力	53.37	5,460	120	6,500	90	250	350	100	A++
4	中国石油PetroChina	中国	石油天然气	51.81	4,830	210	5,000	105	90	400	96	A+
5	苹果Apple	美国	消费电子	51.29	3,830	970	3,650	16	300	200	58	A+
6	微软Microsoft	美国	软件·云计算	50.69	2,450	880	5,120	22	295	200	60	A+
7	中国工商银行ICBC	中国	银行	48.79	2,140	510	6.30万	44	30	200	78	A+
8	谷歌Google	美国	互联网·科技	47.84	3,070	740	4,020	18	450	160	60	A+
9	沃尔玛Walmart	美国	零售	46.43	6,480	155	2,520	210	5	70	82	A+
10	国铁集团China State Railway Group	中国	铁路运输	46.41	1,700	25	1.30万	200	40	90	100	A+
11	中国建设银行CCB	中国	银行	45.78	2,000	470	5.40万	35	25	180	76	A
12	伯克希尔Berkshire	美国	多元控股	43.48	3,640	960	1.07万	36	1	270	60	A
13	中国石化Sinopec	中国	石油石化	42.08	4,300	90	3,000	60	50	300	94	A
14	大众Volkswagen	德国	汽车	41.87	3,220	180	6,300	68	220	90	72	A
15	丰田Toyota	日本	汽车	40.46	3,120	305	5,600	38	115	110	72	A
16	摩根大通JPMorgan Chase	美国	银行	36.2	1,600	500	3.90万	30	1	120	70	A
17	埃克森美孚ExxonMobil	美国	石油天然气	34.58	3,450	360	3,700	6	10	130	78	A
18	Meta	美国	互联网	34.4	1,350	390	2,290	6.7	380	100	55	A
19	中国建筑China State Construction	中国	建筑工程	34.36	3,050	50	3,600	36	40	120	86	A
20	三星电子Samsung Electronics	韩国	电子·半导体	34.25	1,940	220	3,700	27	220	70	65	A
21	中国移动China Mobile	中国	通信	32.73	1,420	180	2,800	45	50	120	84	B
22	联合健康UnitedHealth	美国	医疗健康	32.37	3,720	220	2,740	44	3	80	85	B
23	壳牌Shell	英国	石油天然气	30.86	3,160	200	4,000	10	10	90	76	B
24	梅赛德斯-奔驰Mercedes-Benz	德国	汽车	28.26	1,600	140	2,800	17	75	55	70	B
25	宝马BMW	德国	汽车	27.97	1,700	120	2,700	15	80	50	70	B
26	华为Huawei	中国	通信设备	27.61	990	120	1,800	20	230	160	70	B
27	强生Johnson & Johnson	美国	医药健康	27.29	850	130	1,670	13	150	40	85	B
28	阿里巴巴Alibaba	中国	电商·云计算	27.29	1,300	130	2,700	22	80	60	70	B
29	腾讯Tencent	中国	互联网	26.55	850	220	2,400	11	90	80	62	B
30	台积电TSMC	中国台湾	半导体	26.46	690	270	1,500	7.6	55	30	68	B
31	福特Ford	美国	汽车	26.19	1,760	43	2,700	17	80	20	70	B
32	英伟达NVIDIA	美国	半导体	24.42	610	300	660	3	87	50	55	B
33	罗氏Roche	瑞士	医药	24.19	680	120	1,000	10	150	30	84	B
34	国家能源集团China Energy Investment Group	中国	煤炭·电力	24.17	850	60	2,500	30	25	80	90	B
35	雀巢Nestlé	瑞士	食品饮料	22.95	1,000	120	1,300	27	20	40	82	B
36	比亚迪BYD	中国	汽车·电池	22.13	850	42	850	70	60	40	74	B
37	特斯拉Tesla	美国	汽车·能源	20.41	970	150	1,070	14	40	40	62	B
38	好市多Costco	美国	零售	19.95	2,500	63	690	30	2	25	80	B
39	中粮集团COFCO	中国	食品农业	19.49	1,000	10	1,300	17	15	50	90	B
40	宁德时代CATL	中国	新能源电池	14.71	560	62	1,000	12	30	25	70	B

营收、净利、总资产、研发、税收单位为亿美元（约值），员工为万人；"民生"为 CPPPS 民生保障评分（0-100，能源、电力、铁路、粮食、通信等基础性企业得分较高）。市值因部分央企未整体上市、口径不可比，仅作参考、不纳入评分。全部数据为公开常识近似值，以官方为准。
Revenue, net profit, total assets, R&D, and tax revenue are in USD hundreds of millions (approximate values), and employees are in units of 10,000 people; "livelihood" is the CPPPS livelihood-security score (0-100, with foundational enterprises in energy, power, railway, grain, and communications scoring higher). Market cap is for reference only and not included in scoring, because some central SOEs are not fully listed and the basis is not comparable. All data are public approximate values, subject to official figures.

算法详细论述Methodology

一、评价依据

This algorithm does not depend on the rankings of any existing list, but evaluates the comprehensive strength of enterprises from four dimensions based on verifiable underlying operating data:

本算法不依赖任何既有榜单的名次，而依据企业可核查的底层经营数据，从四个维度评价企业综合实力：

规模：营业收入、总资产、员工人数——衡量企业"做大"的体量与资源掌控力。
质量：净利润、利润率——衡量企业"做强"的盈利能力与经营效率，避免单纯以规模论高低。
创新：研发投入——衡量企业面向未来的技术积淀与持续竞争力。
贡献：税收、就业承载、民生保障——衡量企业对社会财政、就业与国计民生的反哺，其中民生保障为 CPPPS 特色维度，国际规模榜单普遍缺失。

二、模型的数学结构

The comprehensive strength index uses a weighted-sum multi-attribute utility model, linearly combining the four dimensions:

综合实力指数采用加权求和的多属性效用模型，四个维度线性合成：

综合指数 = 100 × ( 0.40·规模 + 0.25·质量 + 0.15·创新 + 0.20·贡献 )

Each dimension is composed of sub-indicators:

各维度由子指标合成：

规模 = 0.45·log₁₀(营收)′ + 0.30·log₁₀(总资产)′ + 0.25·员工′ 质量 = 0.60·净利润′ + 0.40·利润率′ 创新 = log₁₀(研发)′ 贡献 = 0.40·税收′ + 0.20·就业′ + 0.40·民生保障′

Here the prime mark (′) denotes the value after range standardization. Large-span scale indicators (revenue, total assets, R&D) are first log-transformed to suppress the dimensional monopoly of top enterprises; the sum of weights always equals 1, and everything is publicly disclosed, reproducible, and auditable.

其中′表示经极差标准化后的取值。规模类大跨度指标（营收、总资产、研发）先取对数，抑制头部企业的量纲垄断；权重之和恒等于 1，全部公开、可复现、可审计。

三、指标与权重依据

维度	指标	方向	处理	权重	依据
规模	营业收入	正向	对数	0.45	经营体量的核心口径
	总资产	正向	对数	0.30	资源掌控与抗风险家底
	员工人数	正向	线性	0.25	组织体量
质量	净利润	正向	线性	0.60	盈利绝对规模
质量	利润率	正向	线性	0.40	盈利效率
创新	研发投入	正向	对数	1.00	技术积淀与未来竞争力
贡献	税收	正向	线性	0.40	财政贡献
	就业	正向	线性	0.20	就业承载（稳就业）
	民生保障	正向	线性	0.40	能源、电力、铁路、粮食、通信等国计民生贡献（CPPPS 特色）

The overall weights of the four dimensions are: Scale 0.40, Quality 0.25, Innovation 0.15, and Contribution 0.20. The highest weight on Scale reflects "rank by size"; Quality provides efficiency correction; Innovation reflects long-termism; and the Contribution dimension brings in taxes, employment, and livelihood protection that international rankings generally ignore, highlighting the social value of enterprises.

四个维度的总体权重为：规模 0.40、质量 0.25、创新 0.15、贡献 0.20。规模权重最高体现"以体量定座次"，质量做效益修正，创新反映长期主义，贡献维度则把国际榜单普遍忽略的税收、就业与民生保障纳入，凸显企业的社会价值。

四、数据归一化与对数处理

Different indicators differ enormously in scale (revenue is measured in trillions while employees are measured in tens of thousands), so their dimensions must first be removed. This algorithm uses linear min-max normalization to map each indicator to [0,1]:

不同指标量纲差异巨大（营收以万亿计、员工以万人计），须先消除量纲。本算法采用线性 min-max 归一化，将各指标映射到 [0,1]：

x′ = ( x − x_min ) / ( x_max − x_min )

For large-span indicators such as revenue, total assets, and R&D, base-10 logarithms are taken before normalization to make score distributions more balanced and prevent a few giant enterprises from monopolizing high scores. When the maximum and minimum of an indicator are equal, it is uniformly set to 0.5 to avoid division by zero.

对营收、总资产、研发等大跨度指标，先取以 10 为底的对数再归一化，使分值分布更均衡，避免个别巨无霸企业垄断高分。当某指标最大最小值相等时统一取 0.5，避免除零。

五、数据来源与局限性

The data on revenue, net profit, total assets, employees, R&D, and taxes in this list are approximate values from publicly available common knowledge, referenced to annual reports of listed companies, national statistical bulletins, and disclosures by state-owned asset regulators, all marked "subject to official figures." The livelihood protection score is an expert assessment by CPPPS based on the foundational nature, public-interest nature, and livelihood-bearing capacity of each enterprise. Market capitalization is for reference only and is not included in scoring, because some enterprises (especially large central state-owned enterprises) are not fully listed and their figures are not directly comparable.

本榜单的营收、净利润、总资产、员工、研发、税收等数据采用公开常识近似值，参考上市公司年度报告、各国统计公报与国资监管披露，一律标注"以官方为准"；民生保障评分为 CPPPS 依据企业基础性、公益性与民生承载力的专家评估。市值因部分企业（尤其大型央企）未整体上市、口径不可比，仅作参考、不纳入评分。

This model has the following limitations: first, approximate values deviate from official precise figures; second, the metrics of different industries are not fully comparable (for example, the total assets of financial institutions cannot simply be placed side by side with those of physical enterprises); and third, exchange-rate conversions (approximate values) fluctuate. Therefore, this ranking is for research and reference only and does not constitute any investment or decision-making advice.

本模型存在如下局限：其一，近似值与官方精确值存在误差；其二，不同行业口径不完全可比（如金融机构总资产与实体企业不可简单并列）；其三，汇率折算（约值）存在波动。故本排名仅供研究与参考，不构成任何投资或决策建议。

六、参考文献

相关上市公司年度报告、可持续发展（社会责任）报告（公开披露数据）。
各国统计公报、国资监管机构披露数据（公开渠道）。
Keeney, R. L., & Raiffa, H.（1976）：Decisions with Multiple Objectives: Preferences and Value Tradeoffs，多属性效用理论与可加加权法。
《财富》世界500强榜单（仅作底层数据交叉参照，非名次来源）。
国家统计局：《国民经济行业分类》（GB/T 4754）。

说明Notes

本版采用四维（规模+质量+创新+贡献）自有评价模型，不转载任何既有榜单名次。规模类大跨度指标（营收、总资产、研发）取对数后极差标准化，其余指标线性极差标准化至 [0,1] 后加权合成：综合指数 = 100 ×（0.40·规模 + 0.25·质量 + 0.15·创新 + 0.20·贡献）。贡献维度纳入税收、就业与民生保障（CPPPS 特色），肯定关系国计民生的基础性企业。数据为公开常识近似值，以官方为准。排名永久免费公开、永不卖榜。
This edition adopts a four-dimension (scale + quality + innovation + contribution) proprietary evaluation model and does not reproduce the rankings of any existing list. Wide-span scale indicators (revenue, total assets, R&D) are log-transformed and then min-max standardized, while other indicators are linearly min-max standardized to [0,1] and then weight-combined: composite index = 100 × (0.40·scale + 0.25·quality + 0.15·innovation + 0.20·contribution). The contribution dimension incorporates tax revenue, employment, and livelihood security (CPPPS signature), affirming foundational enterprises vital to the national economy and people's livelihood. Data are public approximate values, subject to official figures. The ranking is permanently free and public, and rankings are never sold.

全球企业500强Global 500