China Industrial Robot Statistics 2026
By Axis Intelligence Research
Co-authors: Liam Michael | Last updated: August 31, 2026 | License: CC BY 4.0
China installed 295,000 industrial robots in 2024, the highest annual total ever recorded in any country, and now runs an operational stock of 2,027,000 units. In the same year Chinese robot brands outsold foreign brands at home for the first time. According to Axis Intelligence Research, only about half of the value inside those Chinese-branded machines is itself Chinese.
Quick Answer
China installed 295,000 industrial robots in 2024, or 54% of global deployments, against a world total of 542,000 units, per the International Federation of Robotics. China’s installed base reached 2,027,000 robots, the largest of any country. Chinese suppliers took 57% of their home market, up from 47% in 2023. According to Axis Intelligence Research, the Localization Depth Ratio (LDR) for China’s robotics industry reads 0.53 as of the 2024 reporting year: for every unit of finished-robot market share Chinese brands hold at home, roughly half as much of the content inside the machine is domestically made. China’s published robot density fell from 470 to 166 robots per 10,000 employees between reports, and not one robot was removed to cause it.
Key Findings
- According to Axis Intelligence Research, foreign robot suppliers lost approximately 18,922 units of Chinese demand in 2024 while the market itself grew by 18,712 units — foreign volume fell about 13% to roughly 126,850 units in a year when total installations rose to a record.
- According to Axis Intelligence Research, China’s Localization Depth Ratio stands at 0.53 for the 2024 reporting year, and at 0.58 on the alternative sales basis — the gap between assembling robots domestically and building what goes inside them.
- China’s robot density reads 166 units per 10,000 manufacturing employees in 2024, ranking 22nd worldwide and 6th in Asia, after the IFR recalculated the series on updated National Bureau of Statistics labour-market data. According to Axis Intelligence Research, the implied Chinese manufacturing workforce behind that reading is about 122.1 million people, against roughly 37.3 million implied by the previous report — a 3.27x change in the denominator, not the numerator.
- According to Axis Intelligence Research, the installation-weighted domestic supplier share across the five Chinese customer industries with disclosed supplier splits is 61.4%, above the 57% all-industry figure, because the sectors growing fastest are the ones Chinese suppliers already own.
- Automotive is the last foreign stronghold in Chinese robotics: according to Axis Intelligence Research, approximately 39,468 of the 57,200 robots installed in Chinese car plants in 2024 came from foreign suppliers — about 31.1% of every foreign-supplied robot sold in China that year, concentrated in one shrinking industry.
How Many Industrial Robots Does China Install Each Year?
The number that anchors everything else is 295,000. That is how many industrial robots were installed across Chinese factories in 2024, up about 7% on 2023 and the highest annual figure ever recorded anywhere. The International Federation of Robotics reported the figure in its World Robotics 2025 China release, alongside an operational stock of 2,027,000 units — a base that doubled in three years, passing one million in 2021.
Set against a global market of 542,000 installations and a worldwide installed base of 4,664,000 units, China took 54% of the year’s new deployments. Japan, the second market, installed 44,500 units. The United States installed 34,200.
China industrial robot installations and installed base, 2022 to 2024
| Metric | 2022 | 2023 | 2024 | Source |
|---|---|---|---|---|
| Annual installations | 290,144 | 276,288 | 295,000 | IFR World Robotics 2024 and 2025 |
| Share of global installations | — | 51% | 54% | IFR World Robotics 2024 and 2025 |
| Operational stock | — | 1,755,132 | 2,027,000 | IFR World Robotics 2024 and 2025 |
| Domestic supplier share | — | 47% | 57% | IFR World Robotics 2024 and 2025 |
| Domestic-supplied units | — | 130,516 | 168,150 (calculated) | IFR; Axis Intelligence Research |
| Foreign-supplied units | — | 145,772 | 126,850 (calculated) | IFR; Axis Intelligence Research |
There is no published 2025 installation count for China at the time of writing, and Axis Intelligence Research will not manufacture one. The figure arrives with World Robotics 2026, which the IFR has scheduled for September 24, 2026. Anyone quoting a 2025 China installation number today is quoting a model, not a count.
Liam Michael: Installations are a demand statistic with a lag, and the lag is the whole problem in this sector. The 2024 number describes cells that were commissioned, integrated and handed over — payload chosen, cycle time proven, uptime accepted by a plant manager. The 2026 headlines you are reading are about production capacity and press-conference targets. Those are not the same object. When the market grows by 18,712 units and one group of suppliers loses 18,922, something structural has happened to the installed base that no capacity announcement will tell you.
What Is China’s Robot Density in 2026, and Why Did It Fall From 470 to 166?
This is the most misquoted statistic in Chinese manufacturing, and the confusion is worth resolving carefully because it changes the ranking of the entire world.
In November 2024, the IFR reported China’s robot density at 470 robots per 10,000 manufacturing employees for 2023, placing it third worldwide, ahead of Germany and Japan. In April 2026, the IFR’s density release put China at 166 robots per 10,000 employees for 2024, ranking 22nd worldwide and 6th in Asia — while describing that as a 17% year-on-year increase. Both statements are correct. They are computed on different labour-market data, which the IFR states plainly: the 2024 reading is based on updated employment figures issued by China’s National Bureau of Statistics.
Robot density is a quotient. The numerator is operational stock, which grew. The denominator is manufacturing employment, which was restated. Axis Intelligence Research inverted the published figures to recover what each report assumed about the size of China’s factory workforce.
What the density readings imply about China’s manufacturing workforce
| Basis | Operational stock | Published density | Implied manufacturing employment | Source |
|---|---|---|---|---|
| World Robotics 2024 (2023 data) | 1,755,132 | 470 per 10,000 | approximately 37,343,234 | Axis Intelligence Research calculation from IFR data |
| World Robotics 2025 (2024 data) | 2,027,000 | 166 per 10,000 | approximately 122,108,434 | Axis Intelligence Research calculation from IFR data |
The arithmetic is not hidden: 1,755,132 divided by 470, multiplied by 10,000, returns 37.34 million. 2,027,000 divided by 166, multiplied by 10,000, returns 122.11 million. The ratio between them is 3.27. According to Axis Intelligence Research, China’s robot density did not collapse — the population of workers it is divided by expanded by roughly 84.8 million people between two editions of the same report. Applying the 17% growth the IFR reports for 2024, the restated 2023 reading works out at roughly 142 units per 10,000, against the 470 published at the time.
The effect is visible above China as well. The global average density reads 162 units in the 2023 release and 132 units in the 2024 release; Asia’s regional average reads 182 and then 131. A world average does not fall by 30 units in a year of record installations. A world average that is heavily China-weighted falls when China’s denominator is rewritten.
Robot density by country, 2024
| Rank | Economy | Robots per 10,000 employees | Source |
|---|---|---|---|
| 1 | Republic of Korea | 1,220 | IFR World Robotics 2025 |
| 2 | Singapore | 818 | IFR World Robotics 2025 |
| 3 | Germany | 449 | IFR World Robotics 2025 |
| 4 | Japan | 446 | IFR World Robotics 2025 |
| 5 | Sweden | 377 | IFR World Robotics 2025 |
| 6 | Denmark | 329 | IFR World Robotics 2025 |
| 7 | Slovenia | 315 | IFR World Robotics 2025 |
| 8 | United States | 307 | IFR World Robotics 2025 |
| 9 | Chinese Taipei | 302 | IFR World Robotics 2025 |
| 10 | Switzerland | 294 | IFR World Robotics 2025 |
| 22 | China | 166 | IFR World Robotics 2025 |
| — | World average | 132 | IFR World Robotics 2025 |
| — | European Union (EU-27) | 231 | IFR World Robotics 2025 |
Liam Michael: Density is a ratio pretending to be a fact, and the honest way to use it is to say which denominator you are standing on. Reading 166 as a decline is wrong. Reading 470 as current is worse, because it is the number every slide deck still carries. The defensible statement is the one nobody wants because it is undramatic: China has the largest installed base on earth by a wide margin, deployed across a manufacturing workforce large enough that per-worker penetration remains mid-table. Both halves of that sentence are true at once, and any forecast that uses only one half is selling something.
How Much of China’s Robot Market Do Domestic Brands Hold?
In 2024, Chinese robot suppliers outsold foreign suppliers in their own market for the first time, taking 57% of installations. That share was 47% in 2023, 30% in 2020 per the IFR’s assessment of the 15th Five-Year Plan, and 18% in 2015 according to the U.S.-China Economic and Security Review Commission. It had hovered near 28% for most of the preceding decade.
The unit split is where this gets sharp. In 2023 the IFR counted 130,516 domestically supplied units and 145,772 foreign-supplied units. Applying the 2024 share to the 2024 total gives approximately 168,150 domestic units and 126,850 foreign units.
According to Axis Intelligence Research, foreign robot suppliers sold roughly 18,922 fewer units into China in 2024 than in 2023, a decline of about 13%, during a year when the total market expanded by 18,712 units. Domestic suppliers added approximately 37,634 units, growth of about 28.8%. Foreign brands did not lose share to a shrinking market. They lost volume to a growing one.
Chinese industrial robot installations by customer industry, 2024
| Customer industry | Installations | Year-on-year | Domestic supplier share | Foreign-supplied units (calculated) | Source |
|---|---|---|---|---|---|
| Electrical and electronics | 83,000 | +7% | 59% | 34,030 | IFR; Axis Intelligence Research |
| Automotive | 57,200 | -12% | 31% | 39,468 | IFR; Axis Intelligence Research |
| Metal and machinery | 54,600 | +31% | 90% | 5,460 | IFR; Axis Intelligence Research |
| Food and beverage | 8,900 | +86% | 80% | 1,780 | IFR; Axis Intelligence Research |
| Textiles, leather and apparel | 5,700 | +29% | 100% | 0 | IFR; Axis Intelligence Research |
| Wood and wood products | 4,300 | record level | not disclosed | not calculable | IFR |
Those five sectors with disclosed supplier splits account for 71.0% of China’s 2024 installations. According to Axis Intelligence Research, the installation-weighted domestic supplier share across them is 61.4% — computed as 128,662 domestically supplied units over 209,400 total units — which runs above the 57% all-industry figure. The reason is compositional: metal and machinery, where Chinese suppliers hold 90%, grew 31% and now represents 18.5% of all Chinese installations, while automotive, where they hold 31%, shrank 12%.
China’s electronics sector alone absorbed 64% of the world’s industrial robot installations in that industry. In textiles, leather and apparel, China accounted for 95% of global installations, and every unit came from a Chinese supplier. In wood products, China represented 91% of global installations.
One figure is not reconciled here, and Axis Intelligence Research states it rather than smoothing it. The IFR’s September 2025 China release puts the domestic supplier share in metal and machinery at 90%; its May 2026 release on the 15th Five-Year Plan puts the same figure at 85%. We use 90% because it comes from the release dedicated to the Chinese market data, and we flag the divergence because a reader recomputing our weighted share from the other document will land at 60.5% instead of 61.4%.
Liam Michael: Look at where the foreign units still are. About 31.1% of every foreign-supplied robot sold in China in 2024 went into a car plant, and car plants installed 12% fewer robots than the year before. The remaining foreign volume in China is concentrated in the one large customer industry that is contracting, and it is concentrated there because automotive integrators buy on proven cycle time and warranty history rather than on price. That defence holds until a domestic supplier accumulates enough line-hours to be boring. Foreign automotive volume fell by roughly 11,140 units in a single year. Show me two consecutive years of that and the argument is over.
The Localization Depth Ratio: Measuring How Chinese a Chinese Robot Is
Market share counts badges on machines. It says nothing about what is inside them. Axis Intelligence Research built the Localization Depth Ratio (LDR) to close that gap, because the two figures are routinely quoted as if they were the same claim.
LDR — Localization Depth Ratio. The ratio of a country’s robot-component localization rate to its domestic finished-robot market share. It answers a single question: when a domestic brand wins a sale, how much of the machine did the domestic supply chain actually build?
Formula and inputs, disclosed:
LDR = robot component localization rate / domestic finished-robot market share
Component localization rate: 30% (USCC, November 2025)
Domestic finished-robot market share: 57% (IFR, 2024 reporting year)
LDR = 30 / 57 = 0.53
Reading, as of the 2024 reporting year (computed August 31, 2026): 0.53.
On the alternative basis, using the 52% domestic market share the U.S.-China Economic and Security Review Commission records in its assessment of the Made in China 2025 robotics targets, the ratio reads 0.58. Axis Intelligence Research publishes the range 0.53 to 0.58 rather than a single decimal, because the two market-share readings are measured differently — the IFR counts installations, the USCC cites a sales-based series covering the first three quarters of 2024.
How to read it. An LDR of 1.0 would mean localization inside the machine keeps pace with localization at the badge. At 0.53, roughly half the domestic brand share is carried on imported content. The Commission’s component detail explains where: China’s overall robot component localization rate is 30%, but it reaches 80% to 90% in mid-tier components, while approximately 50% of advanced sensors, 70% of actuators and 90% of ball screws are imported. Chinese firms compete well on unit volume and poorly on unit value — Kuka, the German firm Midea acquired in 2016, holds 11.5% of the global robotics market by value, against under 2% each for Estun, Siasun and EFORT.
Definitions and method. Both inputs are shares, so the ratio is dimensionless and comparable across years and countries once the same two source series are used. The component rate is a weighted assessment across component classes rather than a unit count, so the LDR describes the depth of localization, not a bill-of-materials percentage for any individual robot. Component and market-share data are published on different cadences, so successive readings will not always advance in step; when we restate, we restate the basis and leave the history intact.
Did China Meet Its Made in China 2025 Robotics Targets?
Beijing set the targets a decade ago and they are now testable, which is rare enough in industrial policy to be worth the space.
| Target | Deadline | Met? | Outcome | Source |
|---|---|---|---|---|
| 70% domestic market share in robotics | 2025 | No | 52% domestic market share | USCC, November 2025 |
| 70% indigenous key components | 2025 | Mixed | 30% overall; 80% to 90% in mid-tier | USCC, November 2025 |
| One or two of the top five global robotics firms | 2025 | Yes | One firm, Kuka, at 11.5% of the global market by value | USCC, November 2025 |
The Commission’s judgment is that missing the market-share target obscures how much moved: Chinese firms almost tripled their domestic share between 2015 and 2024 while annual installations also tripled. It also records that China accounts for 41% of the world’s installed industrial robots. Axis Intelligence Research computes 43.5% from the IFR’s own 2024 figures — 2,027,000 divided by 4,664,000 — and notes both readings rather than choosing one, since the two organisations are working from different reference years for the stock.
The scoreboard reads the way the LDR predicts. China met the target that could be bought (a top-five global firm, acquired), missed the target that required displacing incumbents at the badge level, and split the target that required building precision components from scratch.
Liam Michael: The ball-screw number is the one to keep. Ninety percent imported, in the component that decides how accurately an arm can repeat a motion ten million times. You can subsidise an assembly line into existence in three years. Grinding a ball screw to a tolerance that holds through a decade of duty cycles is a metallurgy and machine-tool problem, and those take longer than five-year plans. This is also why the humanoid programme is being framed as a leapfrog rather than an extension — a category with no incumbent has no incumbent’s components to displace. We track that side of the industry separately in our China humanoid robot statistics and in our data on Chinese humanoid robot companies.
How Many Industrial Robots Does China Manufacture?
Installations and output are different statistical objects, and conflating them is the most common error in coverage of this sector. Installations are counted by the IFR from supplier surveys and describe machines deployed in China. Output is counted by China’s National Bureau of Statistics and describes machines built in China, including units built for export, for inventory and by foreign suppliers manufacturing locally.
China produced 773,074 industrial robots in 2025, up 28%, per NBS data. Production reached 143,608 units in January and February 2026 alone, up 31% on the same period a year earlier, when output was 91,088 units. Official data for January to July 2026 put industrial robot output growth at 28.5% year on year, with equipment manufacturing value added up 9.7% and high-tech manufacturing up 13.8%, against 5.3% for industry as a whole. The manufacturing PMI stood at 49.2 in July 2026.
Working backwards from the published 28% growth, 2024 output was approximately 603,964 units. According to Axis Intelligence Research, China built roughly 2.05 industrial robots in 2024 for every one it installed — an estimate, because the growth rate is published rounded, and a comparison between two differently constructed series rather than a like-for-like ratio.
That gap is the number to watch as trade friction rises. It is not evidence of overcapacity on its own; exports, inventory and the foreign suppliers building in China all sit inside it. It does mean China’s robot manufacturing base is now sized well beyond its own record-setting domestic demand, which is the same pattern the country ran in solar cells, lithium cells and electric vehicles before those industries turned outward.
What Changes Next
The IFR expects global installations to reach 575,000 units in 2025 and to pass 700,000 by 2028, with China holding average annual growth potential near 10% through that period. The 15th Five-Year Plan places robotics at the centre of China’s industrial system and pivots artificial intelligence research toward physical applications — while the IFR’s reading of the plan puts humanoid commercialisation toward the end of the 2026 to 2030 window and expects the wide adoption of AI in conventional industrial robotics over five to ten years.
For the pages of this dataset that matter most, the date to mark is September 24, 2026. World Robotics 2026 publishes then, carrying China’s 2025 installation count, its 2025 domestic supplier share and a density reading computed on the restated labour-market basis for a second consecutive year. That second reading is what turns the 166 figure from a revision into a series.
Related Axis Intelligence Research datasets: Unitree Robotics statistics, China embodied AI funding, and the full robotics statistics and statistics libraries.
Methodology
Collection. Every figure in this article was retrieved from a source document opened during production on August 31, 2026, and every figure appears as a row in the accompanying CSV with its source organisation, source document, URL and retrieval date. No figure in this article originates from an unsourced estimate, and where a number does not exist publicly — China’s 2025 installation count — the article says so instead of interpolating.
Source hierarchy. Industry-body statistics compiled from supplier surveys (International Federation of Robotics), government statistical output (National Bureau of Statistics of China, via the State Council’s official English release), and government-commissioned research (U.S.-China Economic and Security Review Commission). No aggregator or secondary compilation is the named source of any headline figure.
Axis calculations. Thirty-four rows carry axis_calculated = yes in the CSV, each with its formula in the method_note column: implied manufacturing employment on both density bases, the denominator revision factor, the restated 2023 density, domestic and foreign unit splits for 2024, the foreign supplier unit change, sector-level foreign unit counts, the installation-weighted domestic supplier share, the automotive share of foreign-supplied units, China’s share of global operational stock, the Localization Depth Ratio on both bases, 2024 output, and the output-to-installation ratio. All were computed twice independently.
Scope. Installation and stock figures are IFR survey-based estimates, not audited counts. Domestic supplier shares count units by supplier nationality and include foreign robots built inside China by non-Chinese suppliers on the foreign side, as the IFR states. Output figures from the NBS cover enterprises above designated size and use a statistical sample that the bureau has adjusted, which the NBS itself flagged when explaining divergent service-robot growth. Density comparisons across report editions are not like-for-like, which is the subject of the reconciliation above rather than a caveat about it.
About This Dataset
china-industrial-robot-statistics-2026.csv contains every quantitative claim in this article as a tidy long-format table: one row per observation, with metric, value, unit, as_of_date, geography, segment, source_org, source_document, source_url, retrieved_date, is_primary, axis_calculated, method_note, data_type and license. Temporal coverage: 2015 to 2028, with forecasts flagged in data_type. Spatial coverage: China, with global and country comparators.
License: CC BY 4.0. Free to use, redistribute and build on with attribution.
Citation
APA — Axis Intelligence Research. (2026). China industrial robot statistics 2026: Installations, robot density, domestic market share and the localization depth gap. https://axis-intelligence.com/china-industrial-robot-statistics/
MLA — Axis Intelligence Research. “China Industrial Robot Statistics 2026: Installations, Robot Density, Domestic Market Share and the Localization Depth Gap.” Axis Intelligence, 31 Aug. 2026, axis-intelligence.com/china-industrial-robot-statistics/.
Chicago — Axis Intelligence Research. “China Industrial Robot Statistics 2026: Installations, Robot Density, Domestic Market Share and the Localization Depth Gap.” Axis Intelligence, August 31, 2026. https://axis-intelligence.com/china-industrial-robot-statistics/.
Frequently Asked Questions
Why did China’s robot density drop from 470 to 166 without any robots being removed?
Because the denominator changed. Robot density divides operational stock by manufacturing employment, and the IFR recomputed China’s figure on updated National Bureau of Statistics labour-market data. According to Axis Intelligence Research, the implied workforce moved from about 37.3 million to about 122.1 million between the two report editions, a factor of 3.27. The stock rose from 1,755,132 to 2,027,000 over the same period. Both the 470 and the 166 are correctly computed; only the 166 is current.
Does a 57% domestic market share mean Chinese robots are 57% Chinese?
No, and this is the most consequential misreading in the sector. The 57% counts finished machines sold by Chinese-headquartered suppliers. Component localization is a separate measurement, assessed at 30% overall by the USCC, which is why the Localization Depth Ratio reads 0.53. Roughly 50% of advanced sensors, 70% of actuators and 90% of ball screws used in Chinese robots are imported.
Which customer industry in China still buys foreign robots, and how many units?
Automotive. According to Axis Intelligence Research, approximately 39,468 of the 57,200 robots installed in Chinese car plants in 2024 were foreign-supplied, about 31.1% of all foreign-supplied units in China that year. Chinese suppliers hold 90% of metal and machinery, 80% of food and beverage and 100% of textiles installations, but only 31% of automotive.
How many industrial robots did China install in 2025?
No figure has been published. Installation counts come from the IFR’s World Robotics series, and the 2025 count is scheduled for publication on September 24, 2026. China’s 2025 industrial robot production figure is available separately at 773,074 units, but production and installation are different measurements and should not be substituted for one another.
Why does China’s robot production figure exceed its installation figure by so much?
They count different things. NBS output counts robots built in China, including units for export, units held in inventory and units built locally by foreign suppliers. IFR installations count robots deployed in Chinese plants. According to Axis Intelligence Research, the implied ratio for 2024 is about 2.05 units produced per unit installed, derived from an estimated 603,964 units of 2024 output against 295,000 installations.
Did China hit its Made in China 2025 robotics targets?
It missed the headline one. The USCC records a 70% domestic market share target against a 52% outcome, a 70% component localization target against 30% overall (80% to 90% in mid-tier components), and a met target of one to two top-five global robotics firms, achieved through Midea’s 2016 acquisition of Kuka rather than indigenous growth.
Is China’s robot density low or high compared with other countries?
Mid-table by density, first by volume. At 166 units per 10,000 employees China ranks 22nd worldwide against Korea’s 1,220, Singapore’s 818 and Germany’s 449, and sits above the global average of 132. It simultaneously holds the world’s largest installed base at 2,027,000 units, roughly 4.5 times Japan’s. Both statements describe the same country in the same year.
What does the Localization Depth Ratio tell an equipment buyer that market share does not?
It separates supplier nationality from supply-chain exposure. A 0.53 reading means a plant sourcing a Chinese-branded robot is still buying meaningful imported content inside the machine, mainly in precision components, which matters for lead times, spare-part availability and tariff exposure in a way that the badge on the housing does not.
How reliable are Chinese industrial robot statistics?
The installation and stock series come from IFR supplier surveys covering around 40 countries on a standardised methodology, which is the most comparable source available and is used by governments and researchers. Output figures come from the NBS and cover enterprises above designated size, on a sample the bureau adjusts and discloses. Neither is an audited census. Every figure in this dataset carries its source and retrieval date so a reader can weight them independently.
What should I watch when World Robotics 2026 publishes?
Three numbers: China’s 2025 installation count against the 295,000 record, the domestic supplier share against 57%, and a second density reading on the restated labour-market basis. The third matters most, because one restated reading is a revision and two make a series.