Analysis Region: Manufacturing Industry Sphere (Changwon, Gimhae, Yangsan, Haman, Jinju, Gyeongsangnam-do)
Core Areas: General Machinery, Machine Tools, Precision Machining, Metals, Automotive Parts, Nuclear Power, Defense Machinery Companies, Robot & AI Solution Companies, Changwon National Industrial Complex
Agenda: Is the concentration of machinery manufacturing in Gyeongnam expanding beyond factories utilizing AI and robots into a new industrial competitiveness that designs and sells AI machinery, industrial robots, and autonomous manufacturing systems?
Golden Time Type: Manufacturing Base Opportunity + AI Value-capture Risk
Reference Date: August 28, 2026
Version: Regional AX Golden Time Intelligence v3.2 Enhanced Criteria
In 2024, the Changwon National Industrial Complex recorded a production value of 62.223 trillion won , exports of 18.429 billion dollars , 3,216 resident companies , and 120,289 employees . Among the 38 national industrial complexes nationwide, it ranked 4th in production value, 5th in exports and resident companies, and 3rd in employment. Machinery production accounted for approximately 27% of the total machinery production in national industrial complexes, ranking first. Gyeongnam's transition to AI manufacturing is not merely a matter of attracting new industries to a region lacking an industrial base, but rather a matter of restructuring a large-scale machinery production system. ( Yonhap News , Status of Changwon National Industrial Complex )
In 2025, Changwon's exports totaled $22.63898 billion , accounting for 47.1% of Gyeongnam's total exports and 3.2% of the national total. Overall exports decreased by 0.7% compared to the previous year; while industrial machinery increased by 0.4%, basic industrial machinery fell by 12.1%, and machine components, tools, and molds decreased by 10.8%. Although defense exports increased, the specific categories of traditional machinery moved in different directions. ( Changwon Chamber of Commerce and Industry )
South Korea recorded the world's highest robot density in 2024, with 1,220 industrial robots per 10,000 manufacturing workers . This represents an increase of approximately 20% from 1,012 units in 2023. However, robot usage is concentrated in large automotive and electronics conglomerates, while the adoption rate of smart factories among small and medium-sized manufacturing enterprises (SMEs) stood at only 18.6% in 2024. Furthermore, 75.5% of the smart factories adopted were still in the basic stage. A high national robot density coexists with a low level of intelligence in the field within SMEs. ( International Federation of Robotics , Ministry of SMEs and Startups )
Policy investment in Gyeongnam has expanded significantly. The Gyeongnam Provincial Government has allocated 1.1909 trillion won for manufacturing AI innovation by 2026, representing approximately four times the 295.9 billion won allocated the previous year. From 2026 to 2030 , physical AI development and demonstration projects totaling 1 trillion won are underway, comprising 600 billion won in state funding and 400 billion won in private investment. The Changwon National Industrial Complex has been designated as a manufacturing demonstration hub where the machinery, shipbuilding, defense, and automotive industries are concentrated. ( Gyeongnam Manufacturing AI Investment , Physical AI Projects )
In Changwon, the following projects are being carried out simultaneously: the development of ultra-large-scale AI services specialized for the manufacturing industry (22.69 billion KRW) , the Manufacturing Convergence PINN data demonstration (32 billion KRW ) , the Smart Green AX Demonstration Industrial Complex (22.2 billion KRW) , the Power Generation Gas Turbine AI Factory (7 billion KRW) , and the Machinery and Defense Manufacturing DX Support Center (25.86 billion KRW) . The scope of these policies and demonstration projects has expanded from process automation to AI based on physical laws, digital twins, and autonomous diagnosis. ( Changwon City Manufacturing AI Plan )
On-site evidence is also being confirmed. CTR's Changwon plant has automated the visual inspection of aluminum parts using Vision AI, while GMB Korea and Taerim Industry have applied autonomous mobile robots and dual-arm robots to mixed-model production. Korens is demonstrating integrated digital twin control, and Shinsung Delta Tech and DX Solutions are testing physical information neural networks. Cases of companies where AI has been integrated into actual manufacturing processes already exist. ( Gyeongnam Manufacturing AI Sites )
However, among the 3,216 tenant companies in the Changwon National Industrial Complex, the number of firms that have applied AI to actual production control, productivity improvement rates by process, sales of proprietary AI and robot products, and the diffusion rate among SMEs are not publicly disclosed on a regular basis. Since 2014, Changwon City has proposed the establishment of smart factories for approximately 1,000 companies, but the basic, intermediate, and advanced stages are not separated from their current operational status. ( Changwon City Manufacturing AI Plan )
From 2026 to 2028, the Physical AI National Project, AX Demonstration Industrial Complex, DX Support Center, AI Factory, and the Yeongnam Region Manufacturing AI Convergence Project will simultaneously enter the demonstration phase. If companies in Gyeongnam remain solely as buyers of external AI and robots during this period, productivity may improve to some extent, but the added value of AI machines and robots will be accumulated by external suppliers.
Therefore, the current Golden Time for Gyeongnam's machinery industry is determined to be Manufacturing Base Opportunity + AI Value-capture Risk .
- Machinery industry clustering ≠ AI machinery industry
- Robot Adoption ≠ Robot Industry Competitiveness
- Building a smart factory ≠ autonomous manufacturing
- AI demonstration ≠ widespread adoption across the entire factory
- Productivity Improvement ≠ AI Product Sales
- Attracting national projects ≠ Local company value acquisition
Gyeongnam's machinery industry is centered around the Changwon National Industrial Complex and extends to the manufacturing zones of Gimhae, Yangsan, Haman, and Jinju. Changwon has a high proportion of machine tools, nuclear power, defense, automotive parts, and electrical machinery; Gimhae focuses on metalworking, industrial machinery, and automotive parts; Yangsan on machinery, electronics, and automotive parts; and Haman on casting, metal, and machine parts.
The Changwon National Industrial Complex is a large-scale production system with a managed area of approximately 25.3 million square meters and over 3,000 companies. Its projected production value of 62.223 trillion won and employment of 120,289 in 2024 serve as evidence that the potential demand for AI applications exists not in a single factory, but throughout the entire industrial city. ( Changwon City , Yonhap News )
However, tenant companies vary in production scale and data levels, ranging from large conglomerate finished product factories to small-scale processing, mold, and parts manufacturers. Even within the same industrial complex, the infrastructure for connecting ERP, MES, sensors, equipment communication, and AI models differs from company to company.
The current structure is assessed as having secured large-scale machine production aggregation, while enterprise-specific Digital Readiness is fragmented .
The production value of the Changwon National Industrial Complex increased from 41.7301 trillion won in 2020 to 45.4477 trillion won in 2021, 51.4283 trillion won in 2022, 60.0597 trillion won in 2023, and 62.2230 trillion won in 2024. The recovery in volumes related to defense, nuclear power, automobiles, and shipbuilding boosted the complex's overall production. ( Yonhap News )
In contrast, Changwon's total exports in 2025 are projected to reach $22.63898 billion, a 0.7% decrease. While weapons and industrial machinery increased by 29% and 0.4%, basic industrial machinery, as well as machine components, tools, and molds, experienced double-digit declines. This structure reflects a disconnect between the performance of large-scale project-based industries and traditional machine parts. ( Changwon Chamber of Commerce and Industry )
The increase in industrial complex production value may be the result of increased existing orders and finished product manufacturing, and is not equivalent to new sales of AI and robot products. Furthermore, item- and company-specific data used to distinguish between production recovery and industrial structure transformation are limited.
The current status is determined as Manufacturing Output Recovery confirmed / AI Machinery Revenue Shift unconfirmed .
The first change is that the value of machines has shifted from precision and durability to intelligence and data. Machine tools, gas turbines, and production facilities generate data for their entire operating period when control software, predictive maintenance, energy optimization, and digital twins are combined.
The second change is the shift in the unit of automation from individual robots to autonomous processes . The productivity structure differs between the stage where a single robot performs repetitive tasks and the stage where AI continuously optimizes orders, designs, materials, process conditions, inspections, and maintenance.
The third change is that the scope of competition has expanded from machine manufacturers to AI platform, robot, and semiconductor companies. If machine companies supply only hardware, control data and service revenue belong to external software companies.
The fourth change is that the unit of skill has shifted from individual experience to training data. If the skill required to judge machining noise, vibration, heat, tool wear, and surface quality is not converted into sensors and AI models, production capacity will disappear with the retirement of older skilled workers.
It is assessed that the Gyeongnam machinery industry is shifting from a competition to build good machines to a competition to enable machines to make their own decisions .
The Yeongnam Region Manufacturing AI Convergence Infrastructure Development Project is being carried out from 2024 to 2026 on a scale of 45 billion won . It is a project that collects and processes manufacturing data to develop and demonstrate AI solutions that resolve safety, productivity, and quality issues. The state-funded project budget for 2025 was 10 billion won. (Korea Information & Communication Technology Promotion Agency , Enterprise Plaza )
Changwon's mega-scale manufacturing AI service development project is worth 22.69 billion won, while the PINN manufacturing convergence data demonstration is worth 32 billion won. Unlike models that learn only production data, physical information neural networks combine physical laws such as heat, fluid dynamics, and stress. Interactions with the design, machining, and energy processes of machinery companies in Gyeongnam have been confirmed. ( Changwon City Manufacturing AI Plan )
22.2 billion won will be invested in the Smart Green AX Demonstration Industrial Complex by 2028, and three smart factories applying AI quality control and autonomous logistics, centered around Doosan Enerbility, Hyundai Wia, and Samhyun, have been established as operating models. 7 billion won has been allocated to the AI factory for the maintenance process of power generation gas turbines.
The Machinery and Defense Manufacturing DX Support Center will be established by 2026 with a budget of 25.86 billion won. It will be responsible for developing manufacturing AI models, applying equipment and solutions, addressing technical challenges faced by companies, and providing specialized personnel support. The infrastructure has been expanded from supporting individual smart factories to industrial complex-type AX demonstrations.
The current level of response is assessed as follows : participation of multiple demonstration projects and anchor companies is confirmed, while mass diffusion of small and medium-sized enterprises and commercialization of AI machine products are unconfirmed.
South Korea's manufacturing robot density is projected to reach 1,220 units per 10,000 workers in 2024, ranking first in the world. The global average in 2023 was 162 units, indicating that South Korea has already demonstrated a high density in the use of automation equipment. ( International Federation of Robotics )
However, high robot density is significantly influenced by investments in large electronics and automotive factories. Among the 163,273 small and medium-sized enterprises (SMEs) and middle-sized companies with factories nationwide, 19.5% had adopted smart factory technologies, while 18.6% were SMEs. Even among the adopting companies, 75.5% were still in the basic stage of partially collecting production information. ( Ministry of SMEs and Startups )
Although Gyeongnam has a high demand for machinery production, exports, and manufacturing demonstration, no data has been found comparing the region's own robot density, AI production control rate, and number of autonomous factories with those of machinery industry regions in Germany, Japan, and China.
The current location is determined to have High Robot Utilization Potential and Unmeasurable Regional Autonomous Manufacturing Readiness .
The production value of the Changwon National Industrial Complex is 62.223 trillion won .
There are 3,216 tenant companies and 120,289 employees .
Exports amounted to $18.429 billion in 2024 .
In 2025, Changwon's total exports amounted to $22.63898 billion , accounting for 47.1% of Gyeongnam's exports .
The robot density in Korea's manufacturing industry is 1,220 units per 10,000 workers .
The adoption rate of smart factories among small and medium-sized enterprises nationwide is 18.6% , and the proportion of smart factories in the basic stage is 75.5% .
Gyeongnam's investment in manufacturing AI for 2026 was projected at 1.1909 trillion won .
The Physical AI National Project is worth 1 trillion won from 2026 to 2030 , the Yeongnam Region Manufacturing AI Project is worth 45 billion won , the Mega Manufacturing AI is worth 22.69 billion won , the PINN Demonstration is worth 32 billion won , the AX Demonstration Industrial Complex is worth 22.2 billion won , and the DX Support Center is worth 25.86 billion won .
On the other hand, the number of AI production control companies in the Changwon National Industrial Complex, the proportion of processes applying AI, sales of AI and robot products, and productivity improvement rates by company are not disclosed as a single statistic.
The bottleneck revealed by the numbers is not a lack of project funds.
It is determined that there is a lack of a common measurement system to connect trillions of won in manufacturing AI investment with the field outcomes of over 3,000 companies .
The first gap lies between automation and autonomy . Even with robots, sensors, and MES, production remains limited to automated repetitive tasks unless AI predicts process conditions and controls equipment.
The second gap lies between user companies and suppliers . If machinery companies in Gyeongnam adopt external AI and robots, productivity improves, but revenue from robots, controllers, AI models, and data services may be accumulated by the external companies.
The third gap lies between demonstration companies and general companies . While there are publicly disclosed cases such as Doosan Enerbility, Hyundai Wia, CTR, and Korens, the actual application rate and diffusion rate among 3,216 tenant companies have not been confirmed.
The fourth gap lies between equipment data and training data . Older machines have different communication standards and sensors, and their operating conditions are not standardized. Even if data is collected, the reuse of common AI models is limited because formats vary by company.
The fifth gap lies between skilled technology and data assets . The number of processes and companies that converted the judgment of field technicians into training data is not disclosed.
Currently, the biggest Readiness GAP is determined to be the difference between the speed at which Gyeongnam becomes a huge demand center for AI and robots and the speed at which it establishes an AI machine and robot supply industry .
Gyeongnam is home to a concentration of the Changwon National Industrial Complex, the Manufacturing DX Support Center, the Industrial DX Collaboration Support Center, Gyeongnam Technopark, the Korea Electrotechnology Research Institute (KERI), the Korea Institute of Materials Science (KIMS), and numerous machinery companies. A plan has also been presented to cultivate 1,200 practical AI talents through the Innovation Academy from 2026 to 2031. ( Changwon City Manufacturing AI Plan )
On the other hand, 1,200 is a six-year plan, and as of 2026, results regarding completion, employment, and placement in SMEs are in the early stages. Compared to the 120,000 jobs at the Changwon National Industrial Complex, the supply of manufacturing and AI convergence talent is limited in scale.
Although the Manufacturing Data Open Lab and PINN demonstration projects exist, company-specific data connection specifications, the number of data-sharing companies, the training data reuse rate, and the security incident response system are not disclosed.
Current conditions indicate that institutional, demonstration, and educational infrastructure is expanding, while on-site AI personnel and data interoperability are in the early stages.
Changwon City has proposed the establishment of smart factories for approximately 1,000 companies since 2014. While this is a significant number when simply compared to the 3,216 companies located in the Changwon National Industrial Complex, it cannot be used as the current diffusion rate because information regarding whether companies outside the complex are included, whether they have closed or relocated, and the operational status of their systems has not been disclosed.
In a nationwide survey, the adoption rate of smart factories among small and medium-sized enterprises was 18.6%, and 75.5% of the factories were in the basic stage. There is no latest data available to determine whether machinery companies in Gyeongnam are higher or lower than the national average.
Companies such as CTR, Taerim Industry, GMB Korea, and Korens represent a group of firms where process changes have been confirmed through the application of AI and robots. In contrast, there is a lack of public evidence regarding the adoption rates of sensors, MES, and AI, as well as changes in productivity, among small-scale machining, casting, and mold companies.
The current level of diffusion is determined as follows : Leading Case confirmed / Long-tail SME Diffusion is judged as Data GAP.
Changwon is home to a concentration of large machinery, nuclear power, and defense companies, research institutes, DX support centers, and national projects. While Gimhae, Yangsan, and Haman have a wide distribution of small and medium-sized processing, metal, and automotive parts companies, there are relatively few public examples of manufacturing AI demonstrations and dedicated infrastructure.
Although the Yeongnam region manufacturing AI convergence project features a wide-area linkage structure, the number of participating companies by city and county, the processes in which AI is applied, and the self-investment rate after demonstration are not disclosed. It is highly likely that data maturity levels differ between anchor companies in Changwon and partner companies in the outskirts, even within the same supply chain.
Machine parts can be produced in Gimhae and Haman, while final assembly takes place in Changwon. However, if only the Changwon factory is AI-equipped and delivery and quality data from external partners are not integrated, the optimization of the entire supply chain will not occur.
The current spatial gap is determined to be between the Changwon National Industrial Complex-centered AX and the entire Gyeongnam machinery supply chain AX .
The schedule for the Physical AI National Project is 2026–2030, the Smart Green AX Demonstration Industrial Complex is 2028, and the Machinery and Defense Manufacturing DX Support Center is 2026. Gyeongnam’s major manufacturing AI projects are concentrated in the development, demonstration, and diffusion phases from 2026 to 2028.
During the same period, Changwon's machinery industry faces the coexistence of increased demand related to defense, nuclear power, and shipbuilding, alongside a decline in exports of traditional machinery parts. Companies that secured process data and investment funds while maintaining current orders will diverge during the next economic adjustment period from those that merely processed volume using existing production methods.
The retirement of older skilled workers is also a time risk over the same period. If personnel leave before judgments regarding processing, assembly, and inspection are converted into data, the field knowledge available for AI training itself decreases.
The period from 2026 to 2028 is determined to be a limited transition period in which existing orders in the machinery industry and national manufacturing AI investment coexist .
12-1. Golden Time Application Case in Basic Local Governments ① — Changwon City
The Changwon National Industrial Complex is home to 3,216 companies, 120,000 jobs, and a production base worth 62 trillion won. The mega-manufacturing AI, PINN, AX demonstration industrial complex, AI factory, and DX support center are also centered in Changwon.
The identified gap is the overall diffusion rate of the industrial complex rather than the scale of the demonstration project. There are three smart factories presented as operational models, and there are over 3,000 companies located in the industrial complex. If the results of the demonstration are not replicated into process standards and commercial solutions, the gap between leading factories and general factories will widen.
Changwon is assessed as High-density Testbed Opportunity + Diffusion Bottleneck Risk .
12-2. Golden Time Application Cases in Basic Local Governments ② — Gimhae and Haman
Gimhae and Haman are densely populated with casting, metalworking, machine parts, and mold companies. While these companies participate in the shipbuilding, automotive, and defense supply chains alongside large conglomerates in Changwon, they face disadvantages in terms of company size and the age of their facilities.
Companies that have not completed equipment sensorization, data standardization, and process documentation prior to AI application find it difficult to directly enter advanced AI demonstration projects. The nationwide smart factory adoption rate of 18.6% for SMEs and 75.5% for the basic stage demonstrate the structural extent of this barrier.
If the gap between basic digitalization and advanced AI demonstration is not narrowed within two to three years, companies in Gimhae and Haman will remain as outsourced processes with unconnected data, rather than participants in the AI supply chain.
Gimhae and Haman are judged to have a high Manufacturing Supply Depth and unmeasured AI Entry Readiness .
The first time risk is the loss of skill data . If a skilled worker retires without their experience in judging tool condition, vibration, thermal deformation, and material variations being recorded, the cost of restoring the same quality increases.
The second time risk is platform dependency . If machinery companies in Gyeongnam individually adopt external AI models and robots without securing their own control and data capabilities, revenue from software, updates, and operations after equipment sales shifts externally.
The third time risk is the isolation of validation . If the performance of the three leading factories and some excellent companies is not standardized, the same consulting and development costs will be repeated for each company after the project ends.
The fourth time risk is the outflow of small and medium-sized enterprises from the supply chain . Prime contractors demand real-time delivery, quality, and carbon data, but if partner companies remain with manual or retroactive reporting, transaction costs increase, and the likelihood of suppliers being replaced by digitally responsive companies grows.
The fifth time risk is the reversal of industrial identity . If a structure becomes entrenched where AI and robots increase in factories in Gyeongnam but robots, controllers, and AI machines are actually purchased externally, the added value of the machinery industry will be limited to hardware production.
Currently, irreversibility is judged to be the transfer of leadership in the AI machine industry to external parties while introducing AI, rather than the delay in AI adoption .
Gyeongnam simultaneously possesses actual data regarding machinery, processes, failures, and quality for AI training, as well as large-scale production sites for the immediate verification of AI products. Physical manufacturing data and the demand for demonstration, which are difficult for other AI hubs to secure, are concentrated in the same region.
Once Physical AI, PINN, Digital Twin, and autonomous logistics are accumulated into a common model, the same technology can be repeatedly applied to nuclear power, defense, shipbuilding, automobiles, and general machinery. There is also the potential for cross-industry development, where demonstrations in one industry can be converted into products for another.
On the other hand, if the results of demonstration remain within individual companies, Gyeongnam will not possess reusable industrial technologies despite having numerous success stories.
The currently achievable profit is determined to go beyond the improvement of productivity by factory and the formation of product lines of AI machines, robots, and manufacturing operations to be sold by Gyeongnam machinery companies .
Observation target | Apply AX | On-site judgment | Verification indicators |
| Machinery company | Machinery AX Registry | Distinction between user and supplier companies | AI product sales |
| aging facilities | Brownfield Data Kit | Assessment of AI applicability | Sensor and communication connectivity rate |
| Processing process | Process Foundation Model | Automatic optimization of conditions and quality | Process time and defect rate |
| skilled technology | Tacit Knowledge Capture | Knowledge assetization before retirement | Number of data conversion processes |
| Equipment failure | Predictive Maintenance | Pre-detection of stop risk | Unplanned stop time |
| Quality inspection | Vision·Sensor AI | Full inspection and cause tracing | Inspection time and rework rate |
| Production Logistics | AMR·Robot Orchestration | Determining bottlenecks between processes | Work-in-Progress and Transit Time |
| energy | AI Energy Twin | Cost and Carbon Assessment by Equipment | Energy produced per unit |
| supply chain | Supplier Digital Thread | Prime Contractor and Partner Data Connection | Real-time connection rate |
| Demonstration project | Replication Tracker | Distinction between one-time demonstrations and diffusion | Demonstration → Number of replicating firms |
| AI solutions | Manufacturing AI Marketplace | Connecting local supply and demand | Local solution purchase rate |
| Robot company | Robot Product Registry | Distinction between introduction and in-house industry | Robot production and export value |
| Talent | Manufacturing AI Skill Graph | Job-specific GAP assessment | Placement and tenure rates |
| regional disparities | Machinery AX Gap Map | Comparison of Changwon, Gimhae, and Haman | Maturity levels by city and county |
Gyeongnam Machinery AX Runtime
Order → Design → Materials → Processing → Assembly → Inspection → Logistics → Delivery → Operational Data → Maintenance → Design Improvement → Next Product
AI Machinery Value Chain
On-site problem → Manufacturing data → AI model → Robot/machine integration → Productivity verification → Commercial product → Expansion to other companies → Export → Service revenue
Manufacturing Base Opportunity + AI Value-capture Risk
Gyeongnam is not a region lacking a manufacturing base.
It is not a region without AI and robot demonstration projects either.
It is not a stage where there are no actual field application cases.
The current risk is that Gyeongnam machinery companies are growing into large-scale buyers of AI and robots, but are failing to grow as suppliers of AI machinery, robots, and manufacturing software .
If field demonstrations lead to the proliferation of reusable products and small and medium-sized enterprises between 2026 and 2028, Gyeongnam’s machinery industry will shift to a physical AI production industry.
If not connected, only a few leading factories will be autonomous, while the majority of SMEs will remain as basic smart factories, and AI added value will accumulate with external suppliers.
Currently, Golden Time is determined by Manufacturing Base Opportunity + AI Value-capture Risk .
- Number of machinery industry businesses in Gyeongnam
- Number of workers in the machinery industry
- Machinery industry production value and value added
- Exports of industrial machinery and basic industrial machinery
- Exports of machine elements, tools, and molds
- Changwon National Industrial Complex production value
- Tenant company operating rate
- Facility Years by Tenant Company
- Number of companies adopting smart factories
- Distribution of basic, intermediate, and advanced stages
- Actual system utilization rate
- Number of companies applying AI
- Number of processes controlled by AI
- Number of companies adopting robots
- Number of industrial robots in operation
- SME robot density
- Productivity before and after AI adoption
- Process time reduction rate
- Defect and rework reduction rate
- Reduction rate of unplanned facility shutdowns
- Change in energy production per unit
- Reduction rate of industrial accidents and hazardous work
- Number of manufacturing data collection companies
- Data connectivity rate between facilities and companies
- Number of processes for digitizing skilled technology
- Number of participating companies in the demonstration project
- Self-investment rate after completion of demonstration
- Number of companies replicating demonstration technology
- Number of AI solution companies in Gyeongnam
- Number of robot companies in Gyeongnam
- AI machine and robot product sales
- Exports of manufacturing AI solutions
- Local business solution purchase rate
- Dependence on external AI platforms
- Training of AI specialists in manufacturing
- Employment rate at local companies after completion of the course
- SME placement and retention rates
- Changwon, Gimhae, Haman, and Yangsan AX gap
- Prime Contractor and Partner Data Connection Rate
- Improvement rate of operating profit relative to AI investment
Data GAP: While industrial complex production and export statistics, as well as manufacturing AI project costs and participation cases, exist, a publicly available runtime connecting the cycle of company → equipment → data → AI application → productivity → AI commercialization → diffusion to other companies → export/service revenue based on the same company and process is not available.
Runtime Chain
Manufacturing Demand → Equipment/Processes → Data → AI/Robotics → Productivity Outcome → Commercial Products → Supply Chain Expansion → Exports → Service Revenue → Reinvestment
There is only one structural insight into Gyeongnam's machinery industry. The machinery industry that uses AI and the industry that sells AI machines are not the same transition.
Introducing external robots, vision AI, and digital twins to factories increases productivity. However, if external suppliers retain the rights to control models, operational data, software updates, and predictive maintenance services, companies in Gyeongnam remain as more efficient hardware producers.
Gyeongnam possesses a physical testbed for verifying AI machinery due to the concentration of machinery, nuclear power, defense, shipbuilding, and automotive manufacturing processes in the region. If this advantage is utilized solely for cost reduction at individual factories, the region becomes a major customer in the AI market. However, if the results of these demonstrations are reconfigured into products such as robots, intelligent machine tools, AI control, and maintenance services, the region becomes a supplier.
Golden Time Thesis — The golden time for Gyeongnam’s machinery industry is not the time to install a few more robots, but the time to preserve the large-scale physical AI demonstrations of 2026–2028 as products and services for sale by Gyeongnam companies. If commercialization is successful, the manufacturing sites in Changwon, worth 62 trillion won, will function as the world's largest market for AI machine demonstrations. If commercialization fails, Gyeongnam’s factories will become intelligent, but a structure will become entrenched where the ownership of that intelligence and recurring revenue belong to external AI and robot companies.
Version | Reference Date/Revision Date | Major changes |
| v1.0 | 2026.08.28 | This report was initially drafted based on the strengthened criteria of Regional AX Golden Time Intelligence v3.2. It cross-verified the following: Changwon National Industrial Complex projected 2024 production value of 62.223 trillion KRW, exports of 18.429 billion USD, 3,216 resident companies, and 120,289 employees; Changwon's 2025 exports of 22.63898 billion USD; Korea's robot density of 1,220 units per 10,000 workers; the adoption rate of smart factories in SMEs of 18.6% (basic stage) and 75.5%; Gyeongnam's manufacturing AI investment of 1.1909 trillion KRW; physical AI investment of 1 trillion KRW; and mega-scale manufacturing AI, PINN, AX Demonstration Industrial Complex, and DX Support Center projects. The "Golden Time" was identified as "Manufacturing Base Opportunity + AI Value-capture Risk," and the structural insight was condensed into a single conclusion: "The machinery industry using AI and the industry selling AI machinery are not the same transition." |
