Textile Mills Must Adopt AI But Should Fix Their Thread: People, Data and Systems

New Products and Innovation

Textile Mills Must Adopt AI But Should Fix Their Thread: People, Data and Systems

Artificial intelligence promises better quality, higher productivity and less waste across the textile value chain. But fragmented data, skills shortages and high investment costs are slowing adoption. For India’s textile industry, the challenge is not whether to embrace AI, but how quickly it can build the capabilities to use it.

Artificial intelligence (AI) is moving into textile manufacturing, promising to transform everything from fabric inspection and predictive maintenance to production planning, energy management and supply chain traceability. The technology is advancing rapidly, and the commercial opportunity is growing just as fast.

The global AI in textile market is estimated at US$ 5.54 billion in 2026, and is expected to reach US$ 17.65 billion by 2030, according to a forecast by The Business Research Company. The report estimates a compound annual growth rate of 33.6% through 2030. Asia-Pacific was the largest regional market in 2025, while North America is expected to record the fastest growth over the forecast period.

These figures point to a market gathering momentum. Textile manufacturers will have to figure out how effectively they can put AI to work on the factory floor.

Like automation and digitalisation, AI can improve overall efficiencies in textile production and planning. AI can identify defects, anticipate machine failures, optimise energy consumption and improve demand forecasting. It can help manufacturers make faster decisions and reduce costly production errors.

The industry is, however, discovering that buying the technology is often the easiest part. Getting it to deliver measurable results is considerably harder.

A new white paper from the International Textile Manufacturers Federation (ITMF) and the Institut für Textiltechnik of RWTH Aachen University (ITA) puts the problem into perspective. Its central finding is that the biggest barriers to AI adoption are not shortcomings in algorithms. They are poor data integration, a shortage of skilled people and gaps in governance.

In other words, the textile industry may have plenty of digital information, but it is not necessarily equipped to use it intelligently.

The industry is digitally fragmented

The ITMF–ITA white paper, AI in the Textile Value Chain: A Strategic Guide to Artificial Intelligence for the Global Textile Industry, draws on ten webinars conducted between 2024 and 2026, involving more than 20 experts from companies, research institutions and policy bodies. It also incorporates discussions with industry practitioners at the ITMF Conference 2025 in Yogyakarta.

The paper describes the textile industry as “not under-digitalized, but digitally fragmented”. Decades of digitalisation have generated substantial amounts of data, but much of it remains scattered across disconnected systems. Information may be available within individual machines, departments or business functions without being accessible in a form that allows the entire operation to benefit.

This is a fundamental problem. AI does not automatically repair poor data. It can amplify its weaknesses. If information is incomplete, inconsistent or inaccessible, the output of even sophisticated systems can be unreliable.

The industry already has practical applications where AI is delivering results. Computer vision is being used for fabric and yarn inspection, while predictive maintenance can identify the likelihood of equipment failure before a breakdown occurs. AI-based energy optimisation and demand forecasting can also improve operational efficiency.

One example cited in the ITMF–ITA paper is a nonwovens production line where causal AI reduced scrap by 10%. This is the kind of measurable gain that makes AI commercially relevant: less waste, better process control and improved use of resources.

The technology is also beginning to move beyond providing recommendations. Agentic AI systems are being developed to coordinate multi-step activities such as sourcing, while causal AI can adjust production settings directly, subject to human supervision. The paper estimates that end-to-end AI workflows could become realistic within two to three years.

But the transition requires more than technology. Manufacturers need to establish control over sensitive data, strengthen cybersecurity, understand the limitations of AI systems and retain human accountability for decisions.

The report recommends that companies build skills and data capabilities before scaling technology. It also advocates starting with clearly defined applications rather than attempting a complete transformation at once. For small and medium-sized enterprises (SMEs), industry associations could play an important role through shared data standards, collaborative platforms and sector-specific tools.

India is still at an early stage

For India’s textile and apparel industry, the challenge is particularly relevant. The sector has a large and diverse manufacturing base, with companies operating at very different levels of digital maturity. While some enterprises are investing in automation and connected systems, many others are still building basic digital capabilities.

A study by the Confederation of Indian Textile Industry (CITI), supported by the Northern India Textile Research Association (NITRA), provides a useful snapshot. The study, AI, Automation & Digitalisation Readiness in the Indian Textile & Apparel Industry, was released at UP TEX 2026 in Lucknow on September 8.

Covering the value chain from spinning to garmenting, including home textiles, technical textiles and allied services, the survey found that 35% of firms had not started adopting AI. More significantly, 38% operated without a digital system, while only around 14% reported having fully integrated digital systems.

Automation adoption was concentrated mainly in routine applications such as machine monitoring and setting. More complex functions, including production scheduling, had lower adoption.

The findings place India in the “early-to-emerging” stage of AI and digital adoption in textiles and apparel, with the country continuing to trail China. High investment costs and shortages of skilled talent were identified as the leading barriers to wider adoption.

The survey also found that training remains largely informal and enterprise-led, while technology's impact on employment is mixed. This points to a gap that cannot be addressed simply by making more software available. Workers, supervisors, technicians and managers all need the capabilities to work with increasingly digital production systems.

The CITI study recommends practical support to help companies, particularly MSMEs, move forward. Demonstration factories, capital subsidies and common digital platforms emerged as important priorities. Industry-specific AI tools, Centres of Excellence, training programmes, pilot projects and financing mechanisms could also help reduce the risks of adoption.

For smaller manufacturers, the ability to test a technology before investing in it could make a significant difference. Shared platforms and sector-level initiatives could also reduce the cost of building capabilities that individual companies may struggle to develop on their own.

From factory data to business performance

The commercial case for AI is not confined to India. Textile exporters in Bangladesh and Vietnam are also looking at smart manufacturing as a way to improve productivity, control costs and meet increasingly demanding buyer requirements.

In Bangladesh, the focus is on connecting factory challenges with practical technology solutions. The TIE–ITET Textile Innovation Expo 2026 is part of an initiative by Textile Innovation Exchange to bring manufacturers and technology providers together through structured implementation programmes. The objective is to help factory teams identify bottlenecks, test digital tools under actual shop-floor conditions and translate technology investments into business results.

The applications being considered include automated cutting, computer vision, digital dyeing, predictive maintenance and energy-efficient machinery. Each addresses a specific manufacturing challenge.

Automated cutting can reduce fabric waste. Computer vision can identify fabric defects earlier and help prevent quality problems from travelling further down the production line. Predictive maintenance can reduce unplanned machine downtime, while digital dyeing can help control water and chemical consumption.

The potential savings extend beyond manufacturing. McKinsey and The Business of Fashion have estimated that digital sourcing tools can reduce fashion sourcing costs by up to 12%. However, around 90% of fashion companies' AI projects reportedly stall at the pilot stage, according to McKinsey research.

That gap between experimentation and implementation is significant. A successful pilot does not automatically translate into a factory-wide solution. Systems must work with existing production processes, data must be reliable, and employees must be trained to use the technology.

A 2026 textile manufacturing study found that data-driven predictive maintenance increased the time between machine failures by 38%. This demonstrates the value of connecting machine data with operational decisions. The gains come not from having more data, but from using it to anticipate problems and act before production is disrupted.

Vietnam's experience offers another perspective. At a conference in Hanoi, industry representatives stressed that digital transformation must extend beyond investment in machinery and software. It must also involve changes in management, production organisation and business practices.

The experience of two Vietnamese manufacturers illustrates why technology must be accompanied by process improvements. Garment 10 Corporation recorded a 52% increase in labour productivity and an 8% reduction in defects after implementing lean manufacturing, while also cutting working time by one hour per day. At Duc Giang Corporation and its member companies, productivity increased by 8–10%, while production-line defect rates fell from 25% to between 10% and 12%.

These results are not presented as AI-driven gains. Rather, they underline an important point for manufacturers considering AI: digital technology is more effective when supported by disciplined production systems, better management and workforce skills.

Vietnam's textile and garment exports reached US$ 35.72 billion in the first nine months of 2026, up 3.3% year-on-year. The industry has set a full-year export target of US$ 48–49 billion. Improving productivity and moving towards higher value-added production are important parts of its competitiveness agenda.

New applications are emerging in India

While the industry's readiness remains uneven, Indian companies are beginning to explore AI-based solutions in different parts of the textile value chain.

One example is the initiative by RePut.ai, TEXPROCIL and Control Union India to develop an AI-powered Textile Sustainability Stack. The collaboration aims to create a unified digital framework to support textile exporters with end-to-end traceability, Digital Product Passports (DPPs), carbon accounting, ESG reporting, Extended Producer Responsibility (EPR) compliance and third-party verification.

The initiative reflects the growing importance of digital information in international textile trade. As sustainability and transparency requirements become more demanding, exporters need reliable information about products, processes and supply chains. A connected data system could help companies organise this information and support compliance-related activities.

The proposed platform brings together TEXPROCIL's industry reach, Control Union's certification and verification expertise, and RePut.ai's AI-based sustainability platform. It is intended to help exporters strengthen supply chain transparency and respond to evolving regulatory requirements, including those in the European Union.

The initiative is still a development project, rather than evidence of an already established, industry-wide system. Its significance lies in the attempt to bring multiple sustainability-related data requirements into one digital framework.

AI is also finding applications in material selection and product development. Response Fabrics has announced an AI Fabric Visualizer aimed at making material selection smarter. While the available information on the tool is limited, it reflects the broader movement towards using AI to support decisions earlier in the product development process.

This is an area where AI's influence extends beyond the factory floor. The Lectra Observatory's latest white paper, The Intelligence Edge Is Here, examines how AI is changing fashion product creation, manufacturing, marketing, traceability and collaboration.

The report brings together perspectives from experts across the fashion ecosystem, including representatives from KIKO Milano and C.P. Company, as well as technology and industry specialists. Its central argument is that AI can support more connected workflows and faster decision-making, but its effectiveness depends on the quality and accessibility of the data being used.

For textile and apparel companies, this means AI could influence several stages of the business: forecasting what the market may need, developing products, selecting materials, planning production and managing information across the supply chain. The value will depend on how well these functions are connected.

Skills must catch up with technology

One of the biggest gaps in the Indian industry's AI journey is the availability of people who can work with new technologies. The CITI–NITRA study identifies the shortage of skilled talent as a major obstacle, while also pointing to the need for more structured training.

The All India Council for Technical Education (AICTE) is expanding its AI and Industry 4.0 skilling initiatives through its Indovation Centres and the TECHNOVATE Skilling Series, conducted in collaboration with the Consortium for Technical Education in Coimbatore.

The programme has conducted more than 16 hands-on training programmes, benefiting over 850 students and faculty members from Rajasthan and other north-western states. Its areas of focus include AI, cybersecurity, robotics and Industry 4.0 applications.

AICTE currently has 11 Indovation Centres across India, aimed at promoting innovation, entrepreneurship, start-up development and technology transfer in higher education institutions. It has also signed agreements with JNTU Hyderabad and Panjab University to establish centres on their campuses. Dedicated skilling laboratories are being planned at the Jaipur centre, while updated curricula covering areas such as Textile Engineering, Electrical Engineering, BBA and BCA are also part of the initiative.

These technical and professional education programmes point to the importance of building a wider talent pool with exposure to emerging technologies. The textile industry will need people who understand both manufacturing processes and digital systems.

That combination is critical. A data scientist may understand an algorithm, but applying it to a spinning mill, weaving unit, processing plant or garment factory requires knowledge of the manufacturing process itself. Similarly, production managers and shop-floor teams need to understand how data can improve decisions without disrupting established workflows.

The ITMF–ITA paper makes the case for investing in people before scaling technology. It also highlights the importance of trust. Manufacturers must know where their data is stored, who can access it, how systems make decisions and who remains accountable when something goes wrong.

These questions will become more important as AI moves from analysing information to taking action within production and business processes.

The next step is implementation

The textile industry's AI journey is often discussed in terms of sophisticated technologies, robotics and the prospect of fully automated factories. But the immediate opportunities are more practical.

A mill does not need to transform its entire operation overnight. It can begin by addressing a specific problem: recurring machine failures, inconsistent quality, excessive energy consumption, fabric wastage or delays in production planning. A well-defined application, supported by reliable data and trained employees, can provide a clearer route to measurable returns.

The difficulty is scaling such applications across departments and production units. This is where integration becomes crucial. A predictive maintenance system, for example, will have limited value if machine data cannot be connected to maintenance schedules, spare-parts availability and production planning.

The same applies to quality control. Identifying a defect is useful, but the greater benefit comes when the information can be traced back to the process responsible for it and used to prevent recurrence.

For India, the opportunity is substantial, but so is the gap between early adopters and the wider industry. The CITI–NITRA findings make clear that many companies still need basic digital systems before they can move towards integrated AI applications. Smaller enterprises will need access to affordable solutions, financing, training and opportunities to test technologies.

Industry associations, research institutions, technology providers and educational institutions will have to work together to make this transition practical. Shared platforms, demonstration facilities and industry-specific tools could help bring down costs and make adoption less intimidating for MSMEs.

AI is unlikely to eliminate the need for textile manufacturing expertise. If anything, its successful application will depend on that expertise. The technology can process information at a scale and speed that people cannot match, but it still needs to be directed towards the right problems.

The winners will not necessarily be the companies with the biggest technology budgets or the most advanced algorithms. They will be the manufacturers and their employees who know where AI can make a difference, have the data to support it, and can translate its output into better production decisions.

The race is no longer about who adopts AI first. It is about who can make it work on the factory floor, and turn it into a lasting competitive advantage.

AI is unlikely to eliminate the need for textile manufacturing expertise. If anything, its successful application will depend on that expertise. The technology can process information at a scale and speed that people cannot match, but it still needs to be directed towards the right problems. The winners will not necessarily be the companies with the biggest technology budgets or the most advanced algorithms. They will be the manufacturers and their employees who know where AI can make a difference, have the data to support it, and can translate its output into better production decisions.

Previous News