As the textile industry continues to modernise, AI will become an integral part of how manufacturers, suppliers and service providers interact in the future, but let us first consider how things are today.

While the speed of textile yarn spinning systems has undergone a remarkable transformation over the past 50 years, fuelled by technological advancements, automation and material innovations, there is still a considerable amount that can go wrong on any spinning machine from day to day.

Ring spinning frames and rotor or air-jet spinning machines combine precision mechanical, pneumatic, and electronic systems. Each machine consists of thousands of parts, ranging from tiny bearings, cots, aprons, rollers, and sensors to large metal frames and automated control units. Added to this are the high number of repeating units, most notably, the spindles and rotors, often across hundreds of positions along the length of the spinning units.

Wear and tear
The latest ring spinning machines can now achieve speeds of up to 30,000 rpm, while rotor spinning has reached 160,000 rpm and air-jet spinning systems—the fastest commercial systems—produce yarns at speeds exceeding 500 m/min. Inevitably, at such speeds, there is plenty of unavoidable wear and tear.

According to the Zurich, Switzerland based ITMF (International Textile Manufacturers Federation) there are approximately 220-250 million ring spindles and an additional 8-10 million open-end rotors currently in operation worldwide, with numbers fluctuating slightly year by year, based on new installations, machine retirements and shifts in global textile production patterns.

Given the wide number of both machine manufacturers and successive machine generations, ensuring they all run optimally through their operational life is no small endeavour.

ESSENTIALorder
Two of the leading specialists in spinning technology, Rieter and Saurer, are both still headquartered in Switzerland due to their historical roots, albeit increasingly with business activities across Asia.

Winterthur-headquartered Rieter has acquired several smaller companies specialising in specific components for spinning machinery over the years. These include Accotex, Bräcker, Graf, Novibra, Suessen and Temco, who are now all members of the Rieter Components Division, which achieved sales of CHF 247.6 million in 2024, making a significant contribution to the company’s overall sales last year of CHF 859.1 million.

Rieter’s advanced ESSENTIALorder platform now offers more than 14.5 million spare parts online and is available 24 hours a day, seven days a week.

Due to AI advances, the webshop is now able to remotely access the machine configurations of participating customers and provide a personalised order experience, enabling spinning mills to optimise their internal stock levels with just a few clicks.

Vardhman endorsement
Vardhman—India’s largest vertically integrated textile manufacturer, with ten spinning mills across the country—is one enthusiastic user of Rieter’s ESSENTIALorder platform.

Vardhman’s central purchasing team uses a monitoring system and bases its purchasing decisions on real-time data. Rieter machines are running in all ten of its spinning mills in multiple locations and the Vardhman purchasing team has established an optimised process for procuring Rieter OEM spare parts across all of them.

Rieter’s aftersales team introduced ESSENTIALorder to two of Vardhman’s units in Northern India back in 2019 and after successful trials, it was introduced to the remaining group units.

ESSENTIALorder enables major customers like Vardhman to check the price, availability, and lead times for all parts before placing an order. Customers can download quotes, order confirmations, and invoices on demand, while order and shipping status can be tracked in real time. A built-in customised visual catalogue allows users to quickly identify and select the correct machine parts. Additional key benefits include ensuring compatibility with equipment design and simplifying recurring orders through the order history function.

Customers can further assign roles allowing different functionalities to different users and choose who can access and create orders, set quantity limits and track delivery times.

“As a user of ESSENTIALorder, our order management process has become much easier and our different spinning mills are optimising stock levels effectively,” says Neeraj Jain, joint managing director at Vardhman. “ESSENTIALorder is a user-friendly and reliable platform because spare parts for Rieter spinning machines can be ordered around the clock in just a few clicks.”

Automated, digital and smart
“The future of the spinning industry is automated, digital and smart,” adds Thomas Oetterli, CEO of the Rieter Group. “As mills struggle to hire and retain workers, automation technology is essential in order to ensure consistent and efficient operations. Rieter has stepped up its R&D activities to fully automate the value stream by 2027 through autonomous transport systems and collaborative robotics. Rieter’s ESSENTIAL digital spinning suite will be the command-and-control centre of smart and high-performing mills that lower cost and maximise returns. Customers will be able to fully focus on their yarn business by outsourcing their operations to Rieter technology and expertise.”

Guaranteed availability
Saurer, headquartered in Arbon, Switzerland, meanwhile holds almost 300,000 original parts for all of its product lines and all machine generations in stock.

“Our customers enjoy the benefits of life-long technical services and the guaranteed availability of original parts for decades,” says Christian von Kannen, Saurer’s vice-president of after sales solutions. “They can also count on support from our technological consulting team across the globe and we have a worldwide network of Customer Service Centres and Saurer agents to support the businesses of our customers.

“Saurer original parts have a long service life, ensure trouble-free production in plants and reduce machine downtime. Naturally, our products come with specified guarantee periods, information concerning expected service life and application recommendations. Many parts are protected by patents to ensure that our customers can benefit from them exclusively.”

Texparts
Saurer also incorporates the independent components businesses Texparts, which celebrated its 25th anniversary in 2024.

“Since 1999, we have been working on innovative solutions to increase the efficiency, productivity and yarn quality of our customers,” says Texparts general manager Marc Kallenberg. “Our main aim is to make machine operation as simple as possible and to improve spinning quality by preventing setting errors and ensuring yarn quality over the entire length of the machine.

“We manufacture spindles, drafting systems, bearing units and top rollers with maximum precision at our advanced plant in Fellbach, Germany, guaranteeing a long service life for these components. Our customers primarily benefit from improved yarn quality, fewer thick spots in the yarn and optimised yarn quality values.”

Competitive advantage
Among recent Texparts innovations are the Eshape spindle and the Spinnfinity underwinding system, which can be combined to optimise energy consumption by up to six per cent while also reducing operating costs through simpler and faster cleaning. This lowers production costs and provides customers with a clear competitive advantage.

Since 2024, Spinnfinity has been available in three different options to additionally fulfil the needs of wool and coarse yarn spinning mills, contributing to a more sustainable manufacturing process by reducing yarn waste.

Texparts bearings are another success story, and the latest RB 105 rotor bearing has been very well received by the market as a complete package.

“Despite widespread competition on price as well as plagiarism and copies, many years of experience and expertise have enabled us to keep gaining market share and we are constantly broadening our product range,” says Kallenberg.

Knitting machine complexity
Arguably even more complex than spinning systems, are knitting machines, and while the number of parts depends on the type, size and complexity of the machine, on average, flatbed knitting machines have between 2,000-4,000 parts and circular knitting machines between 4,000-7,000 parts.

Major parts needing to be monitored and regularly replaced include cams, yarn feeders, motors, control units, sensors, and especially needles and their sinkers.

High speed circular knitting machines can be equipped with thousands of needles, along with their accompanying sinkers, and wear and tear on these components has been estimated to account for around 45 per cent of all faults in production.

This is another area where AI is already making a difference.

Needle damage
In recent years, the identification of worn or damaged needles in industrial knitting machines has become an increasingly important focus for textile manufacturers seeking to maintain consistent quality and avoid costly downtime.

Traditionally, needle inspection was a manual process, relying on the skill and experience of technicians to identify signs of wear or malfunction. This approach, while effective in the hands of experienced operators, is time-consuming and subject to human error, especially in high-speed, large-scale production environments where thousands of needles are in use at once. However, technological developments are now offering more accurate, efficient and automated ways to detect worn needles before they cause defects in the fabric or bring production to a halt.

At the forefront
Several companies are at the forefront of developing and supplying technologies for identifying worn needles and improving maintenance in industrial knitting machines.

Germany’s Mayer & Cie, for example, has been actively integrating digital solutions and machine monitoring technologies into its equipment for many years and is increasingly looking at sensor-based condition monitoring, which can contribute to the early detection of needle wear.

A second German leader in knitting technology, Karl Mayer Group, has developed various digital solutions under its KM.ON brand, which includes smart maintenance and monitoring tools. While not needle-specific, the platform supports predictive maintenance strategies that can incorporate needle condition tracking.

As a leader in seamless knitting production, Italy’s Santoni has already integrated advanced automation features into its machines and continues to work on predictive maintenance technologies, including systems that monitor needle condition indirectly via machine performance.

Off-the-shelf solution
At the ITMA 2023 textile machinery exhibition in Milan, Canmartex—the Barcelona, Spain-based owner of the Jumberca machine brand—launched an off-the-shelf AI system aimed at completely eliminating needle and sinker problems on existing knitting machines. Developed in collaboration with the industrial automation institute Eurecat, also based in Barcelona, it is being marketed by a new spin-off subsidiary, ASB.

“Our system consists of a number of programmes powered under the Fabric Brain umbrella, that address various common problems in circular knitting,” explains ASB managing director Enric Marti. “It can be supplied as a stand-alone system for circular knitting, both on new machines and retrofitted to existing models.”

“Most importantly, the system can identify any defect in the condition of every single needle and sinker, prior to starting up the machine, rather than encountering the fault during production,” adds Xavier Planta Torralba, industrial technological area director at Eurecat. “It is an opportunity to truly move from analogue to digital and Industry 4.0 in circular knitting.”

Over a thousand samples of fabrics with defects were analysed to identify the cause of the defects in the development of the new AI system and it was ascertained that 41 per cent of the defects were caused by wear, breakage or twisting of the needles and sinkers.

All of these defects can now be avoided prior to production due to the predictive needle and sinker control systems jointly developed by Canmartex and Eurecat.

Wear analytics
Groz-Beckert is the world’s leading supplier of industrial machine needles, precision parts and fine tools, distributing more than 50,000 products for the knitting industry.

As such, its customer portals are inevitably critical in enabling customers to quickly find the right products for their applications.

“In addition to comprehensive product information and catalogues, technical data, needle recommendations and training content, our portals offer a web shop, personal customer accounts with various evaluation and ordering options and direct contact points,” explains Eric Schöller, a member of the Groz-Beckert executive board. “The portals are available in dozens of languages, optimised for mobile use and available around the clock.”

Groz-Beckert is now exploring smart needle technologies and wear analytics, being well positioned to contribute to developments in identifying and managing worn needles, especially through its innovation in materials and manufacturing tolerances.

Groundwork
While full automation of needle wear detection is still an emerging area, the groundwork being laid by these firms will shape the next generation of intelligent, self-monitoring knitting machines.

One of the most notable advancements has come through the use of machine vision systems which exploit high-resolution cameras and sophisticated image processing software to inspect the condition of each needle as the machine operates. By capturing detailed images in real time and comparing them against a reference of what a properly functioning needle should look like, these systems can automatically flag anomalies such as bent hooks, cracked shanks or excessive wear. This kind of automated visual inspection enables early detection of problematic needles, often before any visible issues appear in the knitted fabric.

Detecting subtle changes
Complementing machine vision, sensors integrated into the knitting machines are also playing a key role. Vibration and force sensors, in addition to acoustic monitoring devices, can detect subtle changes in machine behaviour that may indicate a worn or broken needle. A slightly altered vibration pattern, for instance, may suggest that a needle is not engaging the yarn correctly or is rubbing against other components. These changes, while imperceptible to the human eye or ear, can be captured and analysed using AI algorithms trained to recognise the early signs of wear. This predictive approach is increasingly being seen as a game-changer for maintenance planning and quality assurance in knitting operations.

Another area of development is in the use of digital twins—virtual replicas of the knitting machine that allow manufacturers to simulate and monitor the performance of individual parts, including needles. By correlating machine data with the digital model, it becomes possible to identify which specific needles are deviating from expected performance parameters. This targeted diagnosis significantly reduces the need for blanket inspections and enables technicians to focus their efforts where they are most needed.

RFID
The use of RFID (radio frequency identification) technology to track needle lifespan has also been explored. While not yet widespread, the idea is to embed micro-tags onto the needle carriers, allowing the machine to log usage time and generate alerts when a needle approaches the end of its expected service life. When combined with the other methods, this provides a comprehensive picture of needle condition and performance over time.

Together, these innovations are pushing the industry towards more intelligent, automated maintenance systems that minimise disruptions and preserve fabric quality.

As these technologies become more affordable and integrated into standard knitting machine designs, the identification of worn needles is expected to shift from a reactive process to a proactive and even predictive one—ensuring smoother operations and greater production reliability for textile manufacturers worldwide.

Predictive maintenance
Overall, one of the most significant areas where AI will have an impact on the management of textile machinery spare parts is in predictive maintenance and inventory planning. Through the use of machine learning algorithms and real-time data from sensors embedded in textile machinery, AI can anticipate when specific parts are likely to fail or degrade.

Rather than relying on scheduled maintenance or reacting to breakdowns, manufacturers can move towards a predictive model that automatically triggers the ordering of required components before a failure occurs. This shift reduces downtime and ensures machinery is kept in optimal condition, which is particularly crucial in the highly competitive textile industry where delays can lead to significant losses.

Demand forecasting
AI will also enhance demand forecasting for spare parts.

Traditional inventory management relies heavily on historical usage data and human judgement. AI, however, can process vast amounts of data from multiple sources—such as production schedules, environmental conditions and usage patterns—to make highly accurate predictions about future needs. This level of insight allows suppliers to optimise stock levels, avoid overstocking or shortages, and provide faster, more reliable service to their customers.

AI-powered visual recognition systems and digital twins can now enable maintenance personnel to identify components quickly by scanning them with a mobile device. These tools not only simplify the ordering process but also reduce errors and improve first-time fix rates.

Furthermore, AI is expected to streamline the logistics and distribution of parts. By integrating AI into supply chain management systems, companies can dynamically reroute shipments based on real-time traffic, weather conditions and customer urgency. This responsiveness helps ensure that the right part reaches the right place at the right time. In some cases, AI might also be used to automate reordering processes, linking machinery monitoring systems directly to supplier inventories, and thus reducing administrative overhead.