Artificial intelligence (AI) is beginning to reshape fleet management beyond conventional telematics that merely track vehicles. In India’s fragmented trucking ecosystem, where cost pressures, ageing fleets and operational inefficiencies remain persistent challenges, AI-led platforms are attempting to shift the industry from reactive monitoring to predictive decision-making. Mumbai-based Taabi Mobility Limited is among the companies advancing this shift, using large-scale data analytics to link driver behaviour, vehicle performance and operating conditions, offering fleets actionable insights aimed at reducing costs, improving safety and optimising asset utilisation.

Generally, most fleet management platforms track location, speed and unauthorised stops, making them mainly descriptive and not prescriptive. Mumbai-based Taabi Mobility Limited is changing the narrative leveraging the computing and predictive power of artificial intelligence (AI).

“Our AI solution adds value by correlating thousands of variables like driver behaviour, road conditions, load, ambient temperature, tyre age etc. and continuously learning in real time. It predicts outcomes. Moreover, traditional reports are static, while AI gets more accurate over time, adapting to different routes. Threshold alerts are not just fixed values. AI detects unusual rates of change and alerts proactively,” explained Chief Executive Officer Pali Tripathi.

Alluding to whether the AI platform only analyses data or also guides operators in real time, she explained that alerts differ by user. “Drivers get in-cabin voice alerts about tyre pressure, fatigue, collision risk etc. Fleet operators receive aggregated, actionable insights across many trucks via a live dashboard with critical exceptions highlighted,” Tripathi said.

She added that the effectiveness of AI relies on high-quality data. The control tower suggests actions like contact drivers, schedule maintenance or recommend coaching but does not fully automate vehicle control. Alert volume is configurable to prevent human fatigue.

She noted that the company’s solution also provides specific corrective actions. “A truck from Delhi to Jaipur showing left-tyre vibration and slow pressure drop triggers an alert for the driver to stop at the next halt. Fleet managers are also notified. The system identifies the issue, potential cause and suggested solution, not just the symptom,” explained Tripathi.

Tripathi contended that the fleet management sector in India is seeing multi-modal transport hubs, digitisation, improved road and waterway connectivity and better warehousing and last-mile efficiency. However, the industry is still not fully organised like in developed countries.

Taabi, she explained, is an operations intelligence platform designed to reduce total operational costs per truck by predicting issues rather than relying on fixed schedules. The system monitors vehicle behaviour, load, road conditions and tyre pressure to flag problems early.

“While fleets focus on fuel cost, tyre health directly impacts safety and performance. Fleet interest in tyre solutions is usually part of a holistic cost-reduction strategy rather than a standalone concern. A 10 percent improvement in tyre life can save crores of rupees for large fleets, making investments in platforms like Taabi worthwhile,” said Tripathi.

Companies in last-mile logistics and cement or steel transporters actively track these metrics through Taabi’s solution.

When asked about collaboration with tyre manufacturers and vehicle OEMs for data sharing, Tripathi indicated that such partnerships are still evolving and not yet fully formalised. She noted that major commercial vehicle OEMs along with tyre manufacturers already collect operational data independently for research and product development.

However, the company’s platform currently prioritises a customer-first approach, focusing on empowering fleet operators with actionable insights. Instead of directly supplying data to OEMs, the system enables fleets to use operational intelligence to hold manufacturers accountable for vehicle performance.

FROM GROUND UP

The company currently serves around 1,300 fleet operators across India. Growth is measured in assets deployed rather than just customers, as a single vehicle may use multiple solutions such as OBD devices, video telematics and fuel monitoring systems. Average deployments are about 272 assets per fleet with ranges from 50 to 4,000 assets.

The company has recorded 130–132 percent year-on-year growth, largely driven by expanding deployments within existing customers.

Nonetheless, Tripathi explained that the primary hurdle for the company was building trust in a completely new category of product. “Since fleets had operated for decades without such technology, convincing operators that the platform could deliver measurable value was difficult. We therefore positioned AI not as a replacement for human judgment but as a tool that enhances decision-making, highlighting hidden operational costs such as tyre wear, vehicle inefficiencies and the financial impact of driver behaviour,” she averred.

Another major challenge was the data ‘chicken-and-egg’ problem. AI systems require large datasets to function accurately, but fleet operators were hesitant to adopt the platform without proof of performance.

Although the company had access to global data, it began collecting India-specific road, load and operational data three to four years before launch to train its models. Early adopters and pilot customers were told transparently that the system would improve as more local data was gathered.

A further complexity involved customising the user interface and experience for different sectors. Construction fleets, buses, trucking companies and enterprise operators such as ambulance services all required different dashboards and operational insights. As a result, persona-based interface design became an important part of product development. When discussing adoption among smaller fleet operators, Tripathi noted that fleets with 5–20 trucks typically adopt the solution through larger enterprises or ecosystem partners.

To improve accessibility, the company offers subscription-based pricing similar to mobile phone plans, avoiding large upfront costs. The base plan provides simple alerts and WhatsApp-style notifications. More advanced features are included in Gold and Platinum plans, which deliver deeper analytics and operational insights.

IMPLEMENTATION

Addressing the challenge of deploying AI-based fleet monitoring on older commercial vehicles, Tripathi noted that a large share of India’s truck and bus fleet is 10–20 years old, meaning many vehicles lack factory-fitted OBD or tyre pressure monitoring systems (TPMS).

“To overcome this, we use a matchbox-sized device that plugs into aftermarket OBD ports typically available on trucks manufactured after 2000. The device captures key operational data such as engine performance, speed, RPM, load conditions and fuel consumption,” she noted.

For older vehicles without such capabilities, additional hardware such as fuel tank sensors are installed to track consumption and detect issues like fuel theft or reverse draining. The system can also monitor gensets and auxiliary equipment, while video telematics can be added when required.

Tripathi explained that this approach can actually make the platform particularly valuable for older fleets, enabling both small and large operators to access AI-driven monitoring and predictive maintenance.

The platform also supports intelligent cameras inside the cabin and facing the road, enhancing driver behaviour monitoring and safety analytics. For tyre monitoring, fleets can use external TPMS units, although these are relatively expensive. As a cost-effective alternative, the system derives proxy performance indicators from OBD data and telematics to estimate tyre health and vehicle performance.

“In minimal deployment scenarios, even a driver’s smartphone can provide basic telematics functions such as GPS tracking, route adherence, geo-fencing and idle detection, enabling gradual adoption of digital fleet management tools,” noted Tripathi.

The platform follows strict data security and privacy standards. All operational data is end-to-end encrypted using AES-256 and stored on cloud infrastructure within India through Microsoft Azure. Fleet data remains private to each operator, meaning one fleet cannot access another’s information.

Internally, only aggregated data is used for model training without exposing raw fleet-level details. Any external data sharing is tightly controlled and compliant with India’s Digital Personal Data Protection framework.

MARKET DEMAND

The company views the retrofit segment as the largest opportunity in India, as most commercial vehicles are older and new truck sales represent only a small share of the total fleet. Its strategy is to democratise access to fleet intelligence by enabling AI-driven monitoring on existing vehicles rather than waiting for fleet modernisation.

“We also see growing relevance in commercial EV fleets, particularly in last-mile delivery networks. Our platform acts as an intelligence layer for mixed fleets transitioning from diesel to electric vehicles, helping operators evaluate return on investment, identify suitable routes for EV deployment and manage operational economics. Vehicle-agnostic solutions such as video telematics can be deployed across cars, vans and EV delivery vehicles,” Tripathi contended.

Rather than relying solely on hardware innovation in tyres or vehicles, the company focuses on AI-driven insights derived from sensor data. “Continuous monitoring allows our system to predict performance issues and recommend interventions. The platform functions as an operational intelligence layer, offering voice-based guidance for drivers, cost-optimisation insights for fleet owners and operational support for fleet managers,” averred Tripathi.

Devices installed in vehicles perform round-the-clock monitoring of engine, fuel, tyre and other operational parameters, delivering predictive alerts and actionable insights. By simplifying complex data into clear recommendations, the AI platform aims to improve fleet efficiency, reduce costs and enable smarter operational decisions.

As generative AI transforms the way consumers and businesses discover products, tyre manufacturers face a fundamental shift in digital marketing. Roshan Mohan, Co-Founder and CMO at FlowBlinq, and Founder of PCG argues that the next battle will no longer be fought on search engine rankings but on whether AI systems choose to recommend a brand in the first place.

For more than two decades, tyre manufacturers have refined their digital strategies around a familiar formula: optimise websites for search engines, invest in paid advertising, strengthen dealer networks and build visibility through reviews and comparison platforms. Success depended largely on securing a prominent position on Google’s search results.

That formula, however, is beginning to change.

The rapid adoption of generative artificial intelligence (AI) platforms such as ChatGPT, Gemini and Claude is reshaping how consumers search for information, compare products and make purchasing decisions. Rather than browsing multiple websites, customers are increasingly asking AI assistants to recommend the most suitable product based on their specific requirements.

For tyre manufacturers, this represents far more than another digital marketing trend. It fundamentally changes how products are discovered.

According to Roshan Mohan, Co-Founder and CMO at FlowBlinq, and Founder of PCG, companies that continue treating AI as simply another marketing channel risk missing a much larger transformation. “The customer journey for tyres has already started bending around AI, and the change over the next three to five years won’t be a redesign of the funnel; it’ll be a shift in where the funnel begins,” he says.

FROM SEARCH ENGINES TO AI CONVERSATIONS

Historically, buying tyres has been an information-intensive process. Consumers often compare technical specifications, dealer recommendations, user reviews, pricing and compatibility before making a purchase. Search engines have traditionally served as the starting point for that journey.

Generative AI is simplifying this process dramatically.  Instead of opening multiple browser tabs and manually comparing products, motorists can simply ask an AI assistant for recommendations based on vehicle type, budget, driving conditions and performance priorities. The AI then synthesises information from numerous sources into a single response. “What changes for the buyer is effort, not intent. They still want the right tyre for their car and budget, but instead of researching options and making the final comparison themselves, they are increasingly describing their needs to AI and letting it identify the best solution,” Mohan explains.

This shift effectively transfers much of the research process from the consumer to the AI model.

Industry forecasts suggest this transition is already underway. Gartner predicts traditional search engine volume will decline by 25 percent by 2026 as generative AI absorbs many queries that previously began with conventional search engines. Meanwhile, Checkout.com’s research indicates that consumers are embracing AI-assisted purchasing faster than many businesses are preparing for.

For tyre companies, the implication is profound: visibility may increasingly depend not on appearing first in search results but on being recommended within AI-generated answers.

THE RISE OF AI VISIBILITY

Search engine optimisation (SEO) has long centred on improving rankings through keywords, backlinks and domain authority. AI discoverability, Mohan argues, follows a very different logic.

“Traditional SEO was about ranking, winning a position on a page of 10 blue links. AI visibility, or Generative Engine Optimisation (GEO), is about being the answer rather than a link to the answer,” he says.

Unlike traditional search engines, large language models evaluate whether product information is sufficiently trustworthy, structured and complete before referencing it. If they cannot confidently interpret a manufacturer’s data, the brand may simply disappear from the recommendation altogether.

This places far greater importance on machine-readable product information than on conventional search optimisation. At the same time, AI systems are looking beyond a company’s own website to understand whether a brand is trustworthy. This makes it important for brands to have a presence across credible, independent sources, where editorial PR and genuine reviews can play a key role, rather than advertisements or advertorials. AI systems bring together these trust signals from multiple sources and present them to users in one place. This means the decision-making journey is increasingly shifting to the AI chat, where consumers can get a more comprehensive view before making a choice. Brands that build credibility across trusted sources will therefore be better placed to influence how AI systems recommend them.

Structured specifications, consistent product descriptions, schema markup and clearly organised technical information become essential because AI systems rely on these elements when generating responses.

FlowBlinq has developed what it describes as 17 Generative Engine Optimisation pillars to assess whether brands are sufficiently prepared for AI discovery. These include structured data quality, technical completeness and AI crawlability.

Perhaps more significantly, Mohan believes many companies have little understanding of how frequently AI platforms mention their products – or whether they are mentioned at all.

“Our citation tool runs a brand across ChatGPT, Claude and Gemini and shows, prompt by prompt, whether the brand gets cited, where it loses out to a competitor and where it’s simply absent from the answer altogether,” he adds.

TECHNICAL ACCURACY BECOMES A COMPETITIVE ASSET

Tyres differ from many consumer products because purchasing decisions depend heavily on technical specifications. Load index, speed rating, rolling resistance, wet grip, tread pattern and vehicle compatibility all influence suitability. Inaccurate recommendations can have genuine safety implications.

Mohan believes this makes structured product information particularly important for the tyre industry. “When product data is thin, a model doesn’t refuse to answer; it defaults to the brand it has the most confident, well-structured information about,” he adds. He warns that this tendency naturally favours manufacturers with richer digital product catalogues rather than necessarily those with superior products.

FlowBlinq’s research suggests considerable room for improvement. According to the company’s findings, 62 percent of Indian brand websites provide product descriptions that are insufficiently detailed for AI systems, while more than half lack product codes needed for accurate identification.

For tyre manufacturers, the solution is relatively straightforward but frequently overlooked.

Rather than embedding specifications within downloadable PDF brochures or image-based catalogues, companies should publish technical information directly on webpages in formats that AI systems can easily interpret.

Equally important is the broader digital reputation surrounding a brand. Mohan notes that AI systems increasingly rely on trusted third-party sources – including established news publications, Wikipedia and community platforms – to validate manufacturer claims before making recommendations.

AI ENTERS FLEET PROCUREMENT

The implications extend well beyond retail consumers. Business purchasing decisions often involve lengthy comparisons of performance, lifecycle costs, regulatory compliance and operational efficiency – precisely the type of structured analysis that generative AI performs well.

According to Mohan, procurement teams, fleet operators and original equipment manufacturers (OEMs) may adopt AI-supported purchasing even faster than retail buyers. “B2B tyre buying was never going to be immune to this, and it may move faster than consumer purchasing because procurement teams are exactly the audience generative AI tools were built to serve,” he says.

A fleet manager could ask AI to compare total cost of ownership across several tyre brands. An OEM purchasing team might request suppliers meeting specified rolling resistance or durability thresholds.

In such scenarios, manufacturers lacking accessible technical documentation risk exclusion before human procurement teams even begin formal evaluation. “The practical response isn’t a new sales deck. It’s making sure spec sheets, compliance documentation and comparative data exist in formats a model can read and trust,” Mohan says.

AI WILL ADVISE, BUT HUMANS WILL STILL DECIDE

While AI is poised to transform product discovery, Mohan believes the actual purchase decision will remain firmly in human hands – at least for high-value, safety-critical products such as tyres.

“I’d separate ‘AI helping me decide’ from ‘AI deciding for me’, because consumers still are the final decision makers,” he says.

Recent consumer research supports this view. While surveys indicate growing confidence in AI agents handling routine shopping tasks, willingness declines sharply when AI is expected to complete purchases autonomously. Most consumers remain comfortable with AI conducting research, comparing alternatives and shortlisting products but prefer to approve the final transaction themselves.

Tyres, Mohan argues, naturally fall into the category where human oversight will continue to matter.

“It’s a purchase people make infrequently, it carries real safety implications, and it typically involves a meaningful amount of money,” he says.

Consequently, AI is likely to dominate the research phase – evaluating specifications, warranty terms, prices and dealer options – while the final purchase decision remains with the customer.

However, one area where agentic commerce could quickly gain traction is in connecting customers directly with dealers. Rather than merely recommending a tyre, future AI assistants may also identify nearby retailers with available stock and book installation appointments automatically.

BECOMING AI-READY STARTS WITH THE BASICS

One of the most striking aspects of Mohan’s assessment is that the industry’s biggest challenge is not technological sophistication but digital housekeeping.

“It’s data, overwhelmingly, and it’s more basic than most companies expect,” he says.

FlowBlinq’s audits suggest that many corporate websites still lack the fundamental structure AI systems require. According to the company’s research, 91 percent of audited websites failed to provide clear information explaining their product catalogues in a way that AI could

understand. Even more concerning, nearly half were unintentionally preventing ChatGPT’s web crawler from accessing their websites because of security settings or plugin configurations.

“These aren’t strategic gaps; they’re operational oversights, and they’re fixable in weeks, not years,” Mohan claims.

For tyre manufacturers, this means that substantial improvements may not necessarily require major investments in new technology. Instead, they require a systematic review of how product information is organised, published and made accessible to AI systems.

Mohan also believes the next phase of digital readiness will involve preparing websites for agentic commerce by enabling real-time inventory visibility and ensuring AI systems can interact directly with product databases.

MEASURING RETURN BEYOND TRADITIONAL SEO

Digital marketing budgets have historically focused on search advertising, social media campaigns and marketplace optimisation. As AI-driven referrals grow, Mohan argues that businesses should begin allocating dedicated budgets towards AI discoverability.

“Yes, and the case for it is measurable rather than speculative now,” he says. Rather than relying solely on website traffic or keyword rankings, he believes organisations should monitor a different set of performance indicators.

Among the most important are how frequently AI systems cite a brand when responding to relevant queries, whether those citations are accurate and whether visitors arriving through AI recommendations convert differently from those originating through conventional digital channels.

Adobe’s Digital Insights research suggests AI-generated referrals are not only increasing rapidly but also producing stronger conversion rates than traditional referral sources. According to Mohan, this reflects the higher purchase intent of consumers who have already completed much of their evaluation through AI before visiting a manufacturer’s website.

TRUST WILL DETERMINE INFLUENCE

The emergence of AI recommendations inevitably raises questions about transparency. If AI systems become influential in shaping purchasing decisions, how can brands improve visibility without manipulating results?

For Mohan, the answer lies in accuracy rather than optimisation. “The honest answer is that AI-powered recommendations only work for a brand in the long run if they’re accurate, because these systems increasingly get checked,” he explains.

He believes manufacturers should resist the temptation to game AI systems through exaggerated marketing claims.

Instead, success will depend upon providing complete, verifiable product information that allows AI to make fair comparisons based on genuine performance characteristics.

“So the lever isn’t gaming a model into over-recommending you. It’s making sure that when a model compares your tyre honestly against a competitor on wet grip, rolling resistance or price, your data is complete enough that you win the comparisons you’re actually built to win,” he says.

In his view, transparency is not a constraint on AI marketing but its most durable competitive advantage.

FROM RECOMMENDATIONS TO TRANSACTIONS

The next evolution extends beyond recommendations. Emerging protocols are enabling AI systems to communicate directly with commerce platforms, inventory databases and pricing systems, allowing them to perform increasingly sophisticated purchasing tasks.

According to Mohan, this represents a significant opportunity for tyre manufacturers and dealers.

“The interesting shift is that AI agents are starting to interact with commerce systems directly... rather than just reading a webpage and stopping there,” Mohan says.

Once connected to live inventory systems, AI assistants could recommend the exact tyre that fits a customer’s vehicle, confirm stock availability at nearby dealers and compare prices in real time. Now, it can also make purchases directly from the chat window. This is something FlowBlinq is uniquely positioned to address as well.

Rather than generic recommendations based on previous purchasing patterns, personalisation could become highly contextual – considering vehicle compatibility, driving conditions, current inventory and even maintenance priorities.

However, Mohan cautions that these benefits will only be realised by organisations whose internal systems can support such interactions. Manufacturers and retailers will need modern, connected back-end infrastructure capable of sharing real-time inventory and pricing information with AI platforms.

ENGINEERING PRODUCTS – AND ENGINEERING DISCOVERABILITY

Looking ahead, Mohan does not believe AI will replace product quality as the defining competitive factor. Instead, he sees AI readiness becoming an equally important complement to engineering excellence.

“Product quality will always be table stakes; nobody wins on AI visibility with a mediocre tyre,” Mohan says. Yet he argues that superior products alone may no longer guarantee commercial success.

As purchasing journeys increasingly begin with AI conversations rather than search engines, brands that fail to present their technical information in formats AI systems can retrieve and trust may simply disappear from consideration.

“The winners will be the manufacturers who treated AI readiness as seriously as they treat product engineering,” Mohan says. He returns to Gartner’s prediction of declining traditional search volumes not as a warning but as an indication of how rapidly digital discovery is evolving.

“The discovery layer is moving to AI faster than most manufacturers’ data infrastructure is moving with it,” he says.

His concluding observation perhaps best captures the industry’s emerging challenge.

“A brand can make the best tyre in its category and still lose the sale simply because it was invisible in the one conversation the buyer had before deciding. That’s a genuinely new way to lose, and avoiding it is now a core marketing responsibility, not a technical footnote,” Mohan says.

AI mobile data capture company Anyline has released the latest version of TireBuddy, a smartphone-based system for automotive tyre inspections. Version 1.8 introduces fully automated sidewall capture that removes human variability from data collection. The tool has already helped service teams achieve faster, more uniform inspections over the past year, leading to increased tyre sales and stronger customer trust.

The automated mechanism uses on-device guidance that evaluates each image against four criteria: full sidewall detection, sharpness, proper distance and angle and overall clarity. This real-time feedback minimises redo scans by guiding technicians to capture optimal images immediately. The system addresses common challenges in busy service bays where accuracy often suffers due to varying experience levels.

Standardisation of inspection quality is a primary benefit, as consistent results are achieved regardless of who holds the phone. This removes dependency on technician skill or training duration. New or seasonal staff can perform scans confidently from day one without extensive instruction. The automated capture now serves as the standard protocol for all inspections across locations and shifts.

Additional features include tyre mismatch alerts that flag size discrepancies, automated email reports to back-office systems and a redesigned results screen consolidating sidewall information and tread measurements. With hundreds of thousands of annual inspections, this update reinforces TireBuddy's role in modernising tyre service operations.

Lukas Kinigadner, CRO, Anyline, said, “A shop is only as consistent as its least experienced inspector. Automated sidewall capture gets every scan to the same standard, so teams can stop treating inspection quality as a variable.”

Epson, a global leader in SCARA robot manufacturing, has unveiled its next-generation industrial robotics portfolio at Automation Expo Mumbai 2026. The newly introduced lineup features the high-end CX-A Series 6-axis robots, the LS-C Series SCARA robots, the RC+ 8.0 programming software and the advanced SafeSense safety technology, all designed to address diverse manufacturing applications such as pick-and-place, precision assembly, parts transfer and material handling.

The new offerings significantly expand Epson’s existing industrial robotics family, which already includes the 6-axis C-Series and SCARA T-Series and LS-Series models with payloads ranging from 3 to 20 kilogrammes. With the addition of the CX-A and LS-C Series, manufacturers across various sectors can achieve heightened productivity, flexibility and operational efficiency. The CX-A Series is engineered for complex tasks with a payload capacity of up to seven kilogrammes and a reach of 900 millimetres, available in IP67, cleanroom and ESD variants, while the LS-C Series provides a compact SCARA platform with a 50-kilogramme payload, a 1,000-millimetre reach and cycle times as fast as 0.298 seconds.

Complementing the hardware, the RC+ 8.0 software offers an integrated environment for programming, simulation and system management, facilitating faster automation deployment with support for Visual Studio and C++ development. Additional efficiency features include enhanced diagnostics, OPC UA, GUI builder and safety functions, alongside co-creation tools like Library Builder and RC+ Extension. Meanwhile, the SafeSense technology promotes safer human-robot collaboration by incorporating Safety Limited Speed and Safety Limited Position functions, which can potentially reduce the need for extensive safety fencing and thereby increase operational flexibility.

With over four decades of industrial robotics expertise and more than 200,000 robotic arms deployed globally, Epson continues to drive operational excellence for businesses. Attendees at Automation Expo Mumbai 2026 have the opportunity to view live demonstrations of these solutions and consult with Epson specialists about transforming their manufacturing operations.

Siva Kumar, Sr General Manager – Sales and Marketing, Epson India, said, "India is rapidly emerging as a global manufacturing hub, and automation will play a pivotal role in shaping its future. With our new industrial robot lineup and RC+ 8.0 platform, Epson is delivering the speed, precision and intelligence manufacturers need to compete in an increasingly dynamic marketplace. We remain committed to enabling businesses to accelerate automation adoption and build smarter, more agile and globally competitive manufacturing operations."

A consortium of Fraunhofer institutes has reached a key milestone in the Technology Platform for Tire Abrasion and the Identification of its Emissions in Road Traffic (TERIS) project, moving closer to establishing standardised laboratory methods for generating, analysing and predicting tyre wear.

The project, led by the Fraunhofer Institute for Structural Durability and System Reliability LBF, together with Fraunhofer ICT, Fraunhofer IGD and Fraunhofer IWM, aims to provide the tyre industry, testing organisations and environmental agencies with reliable and practical laboratory procedures for assessing tyre abrasion emissions.

The first project milestone has been completed following a successful review by an advisory board comprising industry experts.

The consortium has developed reference methods for tyre abrasion, particle analysis, tribological modelling, artificial intelligence-based surface analysis, a laboratory test bench concept, accelerated ageing techniques and volatile organic compound (VOC) detection.

According to the consortium, combining different particle collection and measurement techniques enables more precise analysis of both airborne and deposited tyre wear particles. At the same time, tribological models have been developed to better understand the relationship between loading conditions, material properties, surface structures and particle formation, allowing real-world tyre wear processes to be replicated under laboratory conditions.

Researchers have also developed a specialised test chamber for accelerated ageing, enabling tyre samples to be exposed to controlled environmental conditions before evaluating their abrasion behaviour.

Another development is an optical detection system that uses artificial intelligence to identify and classify surface structures. The system has been validated using substitute materials and is expected to be applied to rubber samples during the next phase of the project.

The consortium has also designed a laboratory test bench that combines multiaxial loading, controlled generation of tyre wear particles, targeted particle collection and integrated optical sensors within a single testing platform.

In addition, the project combines accelerated weathering with chemical analysis of volatile organic compounds released from tyre abrasion to assess the environmental impact of tyre wear particles.

The researchers said the work will provide the foundation for faster and more practical laboratory evaluation of new rubber compounds. The resulting methods are intended to help tyre manufacturers reduce emissions, accelerate product development and support compliance with the requirements of the Euro 7 standard.

At Fraunhofer IWM, researchers focused on refining tribological wear models and friction surface concepts to simulate particle formation under controlled laboratory conditions. The institute designed a parameterisable wear test that studies friction between plate materials and model surfaces with different structures, enabling researchers to investigate the mechanisms responsible for particle generation.

Initial findings indicate that tyre wear results from multiple interacting mechanisms rather than a simple relationship between particle emissions and factors such as speed, contact force or temperature. The researchers collected and analysed particles across a wide range of sizes during the study.