Intangles is on a quest to change the way mobility ecosystem players track their efficiencies. The company has been providing data driven insights on vehicle health, driver behaviour profiling, operations intelligence, fuel management and AdBlue tracking to fleet operators and commercial vehicle OEMs using the digital twin paradigm and machine learning, with a focus on timely predictive analytics. Rajesh Rajgor speaks to the Founders of Intangles in an exclusive interaction.
Predictive analytics is no longer only the domain of mathematicians and statisticians, thanks to the increasing availability of interactive and user-friendly software. Despite the fact that predictive analytics has been around for decades, its demand is quickly gaining momentum globally. In fact, it is increasingly being used by businesses to improve their bottom-line and gain a competitive advantage across the globe. In this segment, Intangles’ real-time analytics and prognostic intelligence are being used to help create the world’s largest archive of repair strategies. Fleet operators can thus predict which parts of their vehicle would need maintenance, and schedule servicing or repairs as per the severity, even before a problem arises.
A system that keeps an eye on all the aggregates while also factoring in various environmental circumstances that may impact the performance of a specific part can prove immensely beneficial for its operational efficiency. Fleet owners, often at the mercy of drivers and garage mechanics, can now have more control over their vehicles through preventive steps, thanks to the Intangles Connected Ecosystem. The Intagles solution can render an abundance of curated data on vehicle performance characteristics in real world usage settings for the OEMs. This enables OEMs to improve the service experience for fleets in the backdrop of better parts planning, time to repair and proactive assistance in preventing large breakdowns. The end result is more up-time, improved efficiency and healthier top-lines for fleet operators.
Digital Twin Concept
The earliest model of Intangles was a child tracking solution, which Anup came up with after his son got separated from the rest of the group during a work picnic. Anup then contacted Jayshri Patil, an embedded systems expert with many years of expertise, to begin developing the child tracking gadget. Soon after, Neil Unadkat, CTO and Co-Founder joined the team to assist with the solution’s web interface. Aman Singh, Head of Analytics and Co-Founder who was following his passion for automotive systems as part of his thesis at PICT, Pune, was introduced to the founding team by Neil. He’d been working on collecting data from vehicles using the OBD interface and running diagnostics. Soon they pivoted from the nascent idea of a child tracking solution to a more complex predictive analytics solution for the mobility industry.
Intangles thus began its operations in 2016 with its digital twin concept and a young workforce from Pune. When it first partnered with a large bus fleet operator, the platform proved its efficacy by accurately identifying the factors that could cause a vehicle to break down. The fleet operator was able to bring down instances of breakdown by 75%. Reminiscing about the journey started, Anup Patil, CEO & Founder, says: “We began to learn about the most common faults that fleet operators encounter such as overheating, alternator failure, untimely discharge of a battery, and poor performance of the fuel and air metering system. Then, we began to map those concerns in our diagnostic data-streams.”
“And it was from there that our initial suite of predictive algorithms began to emerge,” he adds. Feedback gathered from the field, fundamental principles of physics governing drive-lines and knowledge of AI were used to simulate digital twins of real-world assets, capable of reasoning and learning. “After speaking with fleet operators, we recognised that a vehicle has specific components prone to failure and we need to provide component level analytics. We ended up getting into this whole data twin ecosystem by accident when we realized we were using physics-based analytics and integrating it into our Machine Learning platform in the form of hybrid analytics. After doing some more research, we discovered that such technologies are commonly perceived as digital twins. So, we began our adventure long before we were even aware that the concept of a digital twin existed,” Anup says.
The digital twin is an Internet of Things (IoT)-based technology that allows for the true duplication of a physical object into its digital counterpart. Intangles uses digital twins to simulate the behaviour of components and systems in the virtual space, in the process churning mission critical insights on wear and tear. “We develop a virtual reproduction of vehicle aggregates that comprise internal components as well as the environment in which they operate. This is accomplished through the integration of edge-computing with AI data models. Our UX philosophy, content management system and component level algorithms are always kept in tune with the KPIs of the quintessential fleet manager. Intangles has carved out a niche for itself by simplifying complex systems built around telemetry and AI for the everyday fleet operator,” adds Neil Unadkat.
Convincing the Fleets
Even though the Indian transportation sector is largely fragmented, it is dominated by a few big influencers. When the go to market was consolidated, a clear-cut framework was established on preventing vehicle breakdowns and accrual of cost benefits to the fleet operator. Shares Anup: “We are talking about improvement in the overall performance of the vehicle and increasing the life of the asset by reducing its downtime. Therefore, we were able to show customers the significant value they would derive out of the system.”
“Customers in the Heavy Commercial Vehicle segment have witnessed savings of INR 40K to 50K per month after being on-boarded onto the Intagles platform. This is because we quantify the reasons for poor performance related to low mileage or bad driving behaviour. Within 30-45 days of installation a customer is able to achieve the return on investment.” Aman adds: “In the initial stages of marketing, fleet manager’s feedback on component level profiling was addressed, there were certain conventional aspects of driving behaviour which needed attention. That certainly was a daunting challenge.”
“For our key customers, we offer a comprehensive driver training programme. The idea is to educate the drivers of practices that will suit the modern vehicles and lead to improved efficiency of the vehicle. One such example is that many drivers still believe that driving down the hill in neutral saves diesel. However, with modern powertrains they actually end up losing fuel with the added risk of destabilisation during braking and engine overrun resulting from shifting into lower gears at speed. We tell them how drive-by-wire engines work and how they can actually save on diesel by driving the vehicle in a certain gear appropriate for a particular speed or momentum,” he adds.
A Proprietary, Scalable Platform
Intangles has an in-house team that designs and develops its own edge-computing system. The advantage this gives them is to be self-reliant with parallel designs and overcome any shortages and disruptions in the supply chain. “We have a vast network of global supply chain partners for sourcing components with adequate flexibility. The device itself has been designed considering all the network topology issues that we see in India. The intelligent over-the-edge computing reduces the noise from the vehicle by a significant amount. Additionally, since we have an in-house embedded systems team, we have been able to create a common platform capable of reading data from 1600 different specs of drivelines – that’s how dynamic the system is. In manufacturing, we have extensively leveraged the robust manufacturing infrastructure within India, and have no plans to take this outside. We have a state-of-the-art hardware lab for design validation, testing and packaging.” Jayshri informs.
Intangles’ devices are collecting more than 2 billion sensory data points per day. This vast quantum talks of the uphill challenge in making business sense out of telemetry data stream. “We have ensured that the platform delivers insights in the fleet operator’s jargon. We spend a significant amount of time hand-holding fleet operators to train them on the platform. Data consumption can happen through multiple avenues such as web dashboard, mobile app and even SMS alerts. We also have a backend system which monitors how customers engage with each and every module on our platform,” Aman explains. If the system notices that a fleet operator has not consumed a specific feature, the command centre team kicks in to raise a red flag about any critical fault that needs immediate attention.
A Competitive Edge
As mentioned earlier, OBD data has been available for many years and there are companies that can make it available for fleets. In such a scenario, what is the value-addition that sets Intangles apart from the other fledgling players in the segment? Says Anup: “Our team has gathered expertise in reading the data, prioritizing the relevance of a parameter in custom data models and projecting the end results in a simplified manner to end users. If you pick specific criteria that we analyse, while competitors are still talking about mundane alerts for over speeding and hard braking, we are raising the bar with peer-to-peer driver ranking based on intricate aspects of drive-ability such as transmission utilisation and use of engine power modes.”
“And while they are predicting at 80% accuracy, we have already reached a level of 97% with continuously learning digital twins,” he adds. The analytics capability that Intangles has developed has also helped them build customised solutions catering to the needs of OEMs across industries. “While a fleet operator wants to see granular data on day-to-day operations, the OEMs are more interested in component level performance aggregates across geographies. It would help them in limiting warranty costs by profiling drivers who abuse vehicular systems. The difference between failure and time to repair is a key parameter for them. Hence, our offering for OEMs is designed accordingly,” Aman informs.
Plans for Expansion
Predictive analytics is an evolving science subject to research and development across the globe. After setting a strong foundation in India, Intangles is expanding to foreign shores. “We are expanding operations in North America, Europe, Australia and New Zealand. In the next three to four months, with the kind of partnerships that we have established, we will be deploying our solutions in Turkey, Vietnam and other countries in the APAC. Fleet operators, whether they are operating in North America or Europe, are ready to get this level of insight because they know the cost of breakdown far outweighs the cost of predictive analytics as a service,” Anup says.
Prior to the pandemic, Intangles established presence in nine other countries, including Malaysia, Indonesia, Thailand, Philippines, Kuwait, the UAE and Belgium. “As part of our efforts to scale our services and make inroads into developed markets, we have partnered with some very large companies. We have partnered with an oil and gas major to penetrate large fleets in these developed markets. That’s the kind of expansion plan that we have charted out over a period of the next two to three years,” he adds. As regards the numbers in terms of revenue and fleets monitored, Anup discloses: “We have been growing at 250% year-over-year, and hope to be in the same range in terms of revenue growth.”
Probe him about the electric vehicle (EV) fleets making way into mass transportation and how this will translate for the company and Anup says that he believes that it shall open immense opportunities for predictive analytics. “As the number of moving parts reduces, sensors are required to map data that is not easily accessible. That said, the monitoring of all these new electrical components becomes even more important. IC engines have stabilised over the past 100 years while the EV is still evolving. Hence, it would require a higher level of monitoring and understanding about how systems operate under diverse usability conditions,” he adds.
Aman believes that looking from a digital twin perspective, It will be interesting to see how battery performance is predicted in the context of varying weather ambients and load conditions. “The possibilities for integrating digital twins in EV performance management are endless,” he concludes.
—
“We are talking about improvement in the overall performance of the vehicle and increasing the life of the asset by reducing its downtime. Therefore, we were able to show customers the significant value they would derive out of the system.
– Anup Patil, CEO & Founder
—
“Intangles has carved out a niche for itself by simplifying complex systems built around telemetry and AI for the everyday fleet operator.
– Neil Unadkat, CTO & Co-Founder
—
“Customers in the Heavy Commercial Vehicle segment have witnessed savings of INR 40k to 50k per month after being on-boarded onto the Intagles platform. This is because we quantify the reasons for poor performance related to low mileage or bad driving behaviour.
– Aman Singh, Head of Analytics & Co-Founder