Listen to this story
Commemorating over 35 years of unrivalled racing legacy now, TVS Motor Company has been one of the most dynamic motorcycle brands in the world. With four state-of-the-art manufacturing facilities in Mysuru, Hosur and Nalagarh in India and Karawang in Indonesia, the global two and three-wheeler manufacturer endeavours to deliver the most superior customer experience across 80 countries in which they operate. TVS Motors is also a recipient of the prestigious Deming Prize.
“At TVS Motor, we have always focused on bringing great customer experiences driven through robust and innovative technologies, be it in our vehicles, our production engineering systems or through digital and AI technologies,” says Maheshwaran Calavai, Chief Digital and AI Officer at TVS Motor Company.
Sign up for your weekly dose of what's up in emerging technology.
In an exclusive interview with Analytics India Magazine, Maheshwaran sheds light on how the brand leverages AI to build superior products, enhance business performance, and offer an immersive customer experience.
AIM: How is AI driving the latest trends in the automotive industry?
Maheshwaran Calavai: AI is transforming industries across sectors, and mobility is no exception to this. AI-driven digitisation of consumer experiences is one of the key trends being leveraged by the automotive industry. Mobility companies are improving consumer experiences during the initial purchase and throughout the ownership. Since there is a continual engagement with customers and since these touchpoints are getting digitalised end-end, applications for data-based decisions are extensive and are being actively pursued by the industry.
Furthermore, the rapid rise of CASE (Connected, Autonomous, Shared, Electric) mobility is one of the key developments led by the rising interest in technology development. Today’s auto enthusiasts are more tech-savvy, and they choose IoT-enabled automobiles that meet their needs for safety, navigation, audio streaming, and other features. The rising usage of AI-driven technology, along with a surge in electric and hybrid mobility, is one of the biggest trends to watch in the automobile sector.
Customer preferences of ownership vs usership of vehicles for their mobility needs, along with connecting multimodal options for end-end convenience, is a key trend that will shape the industry this decade. Autonomy driven by AI will be a major contributor to this shift. Connectivity also plays a crucial role in designing and manufacturing vehicles and associated components to improve speed, quality and efficiencies in operations.
Reflecting on the challenges, deployment of AI-enabled applications at-scale in sales across the extended value chain, like in dealerships, depends on consumers’ preferences. The automotive buying experience is today largely “phygital“, wherein consumers often discover and research digitally, but the final buying happens in a dealership. Convincing dealer partners that AI algorithms can present better insights on consumer preferences, sales potential, and engagement during vehicle ownership becomes critical to creating a balanced and seamless experience across physical and digital touchpoints. There are several such examples, and we look at human-in-the-loop AI systems as an opportunity to combine the precision of algorithms with the experience of experts.
AIM: How important is data science at TVS Motors?
Maheshwaran Calavai: At TVS Motor Company, data engineering and data sciences remain at the forefront for better decision-making in all aspects of our operations.
We leverage data science-driven applications in both direct-to-customer and digital aided experiences in dealerships. For example, TVS iQUBE Electric scooter offers advanced features for our customers that are rooted in data sciences. On the digital-aided side, arming our dealer partners with customer preferences during sales and vehicle service improves customer experiences. Hyperlocal marketing improves marketing efficiency and AI leveraging vision, and NVH (noise vibrations and harshness) improve supply chain quality and efficiencies.
AIM: Elaborate on ways TVS Motor leverages AI with use cases.
Maheshwaran Calavai: Our international business spans 80 countries across the globe, where customers use our two and three-wheelers for their mobility needs. Due to the specific requirements in various countries, we despatch vehicles in CKD (completely knocked down) & SKD (semi-knocked down) conditions. We won’t assemble the complete vehicle even if one part is missed, mismatched, or has a defect.
We first implemented an IoT-based platform for packing assembly using barcode-based scanning to ensure all the right parts are placed in the right boxes. To make the packing process Poke Yoke (mistake proofing), we wanted to deploy Vision-based AI systems to complement the IoT platform. Further, weighing machines present a good mechanism for parts like chain and cable assemblies that do not have fixed profiles. So, we have developed and fully integrated deep learning algorithms and complementary systems with the manufacturing execution system (MES) in multiple stages of packing assembly. For each vehicle variant, the stagewise parts to be packed are displayed on a computer screen, and each part is identified through the Vision AI system. If it is the right part, the conveyor will run, and assembly will continue. On the other hand, if the vision AI system detects a wrong part, the conveyor will stop since the conveyor PLC is integrated with the result of the vision system. The operator then inspects and replaces it with the correct part.
In another example, we used AI vision during the pandemic to ensure adherence to safety protocols. We developed an “AI-Based – Social Distance Monitoring System & Mask Wearing Detection System” in our factories and global offices of TVS Motor group companies. We used YOLO – deep architecture for Person Detection & RCNN (Region-based Convolutional Neural Network) based deep architecture for mask detection. Both models have their advantages and limitations.
At TVS Motor Company, we see large volumes of consumers’ interests. Such enquiries have different propensities to buy as they may be at different stages in their buying journey. Each lead requires the TVS Motor marketer and dealer sales staff to follow up promptly and personalised. At any point, our dealer partners in India will have 5+ lakh enquiries to follow up. To classify leads and help our dealer partners make profitable decisions, we extensively use data science algorithms powered by AI/ML.
We view AI as a platform that provides opportunities for human-in-the-loop solutions, allowing the best of experts and technology to come together to improve decision accuracy, consistency, and speed.
To ensure all aspects of AI are taken care of, we have several workstreams that strengthen the foundations of AI. First, we ensure the underlying data is managed and governed well across our group companies through an established industry framework. We maintain strict standards when it comes to data robustness. We also have data privacy and protection programs, along with data enrichment techniques.
AIM: What is TVS Motors’ acquisition strategy?
Maheshwaran Calavai: Our focus is to provide the best products and services for our customers through innovation – some we create ourselves, some we co-create, and some we partner with other innovators. Over the last few years, we have made significant investments in startups such as TagBox, Rapido, Scienaptic, Ultraviolette and others.
In addition to investments, we work with several startups in customer-facing, manufacturing, sustainability, and enterprise operations. We are also working on open innovation initiatives with industry partnerships.
AIM: What are your thoughts on Autonomous Vehicles?
Maheshwaran Calavai: Autonomous driving will radically transform the mobility of people and products in the coming years. Self-driving technology will first and foremost provide convenience and ease to commuters. If you look at how automation has helped consumers in the last century, hours and energy spent on tasks have been made easy and converted into a formal economy, with actual GDP being added to countries and revenues to companies. Autonomous vehicles will be no different, freeing up human energy and time that can be devoted to other things they would like to do!
From an economics standpoint, this will simultaneously offer immense opportunities to create and grow new products, business models and innovations whilst also reshaping the communities we live, work and commute in.
With full autonomy, vehicles can be on the road all the time. Vehicles owned by consumers for their personal needs are largely idle today except for a few busy hours on the road daily. If vehicles can ply autonomously, the productive time per vehicle day can be maximised. This will bring into question why one would want to own a vehicle when they can get a vehicle to fetch them whenever they want. So instead of a vehicle being a depreciating asset, it can be a revenue-driving investment. How will this also shape the vehicle financing industry?
If we can maximise the productive hours of a vehicle every day, why would we need the same number of vehicles to meet the same demand? If this decreases the number of active vehicles in the world as a result, why would we need so much parking space in the cities and communities we live in? The same autonomous driving tech will power drones enabling 3-dimensional transportation in micro-geographies. How will this shape urban planning and transportation systems?
If they can play on their own, why can’t vehicles like cars or vans be designed to be bidirectional with no need to reverse? If vehicle-to-vehicle and vehicle-to-infrastructure communication and associated navigation are perfected, what would happen to vehicle insurance? If there is indeed an incident, which is to assume ownership of the failure in communication and navigation? How will these change laws and regulations?
Several industries like motels and highway convenience stores are designed to give tiring drivers a break during their long journeys. What will happen to these?
These are exciting questions that autonomy brings, and it is indeed an exciting time to be in the mobility industry at a time when technology is on the brink of making these questions real and now!
While several specialities from design and engineering to marketing and law will come together to solve these, it is doubly exciting to be in the data engineering and data sciences field during this time as these are large-scale data assimilation, engineering and algorithm solutions that will be at the core of autonomous driving.
Even before full level 5 autonomy, innovations and usage of AI in vision, voice and tactile are key to aid mobility. These are akin to the human-in-the-loop AI examples noted above.
AIM: How far are we from achieving level-5 vehicle autonomy?
Maheshwaran Calavai: The automotive sector is fast evolving in autonomy. We are on the cusp of a global breakthrough, but we still have some way to go before we reach level 5 autonomy as far as two-wheelers are concerned.
Several companies and organisations are researching and developing the technologies, infrastructures and policies needed to gear for full autonomy, including in autonomous two-wheelers. The sensing and computing power needed for full-scale autonomy needs to be finetuned for the form factor and price points of two-wheelers which is an engineering and commercial problem different than for other, larger vehicles.
Similarly, challenges like self-balancing, riding behaviour and safety considerations are also different in the case of two-wheelers. Further, most of the two-wheelers in the world in developing economies with different traffic patterns, usage on un-laned and off roads, road conditions etc., need to be solved.
AIM: What does the future hold for TVS Motors from a technology standpoint?
Maheshwaran Calavai: We are working on several immersive experience technologies in our digital assets and in-store. Some of these, like chatbots, 360 virtual experience, in-dealership augmented reality, and real-time visual feedback of vehicle service, are already deployed, and several others are underway. We continue strengthening our capabilities to provide mobile-first, personalised experiences for our customers in direct-to-customer and digital aided engagements. Such technologies and capabilities also have transformative power in automotive design, manufacturing, supply chain and enterprise operations.
Many of these digital experiences have AI underpinning them to personalise and make them more effective. To power these AI systems, we have adopted a cloud-first data engineering approach, coupled with foundational efforts to enrich, manage and govern data with the requisite data privacy and protection programmes. We leverage open-source software and develop purpose-built algorithms. Digital and AI products are built using an agile framework and DevSecOps methodologies. This digital and AI path we are on has created several wins already, and we want to create a competitive advantage with these technologies by providing immersive customer experiences, superior products, and better business performance in the years to come.