Deep Dive: How Zoomcar Is Using Machine Learning To Grow Its Fleet Size To 1 Lakh Cars


India is one of the most interesting and dynamic markets driven by the youth — with the median age around 28. Being one of the mobile-first economies, the nation over the years have continued to add new digital users to an existing large pool.


Sign up for your weekly dose of what's up in emerging technology.

Today, India the millennials and Gen Z’s as the key consumer — the ones who value accessibility over ownership. And this where OTT apps or several renting businesses models come into the scenario – from housing, office spaces, to furniture. However, when it comes to vehicles, the industry was still in need of a player. And this sheer need opened up the doors of opportunities for Zoomcar, a self-drive car rental company.

In order to know more about the industry and the company, Analytics India Magazine talked to a duo from Zoomcar — Arpit Agarwal, Director, Data Science and Anirudha Jadi, Senior Director, Human Resource.

Transformation From The First Year Until Now

Zoomcar was founded in 2013 by David Back and Greg Moran, with the vision of providing easily accessible mobility services to the Indian audience. The company initially offered only self-drive car rental services; however, it has expanded its offerings which includes car subscriptions.

Today, Zoomcar is now present across 45 cities in the country and have tie-ups with some of the car companies. Talking about funding, in a span of 6 years, it has raised close to $103 million. Another major leap is onboarding and piloting of Electric scooters in its shared mobility fleet to foster the green and sustainable ecosystem.

If we look at the numbers, Zoomcar roughly does 5000 self-drive short duration cars trips in a day and have been clocking close to 700 car subscriptions on day to day basis. “In the self-drive space, we command about 80% of the market share. In the shared subscription mobility market (Zap Subscribe), we are the lone leaders with more than 10,000 cars currently. We are already operating on a 100% marketplace model,” the duo said.

When asked about challenges, the duo said that in a shared mobility ecosystem, an asset goes through multiple hands and is prone to a high amount of wear and tear. From maintaining a high-quality condition of the car to car theft to coping up with trips delays, and breakdowns, the challenges are real. However, Zoomcar takes necessary measures to stay on top of the game.

“Our motto is to encourage better driving amongst Zoomers, so as to minimize instances of accidents or breakdowns, thus lowering our service and maintenance costs,” said the duo.

The Tech Stack

Coming to the tech stack aspect, the core technology is the Internet of Moving Things (IoMT) enabled shared mobility. IoMT empowers a keyless entry (KLE) ecosystem wherein a user can lock or unlock a car through a mobile app and have a seamless trip experience without any human intervention.

Data also plays a major role in the companies game, it uses MySQL, Timescale, Redshift and MemSQL for storing/processing its data. Recently, the company has shifted to Data Lake for storing its application data. “We use Redshift cluster for maintaining our data warehouse to empower self-serve analytics. Python and SQL are used for the same,” the Agarwal explained. “For queuing we use RabbitMQ and Kafka; the backend is on Go, Java, ROR; frontend is on Vue and React; and mobile is on Java and Swift.”

Talking about the role of machine learning, Zoomcar is using this technology in the following use cases:

  • Zoomcar’s fleet management system is built to ensure high asset availability at the right place and time. Linear programming-based optimization models help achieve it.
  •  Dynamic pricing, to curate best prices for the prospective renters and subscribers
  • Retention strategies are built on the concept of relevance and personalization. So, the company utilizes churn, lifetime value models and customer segmentation.
  • Predictive maintenance models are built to service the vehicle at the right time, to minimize breakdowns. This also helps in reducing the maintenance costs of the cars.

That is not all, Zoomcar is also actively making the best use of technologies like computer vision and deep learning algorithms. Here are some use cases:

  • To detect real-time collision warnings and send rash driving alerts to prevent accidents
  • To identify fake use of driving license and Aadhaar card to reduce car misuse

Furthermore, Zoomcar is also planning to come up with an indigenous driver scoring system which will rate the renters and assist them with safer driving, which is going to be the first of its kind.  Also, lastly, customer centricity is an integral part of all the Zoomcar initiatives, and this is why the company uses NPS analytics for all the ongoing process changes.

The Hiring Phase

When it is about hiring and dealing with talent crunch, the company creates a bench strength, by identifying potential candidates who are nurtured & developed for future requirement. When it comes to dealing with talent crunch, the company uses LEAP (or internal transfer). “This step is mostly overlooked otherwise, but we believe that it is the easiest way to access the internal pool of talent who have a good understanding of business, metrics, processes & culture too,” said Jadi.

Zoomcar has about 70 engineers that powers the Pan India shared mobility ecosystem across all product verticals. Being one of the best companies in India from the car rental domain, Zoomcar makes sure that they onboard the best engineering talents.

When hiring for engineers, the company tend to prefer individuals with a deep technical exposure, possessing the ability to quickly learn and adapt. “An ideal candidate for us would have a high end-user focus, good problem solving and communication skills, proficiency in time and task management, and a good team player,” Jadi added.  

The company also works on the personal development plan model where respective employees are asked to design their learning & development calendar and based on that internal & external training are planned & imparted to employees. And talking about the delivery of the training, it is by the mode of online learning courses, conferences, seminars, internal cross-functional learning sessions, classroom coachings etc.

Staying Ahead Of The Curve

“At Zoomcar, we enjoy the first-movers advantage in India. While several other players are now venturing into this space, we have already cemented our position as the category-leaders in India,” said the duo.

Two strong factors working in the favour of Zoomcar are the incredibly loyal user base and its tie-ups with behemoths in the automotive sector. Zoomcar has tie-ups with companies like Volkswagen, Nissan, Mahindra & Mahindra, Ford, and Tata Motors, and all of these offerings are extremely popular with the company’s user base. “In fact, 200 Polo cars were subscribed within 48 hours of us announcing the same on our subscription platform,” the duo added.

The Roadmap Ahead

When asked about the future plans of the company, the duo said that the prime focus is to increase the number of vehicle models which will be offered on a subscription basis and also aggressively partner with all the major automotive OEMs to offer subscription as an alternative to vehicle ownership.

As for the self-drive arm of Zoomcar’s business, the current plan is to nearly treble the fleet size to 30,000. The company wants the availability of a Zoomcar car to be similar to an Ola or Uber cab. However, according to the company, it has penetrated just 1 or 2 per cent. The company believes that growth can easily be 20-30x from the current level in the next five years — the number of cars in Zoomcar’s fleet could grow to 100,000.

Other future plan includes electric vehicles and international expansion in the emerging markets of Africa and Southeast Asia. 

More Great AIM Stories

Harshajit Sarmah
Harshajit is a writer / blogger / vlogger. A passionate music lover whose talents range from dance to video making to cooking. Football runs in his blood. Like literally! He is also a self-proclaimed technician and likes repairing and fixing stuff. When he is not writing or making videos, you can find him reading books/blogs or watching videos that motivate him or teaches him new things.

Our Upcoming Events

Conference, in-person (Bangalore)
MachineCon 2022
24th Jun

Conference, Virtual
Deep Learning DevCon 2022
30th Jul

Conference, in-person (Bangalore)
Cypher 2022
21-23rd Sep

3 Ways to Join our Community

Discord Server

Stay Connected with a larger ecosystem of data science and ML Professionals

Telegram Channel

Discover special offers, top stories, upcoming events, and more.

Subscribe to our newsletter

Get the latest updates from AIM