Today, more than half of the world’s population lives in cities, and more than a billion people annually take to the road or to the skies to travel for tourism (while some tens of millions do the same for work). Indeed, today’s world is one in constant motion, but it’s also one where trends are clearly shifting.
For example, McKinsey predicts that by 2030, private cars will be used less, and autonomous shuttles could account for a quarter of passenger-kilometers. This kind of rapid change in passenger trends and habits – not to mention shifts in environmental concerns – necessitates more than ever before that mobility companies use data and AI systems to better understand and help their customers (as well as, eventually, to predict the future of travel to make the best possible business decisions).
This is the story of Europcar Mobility Group, who has been able to do just that by transforming their approach to executing data-driven projects at the company.
We spoke with Guillaume Giard, Airline Market Manager of Europcar international, to better understand how the company has been able to use machine learning and advanced data strategies for better forecasting, pricing, and field/capacity management.
It’s worth noting that the business is already growing in all sectors (from cars to new mobility), because of an existing combination of initiatives to digitalize, reduce costs, and innovate to scale up. However, Europcar Mobility Group was facing specific challenges regarding car rentals at airports and the ability to accurately predict fluctuations in demand based on changes in the market; for example, the International Air Transport Association — IATA — predicts that routes to, from and within Asia-Pacific will see an extra 2.35 billion annual passengers by 2037.
Our group sees more than 50 percent of its turnover of rental cars at airports, so you can see how getting predictions about how many cars need to be where at what time of the day, month, and year is critically important for the larger business.
– Guillaume Giard, Airline Market Manager | Europcar International
In order to start tackling the problem, Europcar wanted to build an application using data from different sources (both public and private), including fleet traffic and passenger volume, reservation and billing data, data on opening of new airline routes, and more. This presented another challenge, as the data comes from different places and is in different formats, but is also massive in volume.
Given these challenges, executives at Europcar Mobility Group assembled a dedicated team composed of different experts throughout the business, including analysts, IT, and the head of forecasting. The group was tasked with collaborating to build a solution that allows the company to make better decisions regarding the purchase and movement of vehicles between airport hubs.
Given the cross-functional nature of the team as well as large volumes of data they planned to use (dispersed across several databases), the Europcar team needed to build their solution in a tool that was collaborative and flexible. Additionally, they sought to build a solution that would forward the democratization of data science throughout the organization — not just for this single project, but planting the seed for wider and smarter data use throughout the company.
Ultimately, Europcar leveraged Dataiku to build a predictive web application (web app) as well as dashboards that:
The important thing for us was ultimately having a tool that allows us to make use of important data while leaving out useless information. We chose Dataiku because it’s a platform for business users who have little knowledge of data science as well as for technical profiles, so it allows us to immerse ourselves in machine learning and predictive projects with a cross-functional team.
– Guillaume Giard, Airline Market Manager | Europcar International
Importantly, the data and the web app as well as dashboards are shared with everyone, from top management to analysts. This illustrates the company’s progress and overall larger commitment toward data democratization.
The company’s recent initiatives to use innovative data strategies in the car rental sector for better forecasting, pricing, and field/capacity management had the following business results:
Europcar Mobility Group can now flexibly adapt their operations, as they are able to anticipate line openings in airports, allowing them to adjust the staff, the vehicle fleet, and the parking space required to respond correctly to demand.
Europcar uses Dataiku to build machine learning models, create dashboards and a web app for final consumption, automate data ingestion, and clean/prepare data.
They plan to continue to tweak and improve existing models created as a part of this project as well as create new dashboards. As a next step, they will work on automating data injection and improving the data quality overall by collaborating with even more departments to track down additional data sources.
We’ve successfully moved from a word-of-mouth strategy when it comes to making decisions and information access to a solid, reliable, source of truth to which everyone (whether technical or not) has access. Now, we’re all on the same page, and we’re able to make even better decisions.
– Guillaume Giard, Airline Market Manager | Europcar International
But beyond this particular project, Europcar has successfully started their path to enterprise-wide data use. There are additional teams who will start to use Dataiku on their own projects with the end goal of bringing real, tangible business value.
On average, two people armed with the Digital MI team's applications in Dataiku are doing the work of about 70 people limited to spreadsheets. That means increased analyst productivity by a factor of 30 by replacing spreadsheet-based processes with governed self-service analytics.
Read moreCalculating ROI on AI projects is often critical to secure resources; however, these calculations can also be notoriously challenging due to the complexity of quantification, number of teams and people involved, and often long delay in outcome.
Learn MoreThere is no question that elasticity, including on-demand compute resource management, is the future of Enterprise AI.
Learn MoreSee how Dentsu Aegis improved their data department’s agility and speed with the ability to quickly generate customer segmentation reports.
Learn MoreSee how Trainline data engineers have become 3x more efficient in creating aggregated, centralized dashboards focused on Trainline marketing acquisition efforts.
Learn More
