Automotive

Drive business improves with spatial analytics

You wish to assess the market better and find out where the new businesses are, as well as have a better understanding of your customers’ demands and develop new requests? This is possible thanks to the data geolocation. With GIS, integrate your existing systems to have access to different features like SMMT (Society of Motor Manufacturers and Traders) data visualization, trends analysis and use predictive analyses to be always more competitive!

Bruce Wong, Advanced Network Analytics Manager, General Motors

Big Data in automotive industries

Big Data for a better driving experience
Cars have always been information mines - from driver data to their driving environment, to the car itself, and all connected peripherals - smartphones, tablets ... Some manufacturers have even collected up to 25 gigabytes of data from a hybrid vehicle.

But now the automotive industry can go further by taking advantage of the mass of data exchanged inside and outside the vehicles in real time. Car manufacturers improve the safety and quality of our cars. They will also be able to move to higher gear in the driving experience and the services offered to their customers.

Indeed, although manufacturers have been collecting data from vehicles for years, something different, more sophisticated, is happening thanks to Big Data.

The drivers have changed and their cars now have to change with them. In a hyper connected world, they expect their cars to offer them the same features and conveniences as their other smart devices. Let's take the example of an individual preparing for a family vacation. A few weeks before the big start, he posts the news on his Twitter or Facebook account, previously connected to his vehicle. Then, depending on the mileage of the car, the theoretical wear rate of the parts or its driving behavior, the driver receives an alert on his smartphone or his computer on board to warn him 15 days before departure that ' a revision is necessary to make the journey from his home to his holiday place without risk of breakdown.

The Big Data challenge lies in the exploitation of previously untreated data (social networks, forums, vehicle data, etc.) because they were considered too massive, diverse and unstructured, and that the adapted technology did not exist!

Vehicle manufacturers now have the capability to analyze large volumes of vehicle data at speeds that also allow them to connect with drivers in real-time on their dashboards, smartphones or via GPS for example. So the cars of tomorrow will offer a personalized driving aid. This aid will be automated and contextual thanks to the real-time analysis of the data collected by a set of vehicles.

In the event of a hazardous situation detected on the road, it is then possible to propose to the vehicles present in the zone a bypass route. When the airbag is triggered, for example, a signal will be emitted to alert the emergency personnel and to inform them of the exact position of the accident vehicle but also to warn the approaching vehicles.

Drivers are no longer just looking for quality, safe and reliable performance. They now see their car as a very personalized extension of their daily lives, and that is why it must be like all devices at the heart of the Internet of the objects, it must be connected. Not surprisingly, 60% of the vehicles sold worldwide will have connected services by 2017.

The movement is underway throughout the automotive sector
Just like the recent collaboration between PSA Peugeot Citroën and IBM on connected services. This project was created to develop a new range of 2.0 services for the French manufacturer's vehicles. Experts from both entities worked together to integrate and analyze massive amounts of data from cars, telephones, traffic signals and other information sources to provide tailored services to driver in real time.
In addition, the German automaker BMW Group and its quality control division also rely on Big Data solutions to combine and analyze data from the numerous road tests carried out with their prototypes. Vehicle recall campaigns around the world are extremely expensive and BMW has sought out Big Data researchers and experts to guard against these losses. The aim of the manufacturer is to detect and correct in a few days certain defects before the production of new models, which required several months before.

It becomes evident that dramatic changes in the way customers

Combining big data, analytics and geographic information systems (GIS)

General Motors (GM) goes a step further

Many businesses have the wish to discover sale trends and anticipate potential opportunities, that’s why they do to an extended data analysis. Go further with General Motors and associate big systems of data, analyses and geographical information (GIS) to the performances offered by the car distributors.

This confers them a significant advantage: the possibility to determinate how to be competitive in relation to actual results. By these, it is implied that car distributors study local people, local characteristics, regional differences and the competitive environment of the concerned brand. This mechanism allows businesses to isolate the request more easily, to take local preferences as well as market competition into account in their marketing strategy.

Bruce Wong, network analysis manager at GM , explains that the intention is to propose real analyses instead of presenting data in tables or through dots on maps; the point of all this being to actually find out about the customers’ lifestyle, the clients’ profiles, and how far they live to the sales site.

Ever since its apex in 50’s and 60’s, GM has been supervising a large network of 13.000 car distributors. Today, even if it only supervises as many as 4.300, GM still wants to provide the same level of services thanks to information technology. To achieve this goal, GM has been setting up strategies since 1995 and is progressively exporting them.

(Source : Drew Robb, Contributor)

Driving in common agreement

Without proceeding to a spatial analysis, a lot of information would stay hidden among the multitude of things we know. For example, the society has been able to determinate that a customer is willing to drive as much as two hours to buy a car. The location analysis of these buyers based on travel time has showed that they would drive further than the most convenient sales place in order to save 500€ if it does not exceed a two hours drive. However, they wouldn’t go so far just to have their vehicle repaired. The GM analysis engine allows the car distributors to determinate the places where they could satisfy the largest number of potential sales or servicing requests.

Adding a new system GIS has helped GM to add correctly localisations in calculations made by analytics. Wong said that he clearly sees the difference between cutting expenses and satisfying the customers’ expectations despite budget downsizing. He also explains that mapping and the “less is more” strategy results from location analysis. He makes it clear that it is possible to provide required data in order to make better choices and thus, to improve the service.

It is in the marketing field that changes are the most significant. Indeed, GM estimates its cost around 2 billion euros per year. While the strategy was previously to get to be known through prime time television advertising, on traditional channels, or on billboards. Now, the points of reference have changed and time has come to operate differently.

The strategy is to determinate which households who could afford buying the different vehicles according to their budget via extended analyses that aim at identifying and locating the buyers in every sale categories (for example: luxury brands, medium size vehicles …). Thanks to these data, the car manufacturer can therefore develop its advertising investments in the most promising areas.

Studying the households

Wong explains that expenses decrease whereas sales increase. Indeed, the study of the households who buy new cars has been intensified and expenses are now more focused on this type of household rather than those who are still attracted to older models or on second-hand cars. Moreover, this analysis triggers the question about what has been done so far on account of changes on technological and societal levels; thus, in possession of these data, the car distributors can be more efficient.

For twenty years, monumental changes have been made in the way that businesses are run or how the companies reach out to the customers and potential buyers. Wong notices that the rise of internet has proved more useful to research and comparison than to sales. This could nevertheless change with the intervention of the pioneer of electrical vehicles TESLA who would like to bypass car distributors in the sale process. If this is effective, it could trigger the need for the automotive industry for new spatial analyses to find customers. Anyway, Wong stays confident about the fact that his company is on the right track.