During the 3-months data collection campaign, Motion-S retrieved data from the vehicles, including location updates and engine measurements. The data was augmented and profiled using the Motion-S platform, adding contextual elements to the trips and extracting powerful insights into different topics. In particular, Motion-S focused on driving safety, energy efficiency and vehicle health. These topics were analysed in the context of potential data-driven solutions for connected insurance, fleet optimisation, predictive maintenance and EV transition assessment. In the following sections, we will provide some meaningful insights into the potential of connected car data for the aforementioned use-cases.
The auto insurance industry is about to be transformed by the ability of access to vehicle data in real-time. It will enable insurance companies to offer individually tailored policies and premiums to their customers, while enlarging their offerings with new data-driven features. Insurers are turning to usage-based vehicle insurance schemes to increase product coverage and to be more competitive.
Usage-based vehicle insurance premiums are calculated either according to distance driven (pay-as-you-drive) or according to driving behaviour (pay-how-you-drive). Unlike monthly or annual premiums, these plans allow for tariffs that reflect the actual risk exposure. Insurers benefit from better risk assessments and drivers will have access to better premiums and coverage.
The most straightforward usage based vehicle insurances are pay-as-you-drive plans. Here, the insurer just regularly reads the vehicle mileage, e.g. once a month, and calculates the individual premiums accordingly.
More advanced pay-how-you-drive plans take driving behaviour into account. Insurers offering such plans might use data points such as acceleration, braking, cornering and speeding. The combination of car data and contextual information provided by Motion-S allows insurers to create accurate behaviour profiles, with much higher explanatory power than traditional telematics factors.
Currently, usage-based vehicle insurance providers rely either on manual mileage readings or on external devices, such as the driver’s mobile phone or black boxes and dongles to collect relevant data. Solutions like these are not very user-friendly, prone to errors, leave room for fraud, and are expensive to operate and maintain.
Cloud access to vehicle data will remove barriers for customer acquisition, onboarding, and operation. Customer loyalty will improve while transparency in terms of data-privacy is guaranteed. The number of potential customers will increase.
During the project, Motion-S used the data points datetime and vehicle coordinates to calculate average speed per trip section. Despite only checking for updates every 5 minutes, Motion-S could draw conclusions about driving behaviour from the data. The illustration below summarises the approach.
Data-driven and individualised vehicle insurance plans will disrupt the insurance industry. With increasing connectivity of cars and growing acceptance of connected services, pricing pressure will intensify. Cloud access to vehicle data will be the most efficient way to gain access to the relevant data points, and will lead to a new chapter in insurance telematics. Insurance companies can take advantage of integrating High Mobility into the Motion-S platform to enable fast and reliable connected insurance solutions.
For more insights into the history of telematics in the context of insurance, please refer to the blog post from Motion-S.
The benefit of remote access to vehicle data for fleet management companies is obvious. Efficient and cost-effective administration and vehicle maintenance are becoming increasingly important. Better and more reliable vehicle data can often make the difference.
This may affect different areas and various vehicle properties. Just to name a few examples, fleet managers are interested in location, fuel levels, driving behaviour and also the general condition of their fleet vehicles.
During their data collection and analytics campaign, Motion-S could detect vehicle-damaging behaviour by monitoring fuel levels. Very low fuel levels can cause damage to the engine. Remote and regular insights into fuel levels give fleet operators the opportunity to instruct drivers and service personnel to refill the tanks on time.
This example is just one of many ways in which vehicle data can be used for better fleet management.
Access to vehicle data via the cloud will allow fleets to integrate new vehicles quickly and easily. The numerous data points will provide opportunities to save costs, to facilitate administrative work, and to improve the experiences of drivers and fleet managers. Eventually, the market will adapt to the novel opportunities to make processes more effective and cost-efficient.
Predictive maintenance is an emerging and interesting new use case. Combining diagnostics data for maintenance purposes and data analytics enables AI-driven applications and services.
Diagnostics data to improve maintenance has been around for decades. So-called OBD (on-board diagnostics) sockets have been installed in new cars since the 1980s. In 1996, the United States made the OBD-II specification mandatory, and in 2001, the European Union followed suit.
Today, most car repair workshops use OBD-II scanner devices to read diagnostics data from vehicles they repair or maintain. An extensive list of OBD-II parameters can be found online.
A growing number of automakers allow diagnostics data to be accessed remotely. The ability to observe and analyse diagnostics data in real-time opens up entirely new markets for smart applications and AI-powered services.
If potential faults are monitored and analysed from the moment they first occur, drivers and vehicle owners will be able to take action before something breaks or something bad happens and thus save time and money.
During their project, Motion-S recorded an impressive amount of data to work with: 231 different trouble codes and 331,655 trouble code activations coming from only 39 vehicles in 3 months.
You can see an interesting example of a recurring trouble code below. The error code 'Mixture too low' indicates that something is wrong with the engine or the fuel supply.
The example above is just one of many how monitoring and analysing vehicle properties can lead to useful insights. The High Mobility Platform and Auto API includes up to 400 data points. Nobody can keep track of all the parameters of a modern car, but computers and smart machine learning algorithms can. Predictive maintenance is just emerging. There are many opportunities for developers and software companies to become active and bring interesting and useful solutions to the market.
Residual value estimation
Residual value estimation is about the car as an investment. The residual value of a car depends on many variables, many of which are represented by vehicle data. Intelligent algorithms analysing vehicle data can estimate the depreciation and thus the current value of a car.
Smart fleet and vehicle owners want to know how much their vehicles are worth at any given time. They want to make good decisions for themselves and for their companies. Car repair workshops, on the other hand, also need to understand the residual value of vehicles because they have to advise their customers.
The residual value estimation use case is similar to the predictive maintenance use case. While predictive maintenance is about efficient vehicle use, residual value estimation is about determining the best time to sell and when to repair or when to dispose. Residual value estimations can include diagnostics data, vehicle usage data and used car market data.
Motion-S could show that deviations in the wear and tear profile based on vehicle data can be used to determine the relative depreciation of a vehicle.
Both, workshops and vehicle owners will benefit from residual value estimation services. Workshops will be able to provide better recommendations to vehicle owners, and vehicle owners will be able to make better decisions.
EV transition estimation
When does it make sense to switch from a combustion engine to an electric vehicle (EV) or a hybrid car? An interesting proposal by Motion-S for a data-driven service answering this question is electric vehicle transition estimation.
Car data can help analyse patterns in driving style and vehicle use. Creating an objective assessment to evaluate if switching to Battery Electric Vehicles (BEV), Plug-in Hybrid Electric Vehicles (PHEV), or Hybrid Electric Vehicles (HEV) respectively is favourable for an individual driver can rely on multiple factors, e.g., the driving style (e.g., harsh manoeuvres, anticipation), the typical daily distance travelled, the usual speeds, the road environment (e.g., urban versus motorway), the availability of charging infrastructure nearby, the effective deliverable power (kW), and the typical stop times close to chargers. Vehicle owners can make smart decisions about when it is the best time to switch.
Use cases such as EV transition estimation exemplify the advantages of hardware-free access to vehicle data and the subsequent disruption of automotive and mobility industries. Vehicle data driven digital services can be offered to potential customers just like any other software application. There are hardly any barriers for customers to subscribe to a service or an application. At the same time, they have the potential to influence the actions and far-reaching decisions of drivers and vehicle owners.
Now is the right time to start developing and investing in vehicle-data-driven services. The interesting value propositions of Motion-S show that if you put vehicle data and data analytics to work, there is an almost infinite number of imaginable products and business opportunities. Many of these business opportunities not only improve user or driver experience but also allow for cost savings, which is a good sign for investors and start-ups looking to enter or even create markets.