Is text analytics part of your current analytical framework? For many SAS customers, the answer is yes, and they've uncovered significant value by integrating this technology. As text data continues to explode in volume, SAS Event Stream Processing can be used to analyze not only high-velocity structured data, but also unstructured […]
Decades of Westernized television and cinema have featured fantastic imagined car technologies, including many that now actually exist. Think of the autonomous car from Knight Rider and today’s self-parking capabilities. Or the ongoing James Bond series with tracking devices that resemble the well-established GPS systems in our cars, phones and even wearable technology.
Art has become reality, and it continues to evolve for automotive manufacturers. Why? The short answer is that we’re obsessed with innovation. And while the connected car concept isn’t exactly new, the sheer breadth of the concept certainly raises new questions. Consider this thought in light of the opportunity made possible by analytics:
Machine-to-machine (M2M) connections
are expected to reach an estimated
1.8 billion worldwide connections by 2022.
At a recent Connected Vehicle Trade Association (CVTA) summit, the reality of automotive connectivity and associated big data challenges were explored. If an estimated one terabyte of data per hour is generated, according to Andreas Mai, Director of Smart Connected Vehicles for Cisco Systems, consider the opportunities to transform the automotive customer experience.
Some automotive customers use SAS’ Advanced Analytics capabilities with real-time decision making for credit scoring, enabling smoother customer experiences at a critical juncture - during financing negotiations. Others use Predictive Analytics for generating real-time offers at the point of service based on customer feedback, past purchase, service data and other interactions associated with the brand.
However, the connected car revs up a whole new world of customer interaction. Picture better driver safety capabilities by predicting recall and accident-related potential. Or the deployment of SAS® Quality Analytic Suite, currently used by Volvo Truck, could expand to include the use of streaming data for minimizing downtime due to breakdowns.
Envision location-based services that automatize the vehicle to get around traffic or inclement weather, or go to a fuel station when a minimum threshold is reached. Personalize the driving experience with media content using connected car infotainment options based on streaming data. The data collected by SAS customer Maruit-Suzuki could expand into real-time streaming of preferences for even further insight.
So how do we get to all this streaming data? Managing data in motion (real-time, not the moving vehicle), is quite different than data at rest. The SAS® Event Stream Processing Engine, often referred to as ESP, offers access and relies on three principal capabilities – aggregation, correlation and temporal analytics.
- Aggregation. Let’s say you wanted to detect the costs of repairs per vehicle: “Tell me when the value of repairs on any day is more than $x per day (or even hour)”. Rather than waiting for aggregated metrics to tell you if your vehicles are causing repair costs higher than annual averages (impacting poor customer experience potentially leading to defection), you could have a real-time pulse of repair counts.ESP can continuously calculate metrics across sliding time windows of moving data to understand real-time trends. This kind of continuous aggregation would be difficult with traditional tools.
- Correlation. Connect to multiple streams of data in motion and, over a period of time that could be seconds or days, identify that condition A was followed by condition B, then condition C.For example, if we connect to streams of diagnostic codes from thousands or even millions of vehicles on the road, ESP could continuously identify conditions that compare vehicle service events to each other. This might look something like “Generate an alert if a certain diagnostic code is more than 150 % of the average of other vehicles.”
- Temporal analysis. ESP is designed for the concept of using time as a primary computing element, which is critical for scenarios where the rate and momentum of change matters. Take surges of activity as clues to potential quality issues as an example, where ESP could detect such surges as they occur.Consider searching for parameters such as: “If the number of diagnostic codes triggered within four hours is greater than the average number of daily DTC triggers of that vehicle / make / model in the previous week, launch an immediate audit of that part’s quality.” Unlike computing models designed to summarize and roll up historical data, event stream processing asks and answers these questions on data as it changes.
Improved automotive products and customer experiences are the end goal for the connected car. But possibilities go on for miles, or kilometers if you will. It’s an exciting time in the world of automotive big data analytics. While futuristic transportation ideas like the 1960s Jetsons’ flying car may not exist yet, it’s just a matter of time before we’re all talking, or perhaps texting, with our cars.
For now, start with what’s right around the corner for the connected car. Take a look at IIA’s report Making Meaningful Predictions in the Fast Lane for a rundown on a few of the many possibilities for event stream processing, data sensors, data in motion and the Internet of Things.