How Smart Car Tech Applies to IT Operations
Artificial intelligence, analytics and automation help build efficiencies and organizational growth.
Imagine a ride to the airport.
Getting into the car, you’re greeted by the dulcet tones of Tom Jones. He’s been a recent addition to your relaxation playlist, but today’s grind doesn’t quite put you in the mood for Tom’s crooning.
Sensing your shift and correlating this with your jam-packed work schedule, the music changes—time for a short, sharp wake-up from the God Father of Punk—Iggy Pop—“I am the passenger—and I ride, and I ride”. Now we’re talking!
It’s a good 45-minute hop to the airport, but you’ve got plenty of time to make your flight. The vehicle has already assessed the best route, necessary speed changes and steering tweaks based on analysis of truck-loads of cloud and edge data—scheduled roadworks, traffic-signal sequences, accident pattern recognition, weather predictions, infrastructure failure probability—even the fact that Acacia elementary school is holding a sports carnival and Norfolk county is conducting some seasonal bushland back-burning. Everything is a valuable data point—ingested and dissected within a single collective soup of analytical goodness.
As you settle back to assess the sales reports automatically presented on your console, the car detects a potentially problematic condition. According to sensor data gathered and compared against the same model of car with similar mileage and driven under the same conditions, there is an 87 percent likelihood of front-left wheel bearing failure occurring on next month’s family trip to the snow fields. But it’s no big deal. The system has already connected to the dealerships servicing system and ordered the part. Unbeknownst to you, the system has also booked the car in for the work, which will be done after dropping you off at the airport. Plenty of time to do that before the car picks up your daughter up from this evening’s soccer practice.
Cruising from airport, the system orchestrates many other activities. It’s already determined and applied the optimum tire pressures needed for the next trip after you unloaded the heavy cases and golf clubs. It’s applied a software upgrade for the braking system, participated in a ride-sharing program, adjusted the air-conditioning to save energy. It’s even prepared a combo of Taylor Swift and Katie Perry tunes for your daughter and her friends.
And the analytical marvel isn’t just consumer data. It’s also provides critical information within a broader sensor ecosystem. For example, that data on the potential failure has already been aggregated together with many others to optimize the inventory and logistics processes of associated service partners.
So, will the car of the future be a car as we currently know it? Or will it be an adaptive entertainment system, ERP constituent—even a mood regulator? The answer is all the above and much more. It’ll be self-governing car of course, but also a system where every sensor and data point coexist and co-operates with others across an extended analytical value chain—there will be no boundaries, and the question of ownership is blurred.
Artificial Intelligence for IT Operations
While the scenarios presented here belong in our future, there are lessons to be learned for how we should optimally manage IT applications and systems. Far from operating as a single cohesive system, today’s tools are narrowly focused on single technologies or functions. Of course, most use advanced analytical techniques, but that value is questionable when the insights revealed have a limited impact on the performance of the system as a whole—a system comprising multiple technologies, processes, functions and stakeholders.
To address this, modern Artificial Intelligence for IT Operations (AIOPs) systems will work very much like cars of the future. That is, they’ll be fully autonomous within a greater experiential ecosystem.
So, imagine your next IT operations journey …
Arriving refreshed at the office because you’re no longer dealing with late night fire-drills, your AIOPs system greets you with a prioritized list of service-level improvements and how they’ve contributed to improving customer outcomes. You’re no longer worried about lowering mean-time-to-resolution MTTR because MTTR doesn’t exist anymore.
As you settle back and start streaming service analytics to view cost-vs-revenue positions across three cloud services, AIOPs detects a potential problem with an important customer application. By correlating metrics across a data lake supporting the entire tech stack, the system determines that current capacity won’t meet increasing customer demand. But there’s no drama. The system automatically provisions additional containers to manage the load. It also calculates when and how these resources can be throttled back—like the driver-less car—accelerating and braking for maximum efficiency.
But with autonomous operations systems like these you’re not the only passenger; there are multiple optimization journeys simultaneously underway. Over in software engineering, AIOPs analytics are continually assessing the efficacy of coding styles and practices—from design to production. Feeding back critical information within DevOps automation toolchain to continuously drive improvements. Much like the car of the future will operate as an interconnected element, then so does AIOP—analysing the whole and piece parts at the same time—systems thinking in the extreme.
When we can achieve autonomous operations using AI and machine learning to manage all functions, it doesn’t mean that staff sit back or become “passengers” in the worst sense of the word. On the contrary, these solutions will help build a playlist innovation, knowledge sharing and organizational growth. A natural consequence when staff are freed from the routine, mundane and reactive fire-fighting where no one is ever in control.
Learn more at ca.com/aiops
*This post originally appeared on The Modern Software Factory Hub