A machine learning algorithm to improve alarm management? Gadzooks, what will they think of next? But Johnson Controls (JCI) has developed the product, and is marketing it.
Here’s the story: JCI is marketing a false alarm reduction service that applies machine learning algorithms to intrusion alarm panel data. It provides end-use customers with insights to prioritize actions that are said to vastly reduce preventable nuisance alarms.
To train its algorithms and confirm their accuracy, the company utilizes a massive amount of closed loop data. They took five years-worth of that data from the 757,000 panels they have in the field. A customer beta-testing phase followed. That lasted a year-and-a-half, then bore results that JCI says can give end users data-driven recommendations to reduce 50-70 percent of their false alarms.
This is good stuff. At SIAC, we’re excited about the possibility of such algorithms when properly applied taking us to the next level of alarm dispatch reduction. That’s excellent and we applaud these steps.
Great job JCI! Kudos for digging in and finding ways to deliver alarm services while helping to improve alarm management practices.
SIAC also fully supports JCI’s next steps as it rolls out the solution to a limited number of businesses. It’s a Cloud-based, software-as-a-service offering. Physical deployment isn’t necessary. The software system should work with any panel vendor. That’s important, and also an excellent feature to ensure penetration in the security industry.
We’re fascinated to see this hitting the market, and looking forward to see how it is adopted and utilized by others. The success will occur when we see high utilization rates. A great reason to be optimistic about the future!