Edge-based AI solutions that drive efficiency in real-time
Enterprises continue to get access to large budgets to invest into new emerging technologies like AI, however, the businesses that benefit the most are the ones that view AI from the perspective of enhancing short-term business capabilities rather than utilizing it as an R&D project. To do this effectively, organizations needed to have structured databases and organized repositories like data lakes so that the wider organization could also utilize and extract value from the adopted AI solutions in an efficient manner. This is no longer the case, as fortunately when AI moves to the edge, the cognitive intelligence available across an organization also moves to the edge.
There is no longer a need for the central team to spend days slicing and dicing data to find pathways towards ROI, when in fact edge-based AI solutions can help employees across the organization solve problems in real-time. A good example of this is one of the biggest fields of artificial intelligence – computer vision. This subsect of AI seeks to understand and automate tasks that the human visual system can do by analyzing raw images or video data. If there are CCTV cameras, this software can be deployed with unparalleled ease across millions of square feet of commercial, residential, or retail real estate. Some of the features that computer vision can assist are object recognition, image classification and pattern detection. These data points can now be unlocked in real-time across a business’s entire footprint – giving operators access to new data points that they did not have access to historically.
Where this technology gets exciting is that the data harnessed can now be translated in real-time into actionable insights that can team members on the ground. Imagine the workflow of an airport operator looking to manage the flow of passengers from check-in to boarding efficiently. Right now, they must depend on their command center or agents on the ground to monitor traffic and queue times to see where the bottlenecks in the passenger journey are. This problem gets even worse when you have hundreds of thousands of daily passengers across several zones ranging from check-in counters, security, lounges, and gates.
An airport operator can leverage an AI-enabled solution like computer vision to amplify the efficiency of their operations in a few different ways. One way is through real-time alerts where the AI will send the respective check-in, security or gate agents alerts on their instant communication headset telling them exactly what needs to be done to improve the flow of passengers. Has the queue length or wait time exceeded their SLA? If so, it will tell the manager where and when they need to open an additional counter. The other key element of such software is its ability to provide predictive and prescriptive analytics where the AI will use a combination of historical data and knowledge of the business outcomes to accurately predict the future and in turn guide managers to the best overall decisions. For example, this will help them pre-empt which zones and times of the week have the most congestion, and it will then help by accurately forecasting the number of security and immigration officers that are required according to each hour of the week. Not only will this ensure the passenger has a better experience, but it will also make sure they optimize staffing schedules to match forecasted demand.
We are just scratching the surface!
Additionally, these algorithms optimize perpetually over time, turning actionable, real-time insights into rich trend reporting that several stakeholders can utilize. This in turn informs recommendations that allow operators to impact their business both in the moment and with an eye on future outcomes. The pathways towards ROI are also evident across other physical environments such as big box retail, shopping centers, QSRs and warehouses. Therefore, I am confident that we are going to see a lot of enterprises across India start to adjust their long-term strategy with data-driven predictive insights from AI-enabled solutions like computer vision.
(The writer is SVP of Strategy & Business Development, at Deep North)