How Artificial Intelligence Can Benefit Retail Security

Retail security is one of the benchmarks for today’s retailers as it affects customers’ loyalty, their shopping experience, and the reputation of malls and stores. As such, retailers not only strive to provide a wide variety of goods and fast high-quality service but also ensure safety and protection for their customers and staff.

Stores are large public areas with heavy foot traffic, which makes them vulnerable to many threats like thefts, shoplifting, and violent incidents among others. Given that shopping malls are complex environments, come in different sizes, and have different settings, they require a customized loss prevention and security approach to address their issues. That is what artificial intelligence has to offer. Let’s take a look at vulnerabilities that exist in retail security and explore what AI has in store for retailers to address their security issues and prevent losses.

Retailers are using AI worldwide. According to the recent research report, the global artificial intelligence in the retail market in 2019 was approximately USD 2.7 Billion and is expected to reach around USD 20.05 Billion by 2026 growing at a compound annual growth rate of 39%. The applications of machine learning range from cashier-free services and product search solutions to chat-bots and virtual fitting rooms. AI capabilities are also utilized to analyze a huge amount of data to provide deeper business insight.

For instance, retail video analytics can track a customer’s location and actions in a store, analyze them and create heat maps further used for smarter merchandise planning, price optimization, and stock forecasting. With the insights from AI, stores get a clear vision of how a certain supply chain works, can detect inefficiencies, and create ways to make it better. In addition, AI analytics helps to boost sales and improve the shopping experience by providing smart recommendations, sending customized advertisements and personalized offers.

Artificial intelligence and machine learning are also harnessed to improve the physical security of shopping malls, stores, and retail supply chains. They empower security professionals to better understand the security landscape, make data-driven decisions in order to respond to threats before they happen and prevent losses.

Loss prevention

One of the major issues for loss prevention professionals is inventory shrinkage. That is a combination of employee theft, shoplifting, vendor fraud and other errors. According to the latest National Retail Security Survey report, which analyzed theft incidents from 118 of the largest U.S. retail chains, United States retailers lose over $31 billion a year due to retail theft. The two main reasons for inventory shrinkage are employee thefts and shoplifting. The study, conducted by the University of Florida discovered that retail security managers attributed more than 48.5 percent of their losses to employee theft, which cost retailers a record $15 billion. Shoplifting was responsible for 31.7 percent of retail losses which is equal to nearly $10 billion last year.

To prevent inventory shrinkage many retailers are using security technologies and enhanced video surveillance in the first place. Fortunately, cameras are usually already in place at stores and are used by security personnel to monitor the site. But the size of shopping malls and a great number of customers visiting them every day especially during the holiday season leave stores vulnerable to thefts and pose an overwhelming task on security officers who have to constantly watch all the footage from the hundreds of cameras across the store, maintain concentration to detect the cases of suspicious behavior and stay alert in order to provide a proper response to the situation. That is just beyond human abilities. Fatigue and human mistakes are inevitable and consequently lead to missed detections and losses in inventory.

Powered by AI technologies, video surveillance cameras can be effectively used for loss prevention by big and small retail businesses. While the traditional approach to video surveillance using CCTV cameras can offer the opportunity to keep track of the premises and respond rather reactively, for example, for investigating the cases of thefts and shoplifting after they occur, integrated with AI technologies cameras turn into an effective tool of a proactive approach to security. Smart algorithms provide a situational analysis of camera feeds for any potentially suspicious behavior like fidgeting, restlessness that can result in shoplifting and alert a security center to a potential threat. Thus, the implementation of artificial intelligence helps security professionals have a clear understanding of what is going on across a vast area of the shopping center and the entire supply chain and improve inventory control.

Slip and fall detection

AI video analytics can also be effectively used to address slips and falls which are the most common shopping center accidents. They can be caused by slippery floors, spilled drinks and food, uneven floor mats, and other obstacles. Artificial intelligence can detect slips and falls, alert the staff so that they can immediately respond to the incident, and provide the necessary help.

Edge AI solutions in retail security

A new trend in physical security is to utilize edge AI solutions. They assist both small and big retailing businesses to expand their physical security capabilities. They are critical to be deployed in case of unstable connectivity in every area of the shop floor and in the warehouse. Being much cheaper compared with equivalent resources in the cloud due to savings on network bandwidth and hardware, edge AI provides effective real-time analytics. AI algorithms analyze data from cameras directly on the edge device and trigger alerts when a threat is detected. Thus, thanks to real-time video analytics, they can guarantee security and prevent theft for example at automatic pay stations. Edge AI solutions are also critical to prevent inventory losses. When used on drones, they can scan the stock, count the elements, identify wrong placement, and automatically initiate stock provisioning if required.

Mall and store-related violence

According to the statistics, in the last 5 years, another threat that has become more of a priority for retailers emerged. 82% of LP professionals pointed that they are more concerned about mall and store-related violence. Even with widespread store closures, the US saw an increase in retail fatalities and violent behavior including active shooter incidents in brick-and-mortar retail stores. In 2020 there were 485 violent incidents, which is 14 percent higher than the previous year. They resulted in 523 fatalities, of which 76 percent were customers, store associates, and security personnel.

Losses of life could have been reduced if such AI-powered technologies as object detection and anomaly behavior recognition had been utilized. They provide stores and malls with round-the-clock surveillance and intelligent video analytics to detect threats of different kinds and automatically trigger an alert. That draws the security officers’ attention to those incidents as soon as they happen and gives them the opportunity to respond to the situation adequately to prevent its escalation. Intelligent video analytics is available in many different forms and can be used to detect weapons, intruders, bags left behind, read license plates, monitor perimeter and identify unwanted behavior like fights and violence.

Securing parking areas

Intelligent video surveillance either in the cloud or at the edge is vital for securing parking areas. Busy parking lots in shopping centers are full of safety risks. They vary from auto thefts and accidents caused by vehicle and foot traffic or a few stray shopping carts to loitering and the risk of a bomb attack as leaving a secondary device in the garage area has become a common attack strategy. Shoppers going back to their vehicles with bulky items and numerous purchases present inviting targets for thieves as well. Let alone joyriding. Without proper surveillance, parking garages can turn into racing tracks which in the worst scenario can result in death and seriously injuring visitors of the mall. Vandalism and damage from graffiti complete the list of issues that store security has to address.

Leveraging AI capabilities, security officers can mitigate those risks. Due to real-time monitoring and AI analysis stores can improve situational awareness in their parking lots and garages. Being alert to various threats and behavior anomalies security staff have time to make decisions to respond to developing situations and protect lives, property and prevent reputational loss for the store.