How COVID-19 Changed the Way We Shop and Accelerated Automation in the Retail Sector
All of us have had to endure the social and financial consequences of COVID-19. To maintain personal safety, consumers increased their online shopping and decreased their in-store interactions. Retailers, especially essential businesses that were open during the height of the pandemic, had to find ways to satisfy increases in consumer demand while implementing new safety measures for their customers, as well as their employees.
Not only has shopping continued during the pandemic, it grew vastly over a very short period of time as consumers first rushed to stock up on staple items and then found themselves supplementing spend at business like bars and restaurants at home. As social distancing guidelines persist, the businesses that are able to support shopping methods that keep the consumer out of the store, find themselves capturing a much larger portion of that spend. Examples of those online shopping methods are BOPIS (Buy Online, Pickup in Store), Curbside Delivery, or direct ship to the consumer. Companies that adopted these service delivery models reaped the benefits, while others who did not were hit hard financially with a lot of businesses closing shop.
From a retailer’s perspective, COVID-19 accelerated investments in automation to gain operational efficiency and reduce labor costs. Prior to COVID-19, forward-looking organizations were investigating innovative technology (such as robotic picking and automated inventory management) that was planned for implementation in 3 to 5 years. However, the health risks that COVID-19 presented to retail store employees, the window to increased automation has shrunk dramatically. Demand spikes (holiday season shopping) conventionally have been solved by “throwing people at the problem”, now have become extremely hard to support during a pandemic when people are afraid to come to work.
At Cognex, we saw this happen in real time as our logistics operations customers relied on our industrial automation technologies to keep pace. Because so much of the response to COVID-19 set the stage for innovations that will reverberate for years to come, it’s worthwhile to look back on how the pandemic changed retail — and how Cognex technologies enabled the transformation.
Let’s take a look at how retailers evolved to confront the pandemic.
The Transformation of Retail Stores
As automation becomes more central to conventional retail, stores are transforming into distribution centers (DC). Next-generation tools like deep learning and robotics are helping to automate complex activities like removing packages from cartons and placing them on shelves. The grocery industry, which tends to run on thin margins, is testing automated fulfillment strategies. Here are a few examples:
Two large grocery chains are testing small automated "micro-fulfillment centers" in the backs of their own stores that are dedicated to fulfilling deliveries and pickups. At two of their grocery stores, Albertsons is using robots to prepare customers' orders, which it says speeds up the picking process.
Another example is the use of compact, robot-based automated storage and retrieval systems (ASRS) that support goods-to-person or goods-to-robot picking. This technology is being used by H.E.B, a large grocery chain of over 400 stores in Texas, to support its micro-fulfillment facilities rollout.
The use of robotics and deep learning provide considerable cost and speed benefits versus manually picking items off store shelves, particularly now as the COVID-19 pandemic has triggered big increases in demand for online shopping and grocery retailers want to keep their workers protected while addressing customer’s ongoing need for fast fulfillment.
Cognex Technologies Enabling Post-COVID-19 Retail
In increasingly automated retail environments, every product needs a unique identifier that must be traced accurately from the manufacturer’s shipping dock to the customer’s front door. Advanced machine vision cameras, barcode readers, and sensors like the ones developed by Cognex make it all possible.
Cognex helps retailers reinvent themselves in four crucial technology categories:
Barcode reading. Barcodes and other unique product identifiers are mainstays of retail logistics. A scanning device quickly and accurately captures the data in a barcode and submits it to an automated distribution system that traces it on every key step of its journey.
Logistics facilities must scale to meet increased demand. One way to do this is by increasing throughput (read rates). Barcode reading systems must accurately read codes, no matter their condition, or where they are on the package or item. Using Cognex’s patented 1D and 2D algorithms, found in Hotbars and PowerGrid barcode reading technologies, Cognex DataMan image-based barcode readers deliver best-in-class barcode reading performance and real-time monitoring to analyze the root cause of no-reads for further optimization in the future.
Machine vision. Digital cameras built specifically for industrial automation capture 2D and 3D images that help ensure items and packages are sorted to the correct material handling systems and fulfilled accurately. Advanced scan tunnels combine multiple vision systems and barcode readers to visualize all sides of a package, enabling real-time inspections and allowing damaged packages to be routed to the proper area for repair or replacement.
We have over a decade of experience developing and delivering complex machine-vision solutions for distribution centers across the retail sector. These solutions use presence/absence detection to help increase sorter tray efficiency, and capture dimensional data that can be used to solve a variety of applications across a distribution facility, such as verifying manifests of incoming goods, deciding if an item is a box or polybag (proper handling), and estimating shipping costs and optimal packaging solutions.
Deep learning. Barcode readers and machine vision cameras provide immense quantities of digital data that a deep learning vision system can analyze to automate processes that have a high degree of variability.
Given the way that COVID-19 has changed the way consumers shop, retail distribution and e-commerce facilities are more rapidly evaluating the use of deep learning to positively identify objects without having to read a barcode. This capability has value in automated inventory management applications.
This removes the reliance on humans to perform this work and allows the facility to run these processes continuously throughout the day to ensure online inventory numbers are updated in close to real-time.
Many retailers are now testing automated inventory robots that use a combination of deep learning, barcode reading, and OCR technology to do inventory management.
Robotics. The fulfillment phase of a DC typically requires people to pick products and place them into boxes for shipment to buyers. This task has so many variables in shapes, sizes, and weight that it’s extremely difficult to automate.
Though robots move products through a distribution center today, it still requires a human to pick items off the shelf and pack into a box or pick items out of a tote and replenish stock before the next day’s orders are fulfilled. The final frontier is automating the entire process of fulfillment: training learning algorithms to outperform human eyes and robotic grippers to perform as well as hands and fingers. Cognex’s industrial vision systems and deep learning software will be essential to moving robotics into that phase.
Learning the Lessons of COVID-19 in Retail
Consumers rapidly adjusted to online delivery and curbside pickup in response to the COVID-19 pandemic. Retailers and their logistics teams dived into new processes and policies at the same time.
The common thread for everybody is that the public health emergency revealed better ways to do business and please customers. It’s reasonable to suggest the new habits learned in 2020 will persist for years to come.
As consumers and retailers adapt to innovations in automation and robotics, Cognex technologies will be there all the way.
 “Grocery stores turn to robots during the coronavirus”, April 7, 2020, www.cnn.com
 “H.E.B Goes Micro With Auto Store Rollout”, September 15, 2020. WinSight Grocery Business