Edge Computing

At the granularity of individual product level, logistics companies need to know their stock at all stages of the supply chain and at all times so they can meet the growing demand for visibility and enable increasing e-commerce. To address this, logistics providers as well as retailers use IoT devices to monitor temperature, track real-time location, and watch stock levels in order to make data-driven business decisions.

There are numerous opportunities to increase resilience by leveraging edge computing in various supply chain processes. During transportation (trucking and last-mile delivery), an autonomous vehicle is vulnerable to increasing cyberattacks; if vehicle control is compromised, this endangers other road users and the shipment itself. When a self-driving vehicle is connected to the edge, however, it is able to react to situations in real time, rectifying any malfunction and correctly responding to cyberattack without requiring human intervention.

Within a warehouse, edge-enabled devices share and process data in real time. This improves the speed and accuracy of warehouse operations. For example, edge-enabled cameras can scan barcodes on individual pallets to monitor all stock per micro-location in the facility. The cameras can read barcode metadata and indicate if a box has been placed in the correct location. Crunching metadata in real time, the analytics platform can trigger an alert of fraudulent barcodes and incorrect placements. This data can then be streamed, connecting the edge-enable devices to the warehouse management system and enterprise resource planning system via the cloud, so that employees are actioned to rectify the situation.

Another good example is how edge computing supports the efficient management of volume fluctuation during high seasons, helping warehouses to cope with a surge in demand for consumer goods.

Manufacturing and warehouse operations benefit from the way edge computing enables close monitoring. For example, watching the efficiency of equipment and production lines helps to detect failures before they occur, avoiding costly delays caused by downtime. In the energy sector, companies can keep a close eye on assets in order to avoid disruption in meeting consumer demand and overcoming global shortages.

Within a traditional operational setup, closed-circuit television (CCTV) cameras in a warehouse continuously output raw video signals which stream constantly to a cloud server. From there, a motion-detection application identifies any footage containing activity, ensuring this is saved to the cloud server – this represents a constant strain on the warehouse’s internet infrastructure as large bandwidth is needed to transfer all this footage; it also imposes a heavy load on the cloud server which must process all camera footage simultaneously. Edge computing offers an excellent alternative. If each camera’s motion sensor computation is moved to the network edge, using its own internal computer to run the motion-detection application, the camera need only send this footage to the cloud server, resulting in a significant reduction in bandwidth use. Now the cloud server would only be responsible for storing footage of significance and it would be able to communicate with a larger number of cameras without being overloaded.