The factory of the future will have only two employees, a man and a dog. The man will be there to feed the dog. The dog will be there to keep the man from touching the equipment. Warren Bennis, 1991.
Digital transformation is accelerating the transition to smart factories and affecting nearly every stage of the manufacturing process. It is driving exponential growth of data-driven sensors and increasing the use of mobile devices in plant operations. Traditionally-separate operations technology and IT systems are converging. Along with this comes increasingly-stringent compliance requirements for data security, intellectual property, and asset tracking. The Internet of Things is enabling continuous monitoring of tools, inventory, and WiP throughout the manufacturing process without human intervention. Last year, the manufacturing industry invested $183B in IoT. Included in this broad categorization of things are pervasive robots and surveillance cameras.
Internet-connected devices and tools can be located anywhere within the factory operation, including indoors on the manufacturing floor, within equipment mezzanines, and outdoors. An Internet-connected torque wrench, for example, may be disconnected from the network at one end of the plant, and then reconnected somewhere else. The network must provide the flexibility to add, remove, and relocate IoT devices anywhere in the factory, without requiring a maintenance window.
Production facilities now run the gamut from small specialized plants, to Tesla’s enormous 5.5 million square foot Gigafactory. It is critical that secure Wi-Fi RF signals blanket these facilities to connect all the IoT devices and sensors. Production must remain up 24-hours per day. Any events that could bring the network down, such as security breaches, rogue devices, or hardware faults must be protected against. Artificial intelligence and machine learning are now being applied to insure reliability and manageability of the network in these environments.
To connect and enable the smart factory, a strong wired and wireless network infrastructure is critical. The requirements for this infrastructure include flexibility, reliability, security, and easy system management. The smart factory consists of thousands of IoT devices, spread throughout the plant. The network must provide complete coverage and accommodate high-density connections. The smart factory network must provide an adaptable topology to accommodate reconfiguring production lines on the fly, without shutting down for maintenance. It is important that management have clear visibility into all network devices and application traffic, so that when a rogue device is connected, or an intrusion occurs, it is flagged immediately and isolated, preventing any possibility of bringing down the network.
In the highly-reconfigurable environment of the smart factory, the ability to flexibly onboard internet-enabled devices, tools, and sensors is an important requirement. Device authentication based on network policy, rather than long, complicated authorization tables of thousands of devices and users, keeps both productivity and security high. Automatic provisioning eliminates the possibility of human error when new tools or devices are added to the network.
Hypersegmentation as implemented in fabric networks delivers a higher level of security and permits different robot and plant tool vendors to be assigned to separate securely-isolated zones. To provide the reliability demanded by the smart factory, fabric-based networks build in resiliency and redundancy. Should a fabric node go down due to mechanical accident, the topology will automatically reroute within 200ms. With this high level of security, policy-based network access control, and hypersegmentation, the smart factory floor can be on the same network as the business offices, helping to integrate IT with OT.
There’s also a growing demand for network-based location services, as provided by today’s Wi-Fi access points and BLE beacons. These services make it possible to track and display where all the devices and workers are located throughout the factory. The multiple communication standards, complicated by dynamically-changing production areas, create a challenge for maintaining a high-quality RF connection between devices and access points. To meet this requirement, AI and ML are being used to optimize the Wi-Fi signals without human attention.