Evolving component technologies will make Industry 4.0 viable
Manufacturing industries around the world are continually looking for initiatives that will boost productivity and, ultimately, profits. Initiatives like lean manufacturing techniques were widely adopted in the 1980s, while the 1990s brought outsourcing and offshoring of low-skill manufacturing to low cost countries. These initiatives have virtually reached the limit of how much they can improve productivity and, in the case of offshoring, even reduced profits since workforce and transport costs have increased in the interim. Manufacturers are therefore looking for a new type of productivity initiative to implement.
That initiative is coming soon in the shape of Industry 4.0, often referred to as the fourth industrial revolution as it follows the revolutions enabled by steam power, electric power, and electronic automation. While the previous revolution focused on the automation of individual machines or processes, Industry 4.0 is characterised by large scale connectivity, decentralised computing power, data collection and analysis in the cloud, and the flexibility of ‘smart factories’. These concepts are inter-related and all are the result of the increasing digitisation of industrial machinery.
As one of the central tenets of Industry 4.0, connectivity is the first step in implementing the other ideas. Adding network connectivity to previously isolated machines will ensure all pieces of equipment, no matter how important or intelligent they are, will be connected both to local networks and onwards to the cloud, so they can communicate with other machines and be controlled remotely. Decentralising computing power means making all the equipment more intelligent by integrating embedded computing into every individual machine, rather than having them obey a central computer. These intelligent devices will be able to talk to each other directly as necessary, and they will be equipped with extensive sensor and feedback systems in order to collect large amounts of data about environmental conditions, productivity and efficiency. This data is sent straight to the cloud, where there is abundant, cheap computing power available to analyse it and recommend courses of action for maintenance, or for streamlining and optimising production. Some factories will be responsive enough to entirely reconfigure themselves based on this data, while some will simply tweak performance attributes as required.
The result is a smart factory whose machines can monitor their own effectiveness and schedule predictive maintenance and upgrades; machines that can be used flexibly in different ways as products need to be changed or customised; machines that can identify individual products as they travel around the factory floor; and machines that can monitor and optimise their own energy efficiency.
The complex, challenging requirements of Industry 4.0 are driving innovation in industrial electronics in several key areas.
Racks and Enclosures
As control system architectures change to match the decentralised computing vision of Industry 4.0, control computers will need to be placed closer to, or inside the machinery they are controlling. While today’s central control computers are housed in traditional fixed racks and DIN rail cabinets, tomorrow’s machinery will need to be portable as it is moved around when the factory is reconfigured to respond to changing needs. A wider range of enclosure form factors will emerge to meet this need, while providing more efficient cooling options and improved EMC protection for tomorrow’s small and dense computing boards.
The smart factories of the future will depend on extensive sensor systems to collect the raw data on process states and machine status, both for real-time process optimisation and for long-term analysis of productivity. These smart sensor systems need to have a certain level of intelligence themselves, processing and interpreting raw signals before passing on data in a useable format to other parts of the network. Smart sensor systems will need to monitor a greater number of different parameters, with many different types of sensor required in the same piece of equipment, and the data will need to be processed concurrently in a form of ‘sensor fusion’, depending on the exact application. These systems may also have the ability to calibrate and optimise themselves as necessary.
Since the entire factory output is dependent on the accuracy of the data from the front end sensor systems, these systems need to be both precise and reliable, placing extra demands on the components. They will need to provide data in tough conditions such as under vibration and when exposed to electrical noise. Sensors will also need to continue on their path towards integration and miniaturisation; with more sensing required at every step of the manufacturing process, there is simply less room for individual sensor components.
Placing more sensors in industrial machines means a lot more data will need to be transferred, whether that’s within the sensor systems for the fusion of sensor data, or the processed data being transmitted to other machines, or for data upload to the cloud. Simultaneously, machines need to receive vital control information from the network in order to carry out their assigned tasks. All this data has to be transferred in real time. Cables and connectors which implement this network have to be more reliable than ever before, with industrial connector manufacturers increasingly turning to military-grade ingress protection, shielding and locking mechanisms to ensure ruggedness and reliability demands are met. Also, since the smart factory is reconfigurable, machinery is increasingly likely to be moved around as needs change, meaning connectors are expected to cope with many thousands more mating cycles than those in previous generations of industrial equipment.
Cables and connectors are also expected to cope with larger amounts of data travelling at much faster speeds than ever before. Industrial Ethernet connectors are already evolving to keep up, with a new coding system (X-code) designed to allow 10Gbps transmission over four twisted pairs. Some connectors are even being given a level of intelligence, such as signal connectors that can analyse the amount of data they are carrying, and report faults if this level drops below what’s expected.
New communications technologies and protocols are evolving to deal with the sheer volume of data that smart factory networks will need to handle. For example, optical fibre communication is increasing in popularity as it enables extremely fast transfer of large amounts of data over long distances, such as between factory sites, or for connection to cloud infrastructure. Optical fibre is also immune to electrical noise so it can be used in the most challenging of environments.
Wireless communications technologies are also starting to find a place in industrial equipment. Protocols designed for low power transfer of sensor data, such as Bluetooth Low Energy, may be useful in applications where sensors are installed temporarily to diagnose problems, or retrofitted. Technologies that don’t need wires are also very useful where the sensors are on moving parts, such as robotic arms, where cables might restrict movement or potentially get snagged.
Industry 4.0 at a glance
Data access and predictive maintenance
Leverage 'big data' to control schedule and processes in real-time
A single connected infrastructure for all processes allows easier expansion and visibility beyond the factory floor
Physical and cyber assets secured across a single network. Ruggedised industrial infrastructure minimises risk of down time
While the exact nature of Industry 4.0 is still emerging, the electronics industry is already responding with flexible enclosure systems, smaller and smarter sensors, more reliable connectors and faster and more convenient wired and wireless communications technologies. These technologies are set to enable the fourth industrial revolution, promising flexibility, productivity and energy efficiency advantages to the factories of the future.