DCN June 2016 | Page 26

Big Data & IoT For example, on a sandwich biscuit production line, the biscuit sandwiching machine at the heart of the line should be able to communicate with the previous elements of the process, as well as the ones that come after it. The mixing, cutting and baking machines at the very start of the production process should also be able to ‘speak’ to the conveyers, the pile packing sandwich machine, the cream feed system, lane multiplication and packaging machines. This level of communication allows the production line to be more flexible and cater for a wider range of biscuit varieties. Regardless of whether we’re talking about biscuits, automotive manufacturing or even smart grids, IIoT has communication requirements that go beyond the standard client/ server needs and conventional thinking. Instead, the nodes act as peers in a network, each making decisions and reporting to other nodes. Besides performing core tasks, the production system is also connected to an enterprise level that can automatically issue alarms, collect and analyse data and even make predictions or recommendations based on this analysis. Of course, this mass influx of data is not without its challenges. Big Data flexibility The increased amount of sensors on the factory floor facilitates a complex level of machine communication, having the potential to improve operational productivity and efficiency. However, even greater value can be derived from unlocking the valuable datasets an IIoT enabled facility generates. Research published by Accenture and General Electric stated that nearly three quarters of companies are investing 20 per cent of their technology budget into Big Data 26 analytics. Exploiting data from IIoT is a key part of the analysis plans of those organisations. The introduction of Big Data is often regarded as a significant financial investment, but the reality is companies don’t need to fear this change. In fact, setting up a Big Data platform to leverage IIoT can be done without the need for a large upfront investment and often starts paying off straight away. Some organisations may choose to begin by implementing their Big Data strategy with a small proof of concept, to gain insight of which data combinations are valuable, before scaling to a broader enterprise solution. Increasingly, companies are also turning to cloud based approaches, as they provide greater flexibility to scale up and down as the business requirements evolve and change. Regardless of the approach, organisations must consider exactly how the data can derive actionable insights and in turn, benefit the business. A common language IIoT will only work if it uses a compatible language across systems and industries. To help achieve this objective, industry giants AT&T, Cisco, General Electric, IBM and Intel founded the Industrial Internet Consortium in 2014. The Consortium aims to accelerate the development and adoption of interconnected machines and intelligent analytics. As IIoT cuts across all industry sectors, from manufacturing to energy, common standards, harmonised interfaces and languages are crucial for successful implementation of the concept. The consortium hopes to lower the entry barriers to IIoT by creating a favourable ecosystem that promotes collaboration and innovation. The next step is to facilitate interoperability and open standards, allowing machines or systems from different original equipment manufacturers (OEMs) to communicate with each other and with control systems. The old and the new Perhaps one of the biggest challenges when it comes to implementing IIoT on a larger scale comes from integrating legacy systems with the latest generation of smart factory equipment. Learning new things changes the structure of the brain and similarly, in manufacturing, implementing new automation equipment usually results in changes across the entire system. The solution is to use standards based protocol gateways to integrate legacy systems in brownfield environments. This allows organisations to free data from proprietary constraints and use it for real time and historical data collection and analysis. There is as much risk in sticking to a single vendor based on current install base as there is to accepting these new concepts with multiple new vendors and interoperability between intelligent devices. Their concept is something that we have experienced greatly within the energy and infrastructure sector and the concepts behind IEC61850 and interoperability. Much like the human brain, the Industrial Internet of Things is always changing and there are still a lot of questions to be answered before we fully understand its requirements, implementation and potential. Luckily, these conversations are taking place and new ideas are put into practice every day. The next step is to figure out an easy way of practically implementing IIoT innovations in manufacturing environments across the world.