Internet of Things
Industry 4.0 - beginning ROI with data
By Tom Canning, VP of IoT at Canonical, the company behind Ubuntu
Things (IIoT) or Industry 4.0, is the era of smart machines and automation. There is now potential for factory machines to have the capability where they can self-diagnose themselves, leading to predictive maintenance and dramatic reductions of industrial inefficiencies in the factory. The word ‘potential’ here, is key. We’ve been talking about automation and AI for some time within an industrial sphere, but actually its practical deployment and the real insights driven by it are not being seen on a widespread scale. The value of adoption has been much discussed, and proofs of concept (POCs) have been documented, but how do we move beyond this into mainstream deployment?
The adoption myth Tom Canning I
n the case of industrial revolutions, it is impossible to enter a new phase without there being winners and losers as a result of disruption. Industries move at different speeds, as do the businesses within them. As such, some fall by the wayside as new incumbents take centre stage. Manufacturers today know their industry is currently undergoing a revolution, and some are naturally nervous about what the future holds. The Industrial Internet of
30 December/January 2020
Bain & Company carried out its customer IoT survey last year, which showed that its industrial customers expect widespread implementation of IIoT to be slower than they did in 2016. So whilst the years have advanced, companies do not necessarily think tech has grown with it, and consequently deployment seems more complex.
While IIoT opens the door to a host of new opportunities such as cost reduction, quality improvement and business growth, the prospect of putting it into action clearly causes apprehension. Implementing IIoT solutions introduces changes not only in a factory’s IT environment, but also in the ways IT interacts with production systems and field devices. Thus, it can be
Components in Electronics
difficult to grasp the complex requirements and considerations. The value risk of IIoT is still perceived as being too high for a majority to form. This is the case because mainstream adoption would require sizeable divestment in legacy industrial equipment in conjunction with ambitious investment in the likes of IIoT equipment such as gateways.
Grabbing hold of data
Since shopfloor equipment is expensive and needs up to one year for initial calibration and optimisation to become productive, substitution timescales are very long. Additionally, every new investment needs to be economically justified and has to prove its return on investment (ROI), which may be an additional challenge in an already highly optimised environment. This creates a chicken and egg situation, but we can only show ROI by moving into adoption, particularly when looking at data. Historic data enables retrospective analysis on industrial operations, with two key benefits: optimisation and planning. Following the implementation of field connectivity for data collection and automation, data analytics algorithms must be executed. Analytics will lead to new knowledge about the production processes, such as new quality and maintenance patterns, prediction of a machine’s parameters and more. Moreover, data analytics can be used in order to simulate or even drive the behavior of field devices. Correlating time-series data from
production resources with industrial productivity metrics allow a granular understanding of areas of waste and inefficiency. Actions can be taken on these correlations to quickly obtain productivity
gains. Historical data combined with condition monitoring is a prerequisite to apply predictive maintenance, one of the big promises of IIoT.
Having said that, you can’t see the benefit of something you don’t have, and we won’t realise these promises until we see wider adoption. Bigger companies might be waiting around to see more POCs and successes elsewhere, but this isn’t going to accelerate their own ROI. By updating equipment, and investing in the gateways needed, businesses can begin efficient data monitoring and in-turn - their own ROI journey.
Focusing on the wins Through machine to machine communications, manufacturers can now have a treasure trove of data insights at their fingertips. Advances in big data analytics will mean that AI-assisted systems can dig through masses of data to produce valuable insights that were not available previously.
It should be noted that the shift to IIoT and digital manufacturing is a strategic commitment and a long-term journey, not a single project. The deployed solution should be continually tracked to ensure proper operation and to identify any need for additional developments and fine tuning. An agile and iterative approach to developing data monitoring and deploying additional features should be used to boost continuous improvement. Companies need to be brave enough to not wait on POCs and move to broad adoption. Data modelling is one of the easiest ways to show ROI, but unless we start now, the return is only going to move further away.
ubuntu.com
www.cieonline.co.uk
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56
