Plant Management
This is where IoT is really exciting and can play a big role as it forces down prices and democratises data access, delivering a solution enterprise-wide.
Fortune.com highlights that IoT could deliver maintenance savings of 25%, reduce unplanned downtime by 50% and extend the life of machinery by years. Gartner forecast 30% TCO savings through the use of ‘smart’ technologies.
Coupled with advances in analytics and low cost communications infrastructure, the time for providing sophisticated and affordable predictive IoT systems is now. There will always be a need for specialist prognostics systems but most will benefit from a ‘good enough’ cost-effective solution – think of the Pareto principle and how camera-phones (once thought of as an impractical joke) have dramatically reduced sales of point-and-shoot compact cameras.
IoT does not sit in a technology silo: it demands organisational and business model changes. That alone will turn off large swathes of businesses. However, benefits will start becoming too overpowering to ignore and organisations will have no choice but to adopt or fail. Business will need to create their own ‘Internet of Assets’.
At Senseye, we are developing a product to do just that, delivering the benefits of advanced predictive analytics to organisations that have previously considered it unaffordable or too difficult to implement and work with.
Data mining approach
Senseye takes real-time data from IoT devices, machines and the environment and mines the data for complex patterns, trends and anomalies. In a machines environment, this tends to be accelerometers to collect vibration data, together with temperature and humidity feeds. If they exist in machine-digestible form, then maintenance records, usage
patterns and spares management data would help make the results more accurate and uncover more actionable insights.
Careful pattern analysis, correlation and regression makes it possible to detect subtle signatures in the data to point to impending failure or sub- optimal operation. While it is true that the system gets better the more time it ‘sees’ failures, anomalous behaviour and basic trending can be performed from the outset. Moreover, Senseye is using the collective analysis of all the machines and equipment connected to its system to help refine the algorithms so they work better for everyone with a dramatically reduced ‘time to first insight’.
Being able to predict failures and find ways of optimising the use of machines, whether in manufacturing or elsewhere, has serious consequences in terms of maximising machine usage, slimming down maintenance and spares management regimes and increasing productivity – all of which drive underlying profitability. In an increasingly globalised marketplace and supply chain, those that ignore such technology advances do so at their peril.
However, delivering this technology to previously excluded organisations does have design implications. For one, an engineering department is unlikely to exist to interpret data and decide on what needs doing. This means that results must be succinctly described in plain English, direct to the user (even by text, if necessary) with no more hunting through graphs or Excel spreadsheets! Also, system configuration and deployment has to be ridiculously easy. These two key areas are where Senseye are seeking to really differentiate ourselves. The company is currently running exclusive trials in three sectors: agriculture, solar and renewables and manufacturing. l
Simon Kampa is CEO of Senseye.
www.senseye.io www.engineerlive.com 17
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