Predictive maintenance & condition monitoring
The future of maintenance
The data evolution has opened up new pathways for businesses to upgrade their production. In this article, Senseye’s CEO, Dr Simon Kampa, describes how the future of maintenance centres on cloud-based applications, powered by AI, sensors and the industrial IoT
A
fter a period of incredible hype, the evolution of the growing uses of data within industrial manufacturing have
begun to illuminate the path for many businesses to upgrade their production and wider operational environments.
A cornerstone of this change has been the
opportunity to connect and take data from manufacturing assets in order to monitor both the condition and performance of machinery and plant. And while the results of this first phase are set to be substantial - Capgemini expects these practices to deliver $500 billion annually in added value around the world within the next five years - there is a clear need for automation at the earliest instance. Indeed, the potentially huge scale of industrial
data from factory machinery will demand automation to realise savings at scale.
GETTInG buSInESS-SpECIfIC: prEDICTIvE mAInTEnAnCE Consider the relatively focused realm of predictive maintenance: a practice that relies on analysing data produced by industrial machinery to spot the vital signs that show when monitored assets will fail in the future. Until recently, predictive maintenance was a time consuming, manual and expensive process requiring large teams of data scientists, often removed from the operational frontline of a business. It was used in industries where regulation demanded it, but most
38
manufacturers could not capitalise or scale on the opportunity. It was simply too expensive and too complex to gather and analyse enough data to deliver tangible results. However, by introducing machines capable of
recording their own vital statistics with sensors or Industrial IoT platforms, these costs have dropped dramatically. Manufacturers can now collect large amounts of data from across their production environments and have it put to work to create a meaningful course of action. Identifying the anomalies in this vast amount of
data is now the problem. Many businesses underestimate the issue of complexity and making sense of such massive amounts of data. Again, automation is a key strategy. Cloud-based computing specialised applications now exist to undertake much of the heavy lifting for manufacturers in a series of tangible steps: Data selection and storage – taking the
meaningful data direct from the sensors Data transformation – converting the
raw data so it can be processed to automatically provide: Condition monitoring – alerts based
on an asset falling outside prescribed operating limits. Asset health evaluation – a diagnosis
based on trends over time analysis if asset health declines. Prognostics – failure predictions
based on machine learning to estimate remaining life.
October 2019 Instrumentation Monthly Decision support system – a series of
proposed best actions in the face of the above. Of course, all of the above is hidden away from
the users, as modern software interfaces are designed to enable meaningful information that is both easy-to-understand and easy to manipulate. Under the hood, a specialised cloud-based
computing application learns the characteristics of each monitored asset by analysing data outputs and trends related to asset key indicators such as vibration, heat, moisture, pressure, temperature, energy consumption and torque levels. The machine condition is continually compared
with data on known maintenance events for that machine and a relevant range of others like it. This end goal is for engineers or technicians to spot emerging problems up to six months ahead of time, identify when machines are most likely to fail and align maintenance accordingly so that business continuity is maintained.
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 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82
