We can provide advice and assistance at each stage of your journey either directly or through our network of industry and academic partners.
The rapid growth in digital technologies such as Internet of Things (IoT), Cloud, Machine Learning (ML), and Artificial Intelligence (AI) has opened up many opportunities for industries and organisations to deliver significant value and transform their ways of working.
One such opportunity is Predictive Maintenance. The fields of asset management and maintenance, ranging from everyday appliances to the largest industrial engine, can be taken to a new level of value through the application of predictive analytics.
Moving from the basic approach of Corrective / Reactive Maintenance (fix on failure) through Preventive / Planned Maintenance (maintenance and repair on a schedule based upon expected component lifespan) towards Predictive Maintenance can take time and increase overall complexity but the significant benefits offered make this a journey worth pursuing.
GET IN TOUCH
For further information, please contact us.
Predictive Maintenance directly monitors the condition and performance of assets during normal operation to reduce the likelihood of failures. It seeks to predict when maintenance is required; maximising the lifespan of an asset; reducing costs; and mitigating against unplanned breakdowns. Following a PdM approach requires real time monitoring of machines and components. Maintenance is then performed when issues, which are expected to impact future performance, are detected.
There has been a significant growth in interest in Predictive Maintenance over the last 18-24 months and there is a wealth of resources available from a wide range of technology, engineering, and consultancy companies who have developed and/or implemented PdM solutions.
Scottish Industry References
GE Renewable Energy pioneered and developed the digital wind farm to yield the most valuable outcomes for their customers—improving performance, lowering risk and reducing cost. By applying advanced analytics and wind power software built on deep expertise, they are accelerating reliability at the turbine, farm and fleet level.
Through the constant collection of real time data—weather, component messages, service reports, performance of similar models in GE’s fleets—a predictive model is built and the data collected is turned into actionable insights.
The digital wind farm is used to manage GE Renewable’s Scottish Wind Farms
Weir’s products across all sectors are maintained using preventative and predictive maintenance methodologies utilising sensor and operational data captured in often remote and harsh operating environments
About two thousand sensors have been installed on the Queensferry Crossing, carefully positioned to monitor the global behaviour of the bridge and its environment in real time.
All data will be stored on the cloud to allow data analytics and the identification of trends in behaviour.
This will allow the operator to respond quickly to extreme events, to target inspections and to carry out pre-emptive interventions to avoid potential failures
The Data Lab Innovation Projects
This research project is seeking to utilise machine learning, cognitive computing and predictive analytics to detect patterns in the machine data which lead to failures and malfunctions.
The outcome sought will allow proactive action to be taken to reduce the breakdowns and increase reliability for customers.
Data from the assets is being used in combination with ERP data to drive more efficient and cost effective servicing routines such as condition based maintenance or ultimately a ‘just-in-time’ service regime.
Asset data will be used in combination with service and repair data and eternal sources, such as location and weather, to identify opportunities to improve component reliability and drive design enhancements for Aggreko’s product engineers
50% funded by The Data Lab
This collaborative innovative project between OPEX Group and Robert Gordon University investigated the possibility of utilising advanced machine learning methods to enhance a systematic Data Science approach for alerting oil and gas clients of potential process degradation and equipment failures
50% funded by The Data Lab