Now there is a third scientific approach to managing equipment maintenance:
Predictive maintenance arose with the advent of ‘Digital Transformation’, the transformation of an organisation’s activities, processes and models resulting from the introduction of digital technologies.
In the world of equipment maintenance, ‘Digital Transformation’ came with equipment manufacturers introducing real-time sensor technology into their products. This allowed individual equipment to register failure itself or to continually register its latest condition and negate the need for individuals to perform periodic manual maintenance checks on equipment. This transformation also allowed equipment intensive organisations to deploy predictive analytics against their equipment and forecast when maintenance activities should be assigned before the point of failure occurs.
The ‘predictive’ approach to managing equipment maintenance offers significant benefits to organisations whose focus is to ensure they optimise the use of their equipment without risking its failure and the resulting cost of downtime. In many industrial sectors, the approach is revolutionising maintenance activity processes. We are seeing it being adopted by an increasing number of organisations- according to ABI Research, a leading market analyst, the maintenance analytics market is currently growing by 22% annually and is expected to reach a market size of $24.7 billion by 2019.
So, what are the key elements that an organisation should consider if they wish to adopt a ‘predictive’ approach to their maintenance activities? Well, for any organisation wishing to undertake the transformation to predictive maintenance, then there are three distinct ‘layers’ that need to be adopted:
The organisation needs to ensure that the equipment they maintain is equipped to capture data and provide real-time feedback on its current condition and status. Equipment data is normally captured using sensor or accentuator technologies. Sensors act as devices which detect their environment conditions e.g. a house thermostat used to monitor room temperatures. Actuators enable interaction with the physical world around them e.g. a heating system in a smart home which manipulates the temperature control valves to change the temperature.
/>Sensors and actuators form the core infrastructure of what has come to be known as the ‘Internet of Things’ (IoT). The IoT revolves around the ability for connected devices to communicate directly between themselves in real-time. We are seeing an increasing number of connected devices in our everyday lives performing this connectivity – Gartner estimates approximately 3.9 billion connected things were in use in 2014. This figure is expected to rise to 25 billion by 2020.
Increasing numbers and types of devices are being designed to offer connectivity. Many household washing machines offer connectivity to their manufacturers service departments as standard, whilst homeowners are also able to turn house lights on and off from their mobile devices through connected lighting.
In the case of more complex connected equipment, such as railway track sensors or traffic management systems, these connected equipment devices capture significant amounts of useful data, some of which needs to be processed immediately i.e. time sensitive data on threat detections or shutdowns, whereas other data can be processed and analysed on a longer-term basis to predict future problems.
The key consideration here is ensuring that all useful data can be passed back to the organisation and converted from its raw state into digital streams for processing. The conversion of this data is performed using Data Acquisition Systems (DAS) which sample signals from the physical world and convert these into digital values. Data can then be integrated with processes and applications within an organisation making use of Cloud technologies, specifically Platform as a Service (PaaS) ensuring organisations get maximum value from their asset data.
The final layer to the introduction of a ‘predictive’ maintenance approach is the application of data received from equipment and how this can be analysed to generate application specific activities such as maintenance service requests.
IoT analytics applications are used to help organisations understand the IoT data at their disposal. Although the IoT can provide a plethora of useful data to organisations, it is about turning this data into useful results which provides the key challenge. This is where the Big Data analytics tools become useful. Big data analytics tools are used to process large and varied data sets to uncover hidden patterns, correlations, trends and preferences allowing organisations to make more informed business decisions.
In a maintenance context a big data analytics tool can be invaluable in alerting an organisation of any customer issues before they experience the inconvenience of equipment failure. For example, big data maintenance analytics allow utility suppliers to use predictive analytics generated by smart meters to detect early warning signs of supply and demand issues on the grid prior to an outage occurring. Similarly, component manufacturers deploy data analytics to monitor the performance of manufacturing processes to detect and replace deteriorating components before they fail.
As you would expect, Oracle has been quick to recognise the benefit of offering technologies which allow organisations to maximise the benefits of predictive maintenance. Oracle’s Digital FieldService cloud offering, embedded with the Oracle Service Cloud and Oracle Field Service Cloud applications, provides a solution to all three ‘predictive’ maintenance layers.
Standard integration with IoT equipment allows organisations to collect data from device sensors connected to remote IoT equipment. Data is gathered from these equipment sensors through the Oracle Cloud which can then be analysed to predict breakdowns and forecast maintenance activities or to perform repairs or replacement activities.
Alerts on mobile service dashboards can also be received in real-time from equipment or from the results of predictive analytics to trigger either manual investigations, based on observation, or to investigate issues through remote diagnosis. The results of these investigations can then be fed back into the predictive data analytics to update any root-cause analysis.
Sensor data can also be combined with data sourced from social media chatter to identify potential issues before they become problems. This leverages a centralised knowledge base to determine the best maintenance solution. By aligning asset usage with service issues, organisations can also match engineer skillsets, tools and processes to provide an optimal service, matching schedules, spare parts and materials with predicted maintenance requests.
Further, real-time sensor data can also be used for tracking assets, ensuring an accurate record is kept of asset locations, associated service level agreements and their related service contracts.
Finally, sensor data obtained can be used to gain further insight into customer activity to cross-sell, up-sell product lines and generate targeted marketing campaigns based on data analysis.
The digital transformation brought about through the introduction of equipment sensors, Internet of Things, Big Data analytics and Service Cloud technologies have provided the tools for all organisations responsible for asset and equipment maintenance to focus on adopting a ‘predictive’ maintenance approach to managing their service activities. With its ability to drive down maintenance costs, improve customer service levels and optimise equipment lifecycles, the ‘predictive’ maintenance approach is quickly becoming the key driver in determining an organisation’s approach to servicing equipment.
If you would like further information about how your organisation can benefit from the ‘predictive’ maintenance digital transformation provided by Oracle’s Digital Field Service applications – Oracle Service Cloud and Oracle Field Service Cloud, then please contact Claremont.