In organisations that make intensive use of assets in their operation, the management and maintenance of these business assets is a key element in the development of their activity and directly affects the company’s profit and loss account. Specifically, maintenance tasks make it possible to ensure the availability of these assets and their efficiency in operation.
Historically, we have been working with two types of maintenance: corrective maintenance, which is carried out once an incident has occurred (failure or malfunctioning of the asset), and preventive maintenance, which is a planned way carries out certain tasks to try to avoid these failures in the equipment and the economic losses they entail. An appropriate combination of these two types of maintenance ensures maximum availability and efficiency of assets while optimising the cost of such maintenance.
There is a final type, predictive maintenance (PdM), which although not new and widely used, mainly in some energy sectors and industry, generally has less weight in the retail maintenance environment. Predictive maintenance tries to anticipate incidents not in a planned way or based on the specifications of the equipment, but through the monitoring of the assets, using more or less complex algorithms that allow identifying the best moment to carry out maintenance depending on the state of the asset.
The prevalence of preventive maintenance over predictive maintenance has generally been due to the difficulty of obtaining correct and real-time information on the state of the assets, as well as the availability of analysis tools that allow the future to be predicted from this data with the necessary minimum reliability.
The introduction of predictive maintenance cannot be done in a disruptive way, but organizations must move forward by taking firm and safe steps from pilot experiences while maintaining the foundation of preventive maintenance to make a safe transition.
Digital technologies for predictive maintenance
Many of the technologies we can use for predictive maintenance are not necessarily new. For example, the use of sensors in the plant or at the point of sale is quite common today, but in recent years they have indeed evolved to be more intelligent, reliable and, very importantly, economical.
Similarly, computers, storage capacity and communications have become much more powerful and affordable. This reduction in costs is important in sectors such as retail, where an organisation may have thousands of computers to monitor.
In recent years, technological advances and the emergence and development of certain disciplines often linked to Industry 4.0, such as the Internet of Things (IoT) or information analysis tools, including Big Data techniques, have enabled a breakthrough in the development of predictive maintenance and make it a tool within the reach of most companies. Which of these technologies are the best candidates to support predictive maintenance?
Internet of Things (IoT)
Given that predictive maintenance is mainly based on obtaining information directly from the equipment, it is clear that IoT is possibly the technology that can have the greatest impact on its development. IoT allows your devices to send a constant flow of information to the company’s private servers.
This information comes from sensors that can capture a wide range of stimuli, such as temperature, pressure or even visuals. It can also come from other sources, such as programmable devices or enterprise information systems. Through IoT, we can have a digital representation of what is happening in the assets so that this information can be analysed and processed to trigger maintenance actions when necessary. The level of representation can vary, even incorporating virtual representations of the asset, known as Digital Twins.
Data storage and processing infrastructures
Predictive maintenance generally requires the analysis of a very large amount of data coming from different sources with a high frequency (even in real-time). It is therefore important to have powerful data storage infrastructures that allow agile access to the data to analyse and visualise the information using advanced analytical techniques and predictive algorithms, such as Business Intelligence (BI) tools or even Big Data.
There is also the possibility of transferring the analysis of the data to the sources themselves (Edge Computing), taking advantage of the power, cost and availability of computing capacity that can be transferred to the very equipment where the information is generated.
The analyses and information generated by the predictive maintenance tools must be presented in an appropriate way to the different users of the system, from maintenance managers to operators. These tools can range from simple dashboards to augmented reality solutions. It is also important that all this information can be displayed on different devices, both desktop and mobile.
In addition to visualisation, it is important that the systems can generate alarms and trigger events that allow integration into maintenance workflows, to automate tasks and increase the efficiency of asset management.
Predictive maintenance at Retain
Retain is a Strategic Asset Management platform that has a powerful CMMS maintenance manager, capable of integrating with all types of information sources to capture information on the different assets and develop a predictive maintenance strategy through the use of advanced algorithms. The result of these analyses can be visualised through the tool itself and integrated into maintenance processes for maximum efficiency.
Do you want to know more? Contact us and a specialized consultant will solve all your doubts without obligation.