Much of the manufacturing environment is going through a period of significant disruption — and enterprise asset management is no exception.
Unlike asset-intensive industries, in industries such as automotive, machinery, food and beverage, and pharma there has been a greater focus on reactive and autonomous maintenance for some time. With the emergence of Industry 4.0, maintenance has stepped out of the shadows and many companies have started to assess the newest technologies to help them go beyond a lean and traditional “run to failure and preventive maintenance” approach.
Looking at the digital transformation strategies of leading enterprises, the digitization of asset management has become a priority for many of them. The rapid development of the latest technologies and trends such as edge devices, SaaS-based tools and apps, cloud-based platforms, AI/ML-based models, and digital twins of digital twins has put maintenance engineers, managers, and field workers under even greater pressure, as C-level management seeks to leverage the technology to deliver significantly better operational results. This has resulted in the need for continuous education at all levels in the plant and tighter communication and collaboration with IT vendors and system integrators. Enterprises have also learned from experience that when they relied on a single supplier of a solution the outcome was poor as the solution was so complex that a single vendor was unable to deliver the expected value. In the latest IDC Global IoT Decision Maker Survey 2019 only around 4% of respondents said they had started an IoT project and then failed — highlighting the increasing level of project preparation, including identifying the right project and the right vendor.
Legacy systems are a common challenge for many companies, both on the IT and OT sides. Retrofitting and additional instrumentation could be seen as an easy way out, but in real life this brings other challenges such as placing the sensors in a difficult environment, closed control systems, or unsupported connectivity between the machine (level 1) and the data layer (level 2). This gap would be probably bridged first, when the enterprise invests in new production lines instrumented with the latest generations of controllers that also cover the asset condition monitoring area.
Production Together With Maintenance
Production and maintenance often see each other as process silos, especially when it comes to the utilization of the production data and use of supporting IT systems. Even if modern CMMS systems enable the connection of various systems such as ERP, IIoT, and spare part marketplaces, there are still many enterprises utilizing the CMMS more as a work order management and maintenance planning and scheduling and budgeting tool, ignoring the value the system can bring when properly integrated and used.
On the other hand, the modern way of operations management is based on the ability to oversee the production process holistically through integration of the operational data including asset condition monitoring and energy consumption monitoring. A comprehensive control and visualization solution provides real-time information across production units to the depth of aggregate level. The recommendation management function then distributes information to the right person in real time, providing supportive information to the operators or maintenance people. This approach destroys the traditional siloed way of work through the integration of technology, processes, and organization.
What Really Helps: Predictive Maintenance vs Condition Monitoring
Predictive maintenance is one of today’s most talked about technologies when speaking about Industry 4.0 and enterprise digitization. But many enterprises confuse predictive maintenance with advanced condition monitoring. Again, condition monitoring’s function is to observe the defined parameters and compare them against defined thresholds (upper and lower limits). Visualization of final outputs is mostly in the form of traffic lights indicating the condition of the machine or aggregated in real time.
Unlike condition monitoring, predictive maintenance is based on the utilization of machine-learning engines, where the current parameters are the basis for the prediction of what will happen in certain periods with certain probability. The cornerstone of success is the prediction accuracy and connection of the predictive model outputs with the enterprise maintenance processes. Deploying predictive maintenance on a fleet of machines is challenging from a cost and technology perspective. In many cases, most companies implement only condition monitoring while believing it is predictive maintenance.
Maintenance Worker as a Data Platform
Modern maintenance field workers are equipped with a set of tools that effectively transform them into a fully connected real-time data platform. The data could be acquired through intelligent hardware integrated into the worker’s helmet connected via WiFi, Bluetooth, or the mobile network, utilizing cameras, digital microphones, and sunlight-readable displays for high-resolution photo capture and real-time video chat. The latest technology also provides active remote assistance for workers, geolocalization, navigation, asset visualization, and online library access, among others.
The trend is to provide field workers with full mobility and 24 x 7 remote support. To this end, mobile devices are enhanced with a mobile app for the instrumentation, control systems, and safety systems to view tags, trends, and alarms, and provide access for field staff to all aspects, including workers’ travel to and from production assets located in the field or factory.
Key players in EAM/CMMS are not standing by, and are instead enhancing their applications with remote field support functionalities, providing full data visibility across enterprise systems such as ERP, MES, and PLM, and utilizing data from SCADA systems.
Enterprise asset management is going through significant change, and this is having an impact on technology, process, and people. Integrated enterprise systems enable transparency across the production and field network and supply chain, providing huge amounts of data in real time. Though maintenance engineers and field workers are becoming more flexible and effective by having 24 x 7 access to data and information, we should keep in mind that what they really need are clear instructions, guidance, and insights. This really changes roles, from being reactive and almost blind to being able to investigate the immediate future, predict what will happen, and react accordingly.
But it is not just about maintenance. Data engineers and data analysts should also understand the technology and processes that will enable them to turn information into insights, as this is a key success factor when increasing availability and reliability of production assets. Collaboration with data scientists and enterprise maintenance is the real revolution that is driven by rapid technology development.