Asset management fundamentally revolves around getting the maximum output of assets while maintaining minimal costs associated with repairs or downtimes. It can be a complex procedure as there are many factors involved such as scheduling, planning, and more. The purpose of digitisation in asset care and maintenance systems is to automate these processes, allocate resources effectively and finally to drive profitability and reliability.

The digital transformation across industries marks a pivotal shift toward optimising operational efficiency and enhancing asset management practices. At the forefront of this shift is the evolution from traditional, reactive maintenance strategies to proactive and predictive maintenance models, facilitated by the integration of digital technologies. Integration of digital technologies in asset maintenance such as leveraging data analytics, machine learning, and artificial intelligence for asset care and integrity can not only predict but prevent downtime, optimise performance, minimise waste and ensure safety and quality in production environments and other various sectors, setting new benchmarks for reliability in the competitive global marketplace.

‍

Asset Care and Digitisation

The primary goals of asset care in any industry are maximising asset availability, ensuring safety and quality, delivering optimum performance, and minimising wastage. This is not merely about keeping machines operational but ensuring they are ready to deliver whenever required, thus avoiding costly downtime and ensuring a seamless production flow. Ensuring safety and quality goes hand-in-hand with availability, as it safeguards not only the workforce but also upholds the standards of the final product, which in turn, protects the brand’s reputation and customer trust. Central to these objectives is the asset’s life. Extending the usable life of machinery and equipment through meticulous care and regular maintenance or other methods is imperative because not only does it ensure the standard quality in its output but also helps in reducing over all costs associated in asset procurement. It keeps production lines running smoothly but also preserves the integrity and longevity of the assets themselves. Proactive strategies to extend the usable life of machinery and equipment should be done on a regular basis and digitisation comes into play here.

The key components to digitised asset care are data storage, data to useful information, data analysis and actions, predict downtimes via machine learning, autonomous action with AI and safe and self maintained assets. The first step in the digitisation process involves collecting large amounts of data from various assets through sensors and IoT devices. This data collected can include operational metrics, performance indicators, maintenance records, and much more. With the collected data, the next step is transforming this information into useful information and this can be done by using advanced algorithms, machine learning models, or analytics tools. There are various tools in the market available to help achieve this but the key is to identify patterns and predict potential issues like equipment failure. The transformation of data into useful information helps for decision-making and strategic planning for the long term future. Software platforms can help in data analysis and actions while machine learning can help predict downtimes or trigger automated maintenance procedures or alert various teams to take preemptive action. Combining these with autonomous actions facilitated by artificial intelligence, where assets are not only monitored but also maintained autonomously, ensures safe and self-maintaining assets. The ultimate goal of digitisation in asset care, is to get to a stage where assets are not only monitored and maintained with minimal human intervention but also is inherently safer and more reliable. Not to mention this also frees up work allocated to employees and makes use of their potential more efficiently in other areas of one’s business. This strategic planning and utilisation of digitising tools helps businesses to scale up exponentially.

When it comes to data collection, it is not just to gather just any data but to capture the right kind at the right time. One such metric is the Overall Equipment Effectiveness (OEE), which indicates an asset's overall productivity. By tracking OEE, businesses gain insights into how effectively an equipment is being used. Another would be Repair and Maintenance (R&M) expenditure data. Keeping a detailed log of R&M expenses helps to pinpoint where investments in maintenance are yielding results and where they are not, which enables decision-makers on where to focus their maintenance strategies and budgets. Other data types can be thermal data where one can assess excessive heat that can be a sign of over-exertion or friction within mechanical systems, indicating potential wear and tear that could lead to malfunctions. Similarly, vibration data is different from every asset and deviations from baselines can be early warnings of mechanical issues. Capturing and analysing vibration data allows for predictive maintenance. Finally, power consumption data offers insights into the efficiency of asset operation. Fluctuations in power usage can indicate problems or inefficiencies within systems, making this data critical for energy management and cost-saving measures.

The transition from data collection to data analysis signifies a move from raw data to usable, actionable information. Analytical techniques can include stratification, regression analysis, image analysis, vibration analysis, and the application of machine learning for model building and creating digital twins. Data analysis with advanced algorithms,machine learning models and AI, help parse through the collected data, identifying patterns that may indicate potential issues or areas for improvement. Digital twin, for instance, is a virtual replica of a physical asset and allows for simulation and testing without the risk to the actual asset. This culminates in a detailed understanding of asset health, allowing stakeholders to take proactive measures in maintenance and asset utilisation.

The final step in the digitization of asset care is data prescription. This stage is all about "what to do next" with the data. Data leads to alerts, alarms, recommendations for action, solutions, and the automatisation of processes. Combining automation and IT solutions, especially those that provide an intuitive interface through which anyone can design and implement automated solutions without needing in-depth technical knowledge, allows for rapid response to insights provided by data analysis.

‍

Evolution of Maintenance Systems

The progression of maintenance methodologies from reactive strategies to more advanced, predictive, and proactive approaches can be broken down into five distinct stages:

First Generation: Breakdown or Reactive Maintenance

The earliest approach to asset maintenance was purely reactive, also known as breakdown maintenance. This strategy operates on the principle of "fix it when it breaks," leading to maintenance actions only after a failure has occurred. While simple, this approach often results in unexpected downtimes, emergency repairs, and higher costs due to the unplanned nature of breakdowns. In this generation, machinery was often over-designed to tolerate this approach, with redundancy built into the system to cope with the inevitable downtime, which was considered just a fact of life.

Second Generation: Preventive Maintenance

In this approach, maintenance is scheduled at regular intervals to prevent failures before they happen. This could be time-based (scheduled after a certain number of operational hours) or usage-based (after the machinery has performed a certain number of cycles). This strategy reduced the frequency of unexpected failures and prolonged the life span of assets. Tools such as checklists and routine inspections became the norm, and the emphasis shifted towards maintaining assets regularly instead of waiting for a breakdown.

Third Generation: Predictive Maintenance

This represents a more sophisticated stage in the evolution of maintenance systems. It relies on condition monitoring and the use of sensors, IoT, and predictive analytics to assess the condition of equipment during operation. Techniques like Vibration Analysis, Wear Particle Analysis, and Thermography are employed to detect anomalies that could lead to future failures. This approach requires specialised skills but minimises downtime by identifying potential issues well before they would result in a breakdown. Maintenance interventions can be more precisely timed to when they are actually needed, based on the actual condition of the equipment.

Fourth Generation: Proactive or Reliability-Centred Maintenance

There is a focus shift from fixing failures to preventing them by using data and insights gathered from the earlier predictive maintenance approach. This stage is characterised by a systematic process that identifies and addresses the root causes of equipment failures through root cause analysis (RCA) and continuous improvement processes. The objective is to improve the inherent reliability and performance of assets. It involves applying reliability engineering principles to optimise maintenance strategies and tailor them more closely to the unique demands of each asset, with support from advanced tools, cloud storage for data management, etc.

Fifth Generation: Intelligent Maintenance

The fifth and most advanced stage is intelligent maintenance, where the maintenance environment becomes interactive and integrated with real-time training and troubleshooting capabilities. This stage uses digital twins, creating virtual replicas of physical assets, allowing for simulations and optimizations in a virtual environment. Advanced techniques like image analysis, pattern recognition, neural networks, and automated response systems come into play. This generation paves the way for smart factories and central control towers that provide comprehensive oversight and allow for more nuanced control over maintenance operations. Technologies such as augmented reality (AR), virtual reality (VR), and mixed reality (MR) are also harnessed to enhance the capabilities of maintenance teams.

Each step in the evolution of maintenance systems reflects a growing sophistication and integration of technology, moving from passive and reactive tactics to a dynamic, data-driven, and automated approach. This evolution has been driven by the increasing complexity of assets, the higher costs associated with downtimes, and the advancement of digital technologies. Progression through these generations, the maintenance function becomes increasingly strategic.

‍

The Journey from Lagging to Leading Practices

Reactive Maintenance: A Lagging Practice

This approach represents the most basic level of maintenance, where visual inspections are the primary method for detecting issues. Here, maintenance management allows assets to run until failure and interventions are only made after a problem has been identified, which can lead to increased downtime and higher costs due to the lack of preventative measures.

Preventive Maintenance: Create Scheduled Interventions

There is a progression towards more planned maintenance activities, also known as preventive maintenance. This approach involves performing maintenance at pre-scheduled intervals to prevent unplanned breakdowns. While it's a step forward, it's still reactive in nature, as maintenance actions are triggered at regular time intervals rather than the actual condition of the equipment.

Conditional Monitoring: Maintain Proactive Measures

This phase introduces conditional monitoring capabilities, where sensors are employed to monitor equipment issues in real-time and send alerts when anomalies are detected. This allows for maintenance to be performed based on the actual condition of the asset, rather than on a predetermined schedule.

Predictive Maintenance: A Leading Practice

This phase utilises a cloud or data warehouse to consolidate data from various sources such as sensors, Programmable Logic Controllers (PLCs), and Enterprise Resource Planning (ERP) systems. The centralization of data enables the use of predictive algorithms for better accuracy in maintenance activities, leading to an automated workflow that proactively addresses maintenance needs before they become critical.

‍

Approach to Digital Transformation in Asset Care in Business

Digitization in every business is important as it creates digital versions of previously analog systems and digitalization is the transformation within a company or its model that makes use of digital technologies and data to improve processes. However, digital transformation encompasses both these aspects and is a more holistic approach to engaging all stakeholders involved in decision making.

The digitization process begins with the conversion of analog information and processes into digital formats. This fundamental step transforms physical data, such as maintenance logs, machine readings, and inspection results, into a digital form that can be easily accessed, shared, and analysed. 

Moving beyond digitization, digitalization involves the use of digital technologies to enhance and transform business processes. This includes implementing process automation to streamline workflows, integrating sensors to capture real-time data directly from equipment, enhancing communication systems for better coordination among teams, and potentially developing new business models that utilise digital capabilities to deliver value to customers. This creates new opportunities for growth and innovation.

Next would be to embrace Digi-Fi, i.e. Industry 4.0 and the IIoT. This involves the sophisticated integration of cyber-physical systems, cloud computing, and cybersecurity to create interconnected networks of machines and devices. Operational Technology (OT) and Information Technology (IT) integration is a critical aspect of this phase, enabling seamless data flow and analytics across the enterprise. Digi-Fi signifies a transformative stage where businesses not just use digital tools but fundamentally rethink how they can operate within an ecosystem of connected devices and intelligent systems.

After this a business should be Digi-BI, where one derives actionable business intelligence insights through vast amounts of data collected and processed. Businesses should make use of  artificial intelligence (AI), machine learning (ML), and business analytics to extract meaningful insights from data, where normally a person may tend to miss on certain trends in the data. These insights drive informed decision-making, improve performance, and can lead to predictive maintenance strategies that anticipate and preempt potential equipment failures.

Finally, Digi-Cult is all about cultivating a digital culture within an organisation. It involves nurturing digital literacy among all employees, conducting capability assessments to understand and close skill gaps, and providing widespread access to information and analytics dashboards. The transformation here is cultural, as much as it is technological, recognizing that the success of digital initiatives depends heavily on the people driving them. A digital culture is about encouraging continuous learning, innovation, and adaptability, ensuring that an organisation as a whole is aligned and equipped to thrive in a digital future.

‍

Case Study: Predictive Maintenance & CBM in Line Efficiency Improvement

A prime example of digital transformation within asset care focusing on Predictive Maintenance at a leading food processing plant. showcases how integrating cutting-edge technologies can not only streamline maintenance processes but also significantly boost operational efficiency.

Vibration Analysis is used to capture the vibration spectrum across three axes in real-time, providing a detailed understanding of a machine’s health. This method enables maintenance teams to pinpoint anomalies and address them before they escalate into costly failures.

Supplementing this, Acoustic Analysis is used to listen for changes within the ultrasonic range, to detect changes in machine friction and stress waves, which can indicate early signs of deterioration.

The plant's commitment to rigorous maintenance extends to Oil Analysis. There is a regular assessment of the lubricant's integrity, its condition and quality, to check its suitability for continued use. Wear particle analysis within the oil further helps determine the condition of internal machine components, providing insights into the overall mechanical health of the asset.

Additionally, Infrared Monitoring and Thermography is used to spot both mechanical and electrical anomalies. IR thermometers and thermal cameras can detect hot spots and temperature variances, often indicating potential problems such as overheating or electrical issues.

The food processing plant also incorporates the application of Condition-Based Monitoring (CBM) for line efficiency improvements. Through sophisticated Split Type Current Sensors, the plant monitors thousands of data points per second for each motor. This high-resolution monitoring allows for an unprecedented level of precision in fault detection, transforming the plant’s ability to tackle issues before they result in downtime.

The plant finally utilises various software platforms and diagnostic tools to turn these data collected into graphs and dashboards that provide visual representations. It also maintains Real-time alerts, to messaging platforms like WhatsApp, where alerts can be received instantaneously by technicians, no matter where they are. There are many tools available these days for real-time monitoring. No-code platforms are one such tool that can be heavily utilised to not only conduct maintenance or inspection checks but automate the business processes and get real-time alerts, dashboards and automated reports, and this does not require any technical expertise.

‍

Case Study: Mixed Reality for Improved Availability

The same food processing plant used in the case study above also uses the power of AI-driven mixed reality to guide operators through maintenance procedures interactively. It uses voice-activated AI that acts as a virtual assistant to the maintenance staff. As operators engage with the machinery, the AI provides step-by-step instructions through a headset, ensuring that each action is executed correctly and efficiently. This continuous guidance helps in mitigating human error, as the AI systematically confirms the completion of each task before proceeding to the next. The AI system not only directs but also adapts to the technician's pace, providing a personalised experience that aligns with each individual's skill level and familiarity with the task at hand.

By integrating voice commands and visual cues, the system ensures comprehensive coverage of maintenance protocols. Each procedure is meticulously verified by the AI, leaving no room for oversight and elevating the standards of preventive maintenance.

By embracing this cutting-edge technology, the company not only bolsters the efficiency of its asset care but also sets a precedent in adopting innovative solutions for workforce empowerment and skill enhancement.

‍

The Way Ahead

As industries worldwide continue to evolve, the trajectory for asset care and maintenance systems is clear; the path leads toward the full integration of smart factory principles. By leveraging the prowess of generative AI, machine learning, and robotics, businesses are set to redefine asset management. The journey ahead is marked by the continuous development of digital capabilities within teams and the strategic implementation of data-driven use-cases. As organisations adopt these innovative practices, the importance of continuous learning and staying ahead of the technological curve cannot be overstated. The commitment to digitization and Industry 4.0 not only prepares us for the demands of tomorrow but also ensures that we remain at the forefront of industrial excellence.

‍

This article was inspired by the enlightening presentation on ‘Asset Care & Maintenance Systems Through Digitization’

Held at the 7th Annual Asset Integrity and Maintenance Summit in India

Given by Mr. Vijay Krishna - General Manager and ICML Head, ITC Ltd. (Medak Unit), Foods Division, whose expertise and insights have been invaluable in shaping the content of this piece.