Industrial Automation Magazine India

The Changing Face of Maintenance Services

Published on : Tuesday 01-06-2021

How technology is shaping the maintenance strategies companies follow in their facilities.

Maintenance in industrial context, also called plant maintenance, refers to the procedure adopted by enterprises to minimise breakdown, avoid shutdown and eliminate downtime, in the process strengthening plant reliability. All machinery and equipment used in production requires maintenance on account of wear and rear and ageing effect. Besides there are consumables like lubricants – oils and greases – that need periodic replenishment.

Traditionally, maintenance is classified in two broad categories – Corrective and Preventive – though there are sub categories in each. These are self-explanatory terms, where Corrective maintenance implies activity undertaken after a defect or failure occurs, and Preventive means efforts are made to avoid a situation leading to a defect of failure. Today, most modern plants follow a mix of maintenance strategies aided in no small measure by the available technologies.

Impact of technology

So how has technology impacted maintenance practices in recent years? “We are in the era of Industry 4.0. Artificial intelligence is already in maintenance in the garb of Predictive Analytics. In other words, machinery protection data along with process data are being used in predictive maintenance of critical machineries in oil and gas, power including refineries, petrochemicals and utilities. This really would have a major impact on the bottom line by avoiding downtime in major machineries like compressors, steam turbines and gas turbines. Proven results are already in public domain,” says Sateyandra Kumar Singh, Director, intelligent Asset Management, speaking on the strength of his three decades of work with major organisations like Saudi Aramco, Cairn Energy, Reliance and IPCL.

According to PV Sivaram, Evangelist for Industry 4.0, Digital Transformation and Industrial Automation, plant maintenance is a set of activities, which are aimed at keeping production activities uninterrupted. Technology has entered into production plants over the years. By this, we mean tech products entering production machinery and other assets. Production practices and maintenance practices are actually procedures (SOPs), and they are impacted rather less by having automation and other digital gadgets in the machines. There is of course some difference in operating a machine, which is equipped with a touchscreen rather than knobs and switches. Similarly in maintenance practice, steps to maintain the electronic components are different and in addition to the usual steps of maintaining motors, bearings and other components. “Every breakdown or near-miss event throws up a wealth of data. By acquiring and analysing this data, and collating with similar experiences at other installations, insights can be gathered. If this big data is fed to analytical engines and to AI, then the most valuable lessons can be learnt. These lessons are – Root Cause of failure, and CAPA – corrective and preventive actions,” says Sivaram.

“Yes, Industry 4.0 is the ongoing automation process of traditional manufacturing and industrial practices using modern smart technologies,” agrees Suresh Sankaran, CEO, Star Automations, Pondicherry. “Information technology (IT) with Operational technology (OT) is completely changing the industrial process. A few years ago, data collection was a manual process, which required trained persons to access the equipment. The process is gradually changed with latest technologies such as cloud computing, Big Data, virtual and augmented reality, advanced HVAC technology, machine learning and artificial intelligence, Internet-enabled sensors and the (Industrial) Internet-of-Things (IIoT) are being deployed to enhance the scalability, quality, effectiveness and cost-efficiency of processes in large industries like oil and gas, energy, manufacturing and aerospace,” he adds.

“Maintenance at many companies becomes purely reactive, which means they are fixing or replacing parts after they fail or start giving abnormal sound. The second most popular maintenance practice from the past is servicing the equipment on a routine schedule decided by OEMs whether service/replacement of parts is needed or unnecessary,” opines Jasbir Singh, Director, ECPR Technologies. According to him, another maintenance practice becoming popular is condition based maintenance (CBM), which is developed by considering the equipment’s current degradation trend and forecasting the probable date of failure. Engineers get time to arrange spares and resources to take a planned stoppage for maintenance. “Presently all the equipment/processes provide system and process alerts, audio visual alarms, and indicators for initiating attention for call of correction. When the alarm sounds, it’s sometimes late to prevent the stoppage. Alarms are popping up in the DCS system to alert the operator but these are often ignored as it causes disturbance to the operator. They change the upper or lower settings without consulting the engineers or keep the reset button pressed permanently to avoid the alarm sound. The number of alarm alerts (high/low)/settings are done as per process requirement during the engineering phase. Such configured alarms in the DCS system help operators during abnormality. But many times critical alarms go unnoticed due to high frequency of non-critical alarms and then major failure happens,” explains.

Focus on operations or assets?

One important point that is often not getting adequate attention is the asset condition when it comes to maintenance practices. Often companies concentrate on operations rather than assets in matters of maintenance. Is this the correct approach? “Operation is the outcome of availability of assets in best condition. Availability of assets could not be ensured until you have reliable assets to perform. To have optimum performance from assets, Maintenance Management must have best strategies in place. So focus on maintenance and reliability is must if you want an outcome from operation simply,” stresses Sateyandra Kumar Singh.

“Line operations are the purpose of the existence of the company. It is natural that they get the focus of attention. Utilities are more or less a necessary evil. They are needed so that operations can happen,” explains PV Sivaram, reasoning why the focus is on operations. “Therefore, it is not surprising that the proportion of maintenance budget allocated to utilities is rather less than allocation for the production assets. As we have said right at the beginning, the purpose of maintenance is to keep the production ongoing. So both – production facilities and utilities or ancillaries need to be kept going,” he elaborates.

Suresh Sankaran is of the while the operational goal is to help increase production, one must ensure environment safety and support key decisions around asset life and repowering. Some companies are not focused on preventive and predictive maintenance; instead they are replacing failure parts immediately from back up to manage day-to-day operations. “All assets require periodic maintenance to continue running efficiently. But the volume of failure products may accumulate into a huge asset value in the next few years. That invisible value of assets needs to be scrapped due to various reasons like spares not available at that time, corrosion due to the environment, technical sources unavailable, etc. The space for storing the idle products is also a hidden cost to a company,” he explains.

“The Operation Manager consistently focuses on plant/machine operation/line running, availability of raw materials, chemicals, staff and machine health. They are concerned with bottlenecks if any to meet the customer requirements in time. They pay attention to improve manufacturing cycle times, by reducing turnaround time, and avoiding reworks in process or maintenance. The bigger challenge is to continuously improve the process and consider how well the newer challenges in the process can be addressed within the framework of operation,” says Jasbir Singh.

Emerging technologies

How do emerging technologies like digital twins and AR/VR help remote troubleshooting? Digital twin, in conjunction with IIoT, helps build a predictive maintenance model, which can offer better insights into operations and help optimise maintenance cycles, in the process achieving a balance between a corrective and preventative maintenance approach. “Every technology has utility in every field. Since data is cheaper nowadays comparably, you can have real time observation of your running machines and it helps in analysing any problem as you do traditionally being in a control room or field. Troubleshooting is proactive action before any major failure happens. It could be done remotely if information is available with experts in real time,” opines Sateyandra Kumar Singh.

“The digital twin is a digital replica or virtual copy of physical assets or products used for physical model simulation. The digital twin concept impacts a wide range of domains, from engineering to architecture to aerospace to medical industries. It’s a technology outcome from physical drawing to computer aided design to simulation model based system engineering,” says Suresh Sankaran. “The speed and quality of maintenance and repair are critical in modern days. In some situations, some client or customer wants an immediate solution but you are unable to travel due to prevailing conditions. Applying technologies such as the digital twin, AR/VR model to the technical support domain will be truly transformative, with its ability to revolutionise the role of the on-site/field technician,” he adds.

“Remote maintenance is a much-used term today. The issue is that the specialist may need time to physically reach the machine to diagnose and repair. But he can surely observe the machine from his remote location – maybe his office,” says PV Sivaram, referring to the digital twin. “He can then guide a local technician with the procedures. Even better, if the knowhow of the expert in the office can be captured into an AI system, the bot can take over most of the function of this expert. The bot will interact with the technician via advanced HMI concepts like AR. And by the way, AR is not an ‘emerging concept’, it is well matured already,” he asserts.

According to Jasbir Singh, digital twins increase reliability of equipment and improve performance of production lines. It helps improve the overall equipment effectiveness or OEE by reduction of process downtime and improves overall performance which in turn increases productivity, reduces risk in all areas, and improves product availability and market reputation for the organisation. “Most of the companies have pilot plants or laboratories for digital twin function. It largely helps in collection of data and analyses. Augmented reality (AR), virtual reality (VR) and digital twins are business-enabling technologies to visualise insights about machine, equipment or operational technologies. The ability to view virtual environments and compare them with the physical are a distinct tool for training in industries like manufacturing and energy, where machinery and other physical equipment can be discussed in offline training,” he summarises.

Maintenance-as-a-Service

Another trend that is emerging today is Maintenance-as-a-Service, which is also in tune with the contemporary theory of core competency and leaving specialised tasks to domain experts. It also fits in with another emerging trend – leasing equipment or machines rather than owning them, on a subscription based model. But does this make a sound strategy?

“It is good when stakes are not so high in terms of impact of down time on cost, safety and environmental factors. Otherwise my personal opinion is Maintenance should not be left to other agencies who are not responsible for the goodwill of the company in case of a major accident. However some specialised services could be outsourced which are not very important in Maintenance activities,” feels Sateyandra Kumar Singh.

“Maintenance as a Service is an intriguing concept and should be investigated thoroughly. At the present time, however, much infrastructure and ecosystem is weak. We need to develop these entities and also beware of the issues of cybersecurity, confidentiality, etc. Financial models need to be worked out too,” cautions PV Sivaram.

Ideal maintenance practices

What should be the ideal maintenance practice for companies in process or discrete manufacturing industries? “The best maintenance strategy is the outcome after a detailed exercise of analysis and criticality assessment of type of industry, type of asset, organisational culture and a detailed exercise done at the time of construction and installation of a project. It is purely specific to location also,” asserts Sateyandra Kumar Singh. “However the trend is we have to use predictive maintenance more than preventive maintenance. Although you cannot stop failure totally of a machine, equipment, system or asset, yet you have to restrict breakdown maintenance to bare minimum,” he states.

“Maintenance is all about keeping the production running. This is not just about keeping items in good repair. There are other issues for example about components getting obsolete. So tracking the life-cycle of products is also a part of maintenance. With such programs, stocking of spares for legacy products can be planned. Even a renovation/retrofitting program can be scheduled,” says PV Sivaram. “It is not always only about parts becoming obsolete. Some older parts or machines stress the environment excessively. We can draw an analogy to filament lamps versus LED lights. The older generation has a poorer performance compared to the new generation and has benefits for resource consumption and environment. These deliberations form part of a modern maintenance toolkit,” he explains.

“The combination of preventive and predictive maintenance is one of the most cost-effective methods for ensuring reliability, safety, and energy efficiency,” emphasises Suresh Sankaran. “The main purpose of regular maintenance is to ensure that all equipment required for production is operating at 100% efficiency at all times. Ignoring periodic maintenance will result in orders not being fulfilled, customers not receiving their products, and breakdowns in your production targets,” he adds.

“The convergence of digital technology in industry is fuelling the development of a high level technology of predictive maintenance. Leveraging industry knowledge and technological development leads to providing information, when maintenance is to be done and what kind of maintenance is needed. Building the experience gained from asset management/diagnostic software, the computing algorithm behind the software is getting strengthened day by day. With the information thus available, the user can predict the life of critical devices and develop condition guidelines for future deterioration,” explains Jasbir Singh. According to him, systematic storage and graphical display representation of the measurement help the process to streamline quickly. The system abnormalities, transient failure of process and device gets quickly highlighted on the operator screen without user intervention even along with the help window for corrective action if needed by the operator. “These are data intelligence tools, which add value over data acquisition tools. These tools not only capture the data but manipulate, format and present in result form so that the user can take corrective action without the loss of time for analysis even in complex processes,” he elaborates.

Summing up, there is no simple solution when it comes to choosing a perfect strategy. The ideal maintenance solution can only be arrived at based on experience and suitability and could involve a mix of different maintenance strategies to suit the specific facility performance and operational conditions.

(Note: The responses of various experts featured in this story are their personal views and not necessarily of the companies or organisations they represent. The full interviews are hosted online at https://www.iedcommunications.com/interviews)