Condition Monitoring – The Health of Machinery & Equipment
Published on : Wednesday 05-10-2022
How technology is changing the art and science of machinery and equipment maintenance in the IIoT era.
Achieving higher OEE (Overall Equipment Effectiveness) is a major goal of the Fourth Industrial Revolution. Acquiring data from manufacturing machinery, and performing analytics to derive prescriptive statements is the new age tool. The all-important data in this case is machine condition data. Monitoring the condition of machines and other assets with the help of sensors installed strategically to collect data which is then analysed to detect anomalies, has been a common practice in recent years. From the old practice of reactive maintenance (maintained at the verge of failure), industry has moved forward to regular maintenance (at fixed intervals) to counting number of operating hours and so on. There have been forecasts from industry experts about the amazing savings in expenses and downtime by means of adopting new-age maintenance techniques, and equally amazing revenues for vendors of such tools and software. Statistics indicate that on an average, improving OEE by 10% boosts the bottom line by 22%; and driving up OEE by 50% increases profitability by 300%.
Condition monitoring is commonly understood to apply to rotating machinery. What other assets in the plant can benefit from condition monitoring?
“It is just a myth that Condition Monitoring is largely applicable to Rotating Machinery. In fact, condition monitoring has evolved as a science over a period of time and its application bucket, at present, includes almost all types of industrial assets that are critical to environment, safety and production,” says Shivnath Ram, Head – Asset Reliability and Asset Management, Jindal Steel and Power. As example he cites his own workplace, an integrated steel plant, where several other assets are benefitted from condition monitoring including Hydraulics and Lubricating Systems; Electrical systems – transformers, electrical panels, cable joints, distribution lines; Refractory – furnace walls, torpedo liners, ladles, stoves and ducts; Piping, structural and machine foundation base; Locomotives, EOT and mobile cranes, slag pot carriers and HEMM; Pressure vessels and storage tanks; Slow speed or quasi-static bearings – BOF converter and slew; Vibrating screens and hammer mills, and Belt conveyors.
Sijesh Manohar, Vice President, Product Development, SAP Labs LLC, Palo Alto, agrees it is not just the rotating machinery. “Condition monitoring can be applied to any kind of asset that is operational and requires maintenance. The sensors or telemetry data from the asset is analysed to predict the probability of downtime for any asset. Today advanced maintenance software can calculate the health of the asset based on its telemetry data to determine the most optimal maintenance schedule,” he says. According to him, this method as opposed to the normal periodic maintenance reduces the cost of maintenance significantly and avoids downtimes more deterministically. In addition to this use case, the term condition monitoring is also used to monitor the telemetry data from containers or vehicles that are used for carrying perishable items such as medicines and food. “An example of such telemetry data is temperature and humidity, which are often monitored in the transportation of perishable goods,” quotes Sijesh.
“Condition monitoring is the process of collecting and analysing sensor data from equipment to evaluate its health state during operation. Hence, this could be applied to a range of machinery, including rotors, compressors, pumps, motors, and any critical machinery that requires maintenance monitoring,” explains Prashant Rao, Head – Application Engineering Team, MathWorks India. “Typically, these machines will have moving components that need monitoring to reduce downtime and better component inventory availability. Lately, we are noticing a steep increase in interest in batteries and other components related to electrification, including electric motors and power converters. MathWorks is working with customers across most of these domains,” he adds.
“Though the origin of condition monitoring is difficult to identify, it is possibly an advancement to predictive maintenance, which is widely used to ascertain the health of machinery, vehicle parts, equipment and other valuable assets, says Titli Chatterjee, Senior Lead, SME – Smart Manufacturing Practice, ISG Group. “Additionally, it is also an essential aspect in rotating machines which are very critical to process plants. The ‘machine behaviour’ is often addressed by Condition Monitoring.”
What is an estimate of the size of the market? How big are the opportunities? How much of this market is accessible to Indian companies?
According to Gangadhar Krishnamoorthy, Associate Partner, KPMG in India, the market for condition based monitoring (CBM), predictive and prescriptive maintenance exists across all sectors, viz., manufacturing, utilities, mining, infrastructure, agriculture that have usage of assets/machines whose performance drives profitability of operations. “The aerospace sector can be viewed as a pioneer in adoption and has significantly benefited from CBM. Service and repair costs of each aero gas turbine range approximately from $1 to $5 million. CBM plays a key role in optimising service intervals without impacting safety. Almost all CBM and assessments are performed in-house by aero-engine OEMs; however, these companies rely on support from analytics service providers to provide them better insights. This market is accessible to Indian companies and its size is estimated to be approximately few million dollars per year,” he estimates.
“Experts estimate that the current market size of the Predictive Maintenance (PdM) Industry may be around $4 bn worldwide. However, with the advent of IoT, edge computing, big data, Machine to Machine communications, and business drivers such as reducing the TCO through asset optimisation, the need for operational efficiency, higher plant reliability, and widespread use-cases across industrial applications will catalyse the adoption of PdM,” says Dr Raunak Bhinge, Founder and CEO – Infinite Uptime. “Again, the experts believe we are looking at a potential of $15-18 bn by 2027 which may grow exponentially at a 25-30% CAGR.”
Sureshbabu Chigurupalli, Board Member-Director, Balasore Alloys Ltd believes the global machine condition monitoring market size reached US$ 2.4 bn in 2021. Looking forward, the market is predicted to reach US$ 3.6 bn by 2027, exhibiting a growth rate (CAGR) of 7% during 2022-2027. “Increasing digitisation, along with significant growth in the oil and gas, automotive, defense, aerospace, manufacturing, food and beverage and marine industries, is one of the key factors creating a positive outlook for the market. Additionally, various technological advancements, such as the utilisation of secure cloud computing platforms, wireless technologies and the integration with the Internet of Things (IoT), are acting as other major growth-inducing factors. Furthermore, increasing awareness regarding preventive maintenance among the masses is also driving the market growth,” he opines.
While the process of collecting and monitoring conditions is vital to the health and efficiency of our machines and companies, condition monitoring and prescriptive analytics is a business activity different from legacy sales and services activities. Do you think MSMEs and start-ups have an inherent advantage in getting market share?
“Condition monitoring and prescriptive analytics requires a technology and data-driven mindset. While large existing players will have to overcome the challenge of managing the transition and change of the overall business processes; MSMEs and start-ups will have great advantages owing to their lean structure, faster decision-making, higher adaptability, and innovation-led creative approach,” says Sunil Joshi, Founder & CEO, DGTRA. According to him, digital data driven approach integrated with cloud and mobile technologies, network enhancement like 5G will result in democratisation of the Services & Solutions business built around condition monitoring and prescriptive analysis. “Cost has been one of the barriers for several SMEs and mid-size firms for deploying an effective condition monitoring program. Start-ups have great potential to deliver lean, cost effective solutions, with abilities to provide instant gratification to its implementers,” he suggests.
“Predictive maintenance provides information about the single decision with no options or choice to consider, such as perform the maintenance or take conscious decision for the maintenance, whereas prescriptive maintenance offers a number of options and its possible outcomes from which the engineer can select the most suited suggestion,” says Jasbir Singh, Automation Expert, Consultant & Implementation Strategist. As such, he believes, the complete stopping of the production line can be avoided by running the machine at a lower speed, if upstream and downstream processes allow it with the sacrifice of productivity. Or, the plant may desire to run the machine at lower speed, in order to delay the planned downtime to coincide with the delivery of high inventory finished products which is delayed to ship out due to reasons beyond control.
Saikat Bhowal, Chief General Manager – Instrumentation, EIL maintains that with the growth and expansion of condition monitoring, the demand for a large number of IIoT enabled sensors, AI/ML applications and data analysis are increasing every day. The requirement is beyond in-house capacity and capability of any company, irrespective of its size and market share. Already some of the big players are outsourcing sensors and other components. “Collaborations and partnerships are anticipated for joint technology development. MSMEs and startups with their diverse fields of expertise in hardware and software can bridge the gap and have potential to gain significant market share,” he asserts.
More and more companies are now implementing automation and remote condition monitoring systems, etc. Many are implementing simulation techniques as well. Are there real studies done to establish proof of concept (PoC) using simulation and digital twin techniques? This needs an intensive collaboration between prospective buyers and vendors. Is it happening in India? With what success?
“Yes, real studies are being done to establish PoC using simulation and digital twin techniques. There are many players in the markets including MSMEs and start-ups who are constantly approaching the industries; show-casing their technical understandings; discussing feasible options and explaining cost-to-benefit ratios in terms of safe work; availability, quality and productivity. This has already started in Indian industries,” confirms Shivnath Ram. The prospective buyers and vendors are now focusing on symbiotic business relationships although these initiatives are in the budding stage when we talk of digitalisation of condition monitoring activities. “New technology has a cost and most industries are still reluctant as they cannot clearly visualise the modus operandi related to asset digitisation and data acquisition sensors; human-machine data interface; data analytics and action prompts,” he laments.
According to Prashant Rao, the pandemic made digital twins and simulation technologies vital technologies for engineers and researchers. Digital twin models are used to reduce time to market, introduce rapidly new models, increase safety, and reduce risks. Virtual commissioning enables early testing and verification of machine software by using a model of the machine. With simulation, the interaction between mechanics, machine software, and the manufactured product can be tested, optimised, and verified in different scenarios before prototypes are available. This approach lowers costs, ensures high product quality, and speeds up commissioning times. “We work with organisations like Foundation for Smart Manufacturing (FSM) to help build real-life prototypes that demonstrate some of these simulation and digital twin technologies. These engagements help prospective buyers to see the technology in action while fostering faster decision-making,” he elaborates.
“The sensor data collected from assets is traditionally used in two different ways. One way is to apply machine learning techniques such as prediction and anomaly detection algorithms to calculate health indicators and predict failures. The second method is to apply engineering simulation to the digital twin representation of the asset,” explains Sijesh Manohar. In this method, the data collected from physical sensors are used to extrapolate virtual sensor readings on other parts of the asset using digital simulation. This is useful in assets where it is highly expensive to place sensors on parts of large physical assets (e.g., large wind turbines) or practically impossible to place sensors (e.g., a deep sea drilling machine). “There are real-world examples where this simulation technique is being used – for example in Scandinavian countries certain roadway bridges, digital twin representation, and simulation techniques are used to calculate stress and strain on different parts of the bridge to avoid failures,” he elaborates.
For Titli Chatterjee, the Digital Twin concept has a crucial role at the factory/plant level, also in shaping the smart manufacturing industry. The emergent technology which has witnessed a surge in recent years mostly focused on predictive analysis for various industries and domains offers the ability to get an in-depth insight on the inner operations of any system. “Digital twin research and implementation is vigorously being taken up and has especially become more popular in certain domains like urban spaces and mobility, freight logistics, smart cities, engineering and automotive industry, and manufacturing amongst others. Comprehensive research on the applications has been published in various journals and available in databases based on application industries,” she adds.
The larger and more highly capital-intensive industry, larger is the number of assets and the need to maintain them. This is especially true in case of process industries, which are also looking for ways to improve efficiency and reduce costs. Is the appetite for such systems bigger at large plant operators like power plants and refineries?
The appetite is certainly large in asset-intensive industries, agrees Dr Raunak Bhinge. Underperforming critical assets like gas turbines, pumps, heat exchangers, motors, compressors, and many more can wreak havoc in process plants. “For example, early detection and prediction of a crack in a turbine rotor can save a power plant from an unexpected breakdown and save a few million in lost production,” he says. Monitoring a host of assets in the Oil & Gas industry for real-time equipment inspection using industrial IoT and edge analytics cuts maintenance costs and reduces failures. Equipment in oil drilling operates at variable speeds and generates vibrations. “This equipment is at the heart of the drilling operation and must be well-maintained. Diagnostics and fault notifications can alert the rig managers in advance before any actual occurrences and losses,” says Dr Bhinge.
“Large plants like the power plants and refineries are supplied and built by OEMs who supply the design, operational technology and the machine while building the plant. The CBM systems are part of the plant erection and commissioning packages offered by the OEM. But there exists opportunities for CBM based asset performance management to be tapped into at these power plants and refineries,” says Gangadhar Krishnamoorthy, who also draws attention to the fact that today many older plants and refineries are being retrofitted with sensors to track performance of critical assets and optimise service and repair intervals. Additionally, refineries have their own complexities, owing to oil characteristics, in building and maintaining digital twins. “For example, crude Pre-Heat train (PHT) is a classic example where all refineries are interested in getting the most out of the PHT so as to minimise fuel fired in the furnace, but a digital twin that adapts to changing crude basket is not easy to build/maintain which makes traditional ways ineffective. Thus opportunities for modern CBM exist at such refineries,” he explains.
“Condition monitoring offers businesses and heavy machinery industries the unique opportunity to assess the viability of vital components without needing to schedule routine maintenance checks. This means we can now monitor the condition of our machinery while it stays operating, ensuring that downtime and unnecessary maintenance fees are a thing of the past,” says Sureshbabu Chigurupalli. The market conditions are indeed challenging for refineries in the 21st century. Fluctuating crude prices, demand ebbs and flows and changing energy costs cut into profit margins. And compliance with tighter environmental and safety regulations can be both difficult and expensive. At the same time, the opportunity to increase profit by refining poorer grade crudes requires machines to run at higher temperatures, putting greater stress on machines. “With so much at stake, mechanical assets need to run at or even beyond original design condition or capacity, reliably and predictably. Condition monitoring solutions can help to improve the PQCDSME (Productivity, Quality, Cost, Delivery, Safety, Morale and Environment),” he adds.
Can the government play a role in accelerating deployment of such systems? Are there any schemes for promoting better maintenance using data technologies?
“Absolutely, the government can use policy instruments to organise interest and motivation to help accelerate the adoption levels,” says Sunil Joshi. “We already see initiatives from institutions like the Defence Research & Development Organisation (DRDO), which has recently initiated the ‘Development of Indigenous Development of Equipment, Sensors, And Software for Condition Monitoring and Machinery Diagnostics Equipment Within The Country’ program. Also, apex bodies like National Power Training Institutes run programs like ‘Training Program on Condition Monitoring and Preventive Maintenance of Hydro Turbine’. Adoption of a data-driven approach is in progress across sectors and there have been several visible initiatives that demonstrate the focus on data-driven predictive solutions.”
“The government can ensure fair policies through the following ways,” agrees Jasbir Singh, who has the following suggestions:
1. Making policies, which must protect the interests of product developers, not only of the product development but also give free access to the government organisations for implementation of it as proof of concept.
2. The government can ensure a collaborative team with public sector organisations to support proper implementation and issue a certificate once PoC is successful for the others to accept.
3. The government can support small developers to improve their performance till the time they become strong enough to compete with international products.
“The government can encourage the accelerating deployment of IIoT-based condition monitoring systems by promoting these initiatives under the schemes and rewards undertaken for digital transformation of industry. Some of the applications like monitoring of health of bridges on railways or other public bridges, Heritage Structure Health Monitoring, monitoring health of municipality sewage pipes by employing digital twin technology, need investment from government institutions and agencies,” concludes Saikat Bhowal.
(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)