Industrial Automation Magazine India

Future production lines will rely heavily on AI for health monitoring

Published on : Friday 07-10-2022

Prashant Rao, Head – Application Engineering Team, MathWorks India.

Condition monitoring is commonly understood to apply to rotating machinery. What other assets in the plant can benefit from condition monitoring?
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. 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.
 
What is an estimate of the size of the market? How much of this market is accessible to Indian companies?
According to Industry Arc, a research agency, the Indian machine condition monitoring market is set to grow at a CAGR of 10.95% during the forecast period 2020-2025 and is projected to reach $193.3 million by 2025. With increased activity levels in the machinery and engineering tools sector, we are set to see more machines used in India. The ‘Make in India’ initiative and schemes like PLI (Production Linked Incentive) are encouraging more Indian players to get into these sectors. The breadth of this landscape covers Machine Tools, Textile Machinery, Construction and Earthmoving, Construction and Mining Machinery, and other heavy industrial machinery, such as Cement Machinery, Rubber Machinery, Metallurgical Machinery, Chemical and Fertilizer Machinery, Printing Machinery, Dairy Machinery, Material Handling Equipment, Oil Field Equipment, Paper Machinery, and others. Machine monitoring is a vital part of this industry and is primarily a mechanical process. Each of these industries is working towards reducing downtime which is why the market for condition monitoring is large and growing.
 
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?
Though a late starter in adopting digital technologies, the manufacturing industry has been now picking pace with Industry 4.0. Future production lines will rely heavily on AI for health monitoring and predictive maintenance services, visual inspection systems, and manufacturing process optimisation. AI techniques will lead the way towards the vision of a fully automated factory that flexibly manufactures goods in small batches – all the way to “sample size one” production. We will see more AI-enabled applications that are not only technologically interesting but also deliver significant economic returns.
 
The core competency of most manufacturing sector companies is not digital technologies; hence, they don’t possess in-house skills to plan and execute. They look for the right partners to help them in their transformation journeys. We see large and small organisations looking for help in several areas, including condition monitoring and predictive analytics. Technologies like Industrial Internet of Things (IIoT) that enable monitoring and analytical tools to enable predictive decision-making and help facilities with energy savings, labour savings, employee safety, and more. With computing on edge and the cloud becoming pervasive, MSMEs and start-ups will provide the rapid development and deployment of solutions.
 
The rapidly increasing calculation power of industrial controllers and edge computing devices and the use of cloud systems are paving the ground for a new dimension of software functionality on production systems. AI-based algorithms will dynamically optimise the throughput of the entire production line while minimising the consumption of energy and other resources. Predictive maintenance will evolve to consider data not only from one machine or site, but across multiple factories and equipment from different vendors. Depending on the requirements, the algorithms will be deployed on non-real-time platforms as well as on real-time systems like PLCs as Beckhoff recently demonstrated at Hannover Messe in Germany.
 
Are there real studies done to establish proof of concept using simulation and digital twin techniques? This needs an intensive collaboration between prospective buyers and vendors. Is it happening in India? With what success?
The pandemic made digital twins and simulation technologies now 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.
 
Is there an appetite for these types of systems more at large plant operators like power plants and refineries?
We are seeing strong interest in these technologies from a broad cross-section of industries, including large steel manufacturers, mining companies, and food processing companies. Small and large companies are using technologies that enable Industry 4.0, like IIoT, Deep Learning, Machine Learning, and data analytics to track, monitor, and take preventive actions when appropriate. Companies are going beyond condition monitoring to predictive maintenance. Predictive models use sensor data and other relevant information to detect anomalies, monitor the health of components, and estimate remaining useful life (RUL) of machinery. With predictive maintenance, engineers can schedule maintenance at just the right time – not too early or too late.
 
Can the government play a role in accelerating deployment of such systems? Are there any schemes for promoting better maintenance using data technologies?
We are seeing a couple of areas where start-ups and MSMEs need institutional and governmental help. To have access to data that could help test their algorithms and solutions with real historical data and to have some support for market access. These could be done through government-led centres or forums where customers and solution providers interact. The centre can have the required hardware or machines where customer data samples are made available and solution providers can deploy and test their solutions in these labs. These centres can be seen as capability demonstration areas to help smaller organisations demonstrate their capabilities and assuage fears from risk-averse larger organisations to deploy these technologies within their organisations. Finally, these centres can be used to develop competencies and the skills required within the MSMEs enabling them to develop their applications and solutions in the future.
 
The role of the government is to be the intermediary that facilitates trust amongst these various players in the ecosystem. This is in addition to the various incentives that the government is providing, specifically for start-ups and MSMEs.
 
Does engineering education prepare graduates/postgraduates to design, specify, evaluate, and implement such new age solutions?
We believe the engineering curriculum should constantly evolve. As the pace at which newer technologies are adopted, it is the responsibility of educational institutions to equip engineers with the right skills. We’re encouraged to see many institutions in India have been incorporating courses on data science and AI into their curriculum. MathWorks has been working with several organisations in multiple ways to accelerate this change. We do broadcast classes with students and focused faculty development sessions, in addition to working with faculty heads to modify the curriculums incorporating newer technologies. Our collaborations with specific Tier-1 institutes as well as with education consortiums like ICT Academy and IUCEE help us work in a focussed manner and enable us to reach large sections of faculty and students. We also extend support to institutes to develop courses incorporating project-based learning techniques that in turn help students apply the classroom learnings to real-life projects.
 
Prashant Rao heads the Application Engineering team at MathWorks India. Prashant is a regular contributor at industry forums, sharing his views around megatrends in technology and how Artificial Intelligence (AI) is getting adopted across industries and other technologies. He works closely with the academic community to help develop analytical and AI-related skills that make them industry-ready. He can be reached at [email protected]
 
(The views expressed in interviews are personal, not necessarily of the organisations represented)