According to a recent Deloitte survey, 93% of manufacturing companies believe that AI will be a crucial technology to drive growth and innovation. With the advent of AI, manufacturers can now leverage advanced algorithms and machine learning technologies to significantly improve their operations. AI-powered systems can help manufacturers monitor and analyze vast amounts of data in real time, enabling them to optimize their manufacturing processes.
However, there are a few challenges faced by manufacturers in the adoption of AI. Integrating AI technology with existing systems can be difficult. Moreover, as AI systems require vast amounts of data to operate effectively, the quality and availability of data can be a challenge.
Furthermore, connected devices and AI systems can pose a security risk for manufacturing companies.
Industry Outlook organized a webinar on AI adoption in manufacturing to understand how these challenges can be addressed to harness the potential of AI and tap into the new opportunities it offers. Through this webinar, we tried to throw light on ways to address such challenges to harness the potential of AI and tap into the new opportunities it offers.
On the panel, we had some esteemed guests from different industries who shared their views on the adoption of AI in manufacturing.
Excerpts: The webinar started with an address by Sreeji Gopinathan, Chief Information Officer (CIO), Lupin
Digital Transformation: Adoption of AI in Manufacturing
Today, we will be looking at how AI is making an impact on manufacturing outcomes, the key areas and use cases, and the challenges faced by the manufacturing industry. However, the question is, that in many instances in the conversations internally as well as externally, there are people who are very enthusiastic about the possibilities that artificial intelligence, machine learning, deep learning, and generative AI can bring to manufacturing companies, but there are also individuals who are not very convinced yet about whether this is going to have an impact or not. Therefore, it does bring in different organizational dynamics, when we try to do these kinds of initiatives in the organization. However, this is not new to any kind of new technology and technological advancements that are new to the market.
A few decades ago, when computers were invented certain organizations' employees found it challenging to adapt to using them. They were used to working on paper, storing data in files, and referring them and hence it took a lot of effort, and a lot of proof points to show people that the computers within the organization are making people's life easier.
Also, before the pandemic in a country like India, working remotely and collaborating remotely using technology was never accepted. However, overnight, that had to become the reality which proved effective. And, similarly, the AI journey is also going through a similar kind of learning process, an organizational change process, before it becomes a real way of life.
There is already an impact that can be seen and different organizations are witnessing it in numerous areas such as process automation. Although we used to do process automation in various forms and shapes in the past as well, today AI has brought in another dimension to it along with robotic process automation and enterprise BI platforms and digital performance management dashboards and when all these technology advancements put together on top of the AI-enabled process automation is giving great results.
Similarly, in the front end, when it comes to sales and marketing, the market trends prediction, and the sales forecasting prediction are already witnessing improvements. Quality Inspection is another area where some organizations are successfully leveraging AI-enabled technology. Also, product yield improvements are another area where you will witness that some companies have started seeing improvements due to the machine learning algorithms that have been deployed looking at all the different parameters that contribute to the outcome of a production process. It could be actual in-process parameters of the compound or the device or a car or any product that is being manufactured, however, all the other parameters like temperature, humidity, all those things are brought into that data science model that is built, which helps isolate parameters, which has more impact on the outcome compared to other parameters and that is how the yield impact is improved upon.
Therefore, some manufacturers today are claiming that they have realized up to a 50 percent reduction in production time. And this is not in a pharma sector scenario, but in more component-oriented and discrete-oriented manufacturing. However, when there is a continuous automated manufacturing process, present in sectors such as steel, automobiles, etc, that end-to-end approach is easier to put in place. And hence, the bottlenecks that get eliminated as they go through the journey will catapult the journey in reducing the overall production time and reducing the errors and the failures that occur, so that there is a more effective outcome that comes in that particular process.
If you look at the AI contribution to the manufacturing industry, and the impact that AI can have on the manufacturing industry to drive their outcomes can be as high as about $16 trillion, in a couple of years. That is the impact AI can bring to manufacturing industries, as per the latest estimates. Therefore, it is a huge opportunity that is staring at the manufacturing organizations in terms of how and what AI can do to get the outcomes that they want to drive.
Unleashing the Power of AI in Industrial Machinery Manufacturing
Shailesh Sharma, Director of Operations, SKF India and South East Asia spoke on how AI is revolutionizing the manufacturing industry.
In the bearing manufacturing industry, the main element is the operations. Here we have to be aware of what exactly we need, why we are doing it, and also know if we have sufficient competency available, and whether the basic infrastructure is available or not. To address this, a lot of preparation is required. As we have started taking baby steps in artificial intelligence, we have not completed any process in full-scale and most of it is at the POC level. Before going to artificial intelligence, it is important to build the infrastructure.
Artificial intelligence depends on the data. It depends on the source of data, and type of data, to check whether our machines are capable to generate those data or not. We have passed through this particular phase, and have upgraded our machines as well. Our control system has now become a single PLC. We have tried to bring all the uniformity as far as possible.
We faced a couple of challenges related to competencies and infrastructure and the machine condition. However, we have solved these issues today. Our main focus is to deploy advanced technology and AI in production, quality, supply chain, and maintenance. One of the main objectives of this is to improve efficiency or loss reduction, which also turns out to be productivity improvement. And, when we talk about quality, we focus on reducing our rejection, reducing every scrap, reduce our customer complaints. And in the supply chain, the main objective is forecasting, which is the biggest problem in today's world.
SKF also has a business in services where the company takes care of the performance of the rotating equipment and has developed its sensors which are mainly for vibration-checking systems. Bearings have essentially important components such as an inner ring, outer ring, ball or rolling element, and others. They work in general concepts that include the inner line and outer line which means, these are two different components. There is one main process which is the grinding process and a sequence of similar operations. Similarly, in the outer ring, the same process occurs.
One of the most critical quality parameters for bearing is the noise and noise in SKF terminology is calculated by checking vibration and we have developed a machine that checks for the vibration. We cannot predict what is coming, and how macro or microeconomic situations will disrupt situations, which creates a lot of inventory not only in our own business but also for tier one, tier two, and all the suppliers. Therefore, these are a few innovations that we are trying to incorporate into our processes.
Overall, the webinar provided attendees with insights to assess their preparedness for incorporating artificial intelligence in the manufacturing industry. It also guided them in charting a path to overcome the challenges specific to their respective sectors.
