AI’s Growing Role In Industrial Manufacturing

Topic: AI’s growing role in industrial manufacturing

AI’s growing role in industrial manufacturing

The industrial manufacturing sector is beginning to embrace AI and ML, but there are still many barriers to entry.

Artificial intelligence has an important role to play in driving the digital transformation of industrial manufacturing and delivering the benefits of Industry 4.0. Two key applications of AI in manufacturing include predictive maintenance and machine vision for defect inspection and product quality. AI also offers the potential to deliver benefits throughout the product lifecycle, ranging from generative design to product development to post-sale and warranty support.

However, there are a number of challenges preventing AI from making inroads into the industrial sector. Some of these obstacles are simply due to the makeup of the industry with a variety of use cases, company sizes, and lack of experience, adding to the complexity of implementation.

There is a lot of interest in AI and machine learning (ML), but there is a lot of fragmentation in the industrial manufacturing market, said Ryan Martin, director of industrial and manufacturing research at ABI Research.

It’s hard to talk in general terms about those technologies given the diversity of manufacturers and the problems they face, Martin said. There is a large base of small and medium-sized businesses and machine shops, especially in the US, that employs one to 10 or one to 50 employees who are really critical and could benefit from these technologies but don’t necessarily employ or have great access to them.

Martin said that only a few large companies, such as John Deere, Caterpillar, GM, Tesla, Apple, and Samsung, are driving much of the innovation.

These types of larger companies, typically employ AI on both the information technology and operational technology sides of their businesses: the former for data management and the latter for monitoring and control of industrial equipment.

The two main areas are related to data, which could be data normalization, data cleaning, or data analysis, thus extracting insights from data and making that data accessible, Martin said. The other area is quality, which can be quality in products or machines, including predictive maintenance type applications, he added.

Computer vision is another area where AI is being more widely deployed for applications like anomaly detection, Martin said. What is changing there is the difference between AI and deep learning applications that look for anomalies that are not predetermined. AI’s Growing Role In Industrial Manufacturing.

That’s really important because historically, these systems are good at finding problems if they’re known problems, but one of the biggest challenges is finding and fixing problems you don’t know you’re looking for, and that’s where AI, in general, comes into play, he added. . Ideally, you could link that anomaly to reason and then initiate action from it. Perhaps it is identifying that the defect occurred due to a problem with the machine producing it, or it could be that the defect occurred due to a vendor issue and that action could be taken automatically.

Today, it is more likely that one or more people will participate in those processes to identify the problem and then implement its corrective action, he added.

AI is also employed in the design phase, with AI built into the design software, Martin said. With generative design, designers input the key parameters of their products, and software using AI ideally generates a number of designs, he added.

The designer then narrows the design list of possibilities based on their criteria.

An example of a metric might be sustainability, in which product selection is based on the least amount of materials that are sourced locally, Martin said. This saves the designer an enormous amount of time and effort.

Challenges

However, industrial manufacturers still face challenges in implementing AI models. A big part of this is a knowledge barrier to entry.

Getting an AI model up and running doesn’t happen overnight, Martin said. You need to understand all of your inputs and, more importantly, what it is you’re trying to accomplish. There is often a lot of setups required, and even where companies can get up and running in 24-48 hours or even a week, which may be true, you still need highly trained people.

Martin said it’s highly unlikely that an AI solution can be set up overnight because you need to collect and analyze data and develop algorithms. Even if the common scenario these days is that a provider can come with algorithms that can help you get 80% of the way there, then you have to customize the last mile or the last 20%, which is a great approach, but it still requires that last-mile customization, which requires time or partners.

But changes are taking place in the industry thanks to new software delivery options. Software as a service (SaaS) and cloud services make newer technologies more accessible.

Cloud as architecture and SaaS as a delivery mechanism means there are much lower barriers to entry and manufacturers can get up and running in a very short time and ideally without as much training because the entire infrastructure would be supported and enabled by another partner or by the vendor. Martin said.

Examples include Siemens’ recently launched Xcelerator as a Service portfolio, PTC’s evolving portfolio on its Atlas SaaS platform, and AutoDesk’s Fusion 360.

Component manufacturers are also focused on lowering barriers to entry. An example is Sensata Technologies’ Sensata IQ platform, which makes it easy to implement asset condition monitoring to prevent unplanned downtime in manufacturing environments. This cloud-based platform uses AI to process data from Sensata sensors as well as qualified third-party sensors to monitor assets from anywhere, including a PC, smartphone, or tablet.

Sensata’s solution targets 85% of a plant’s assets that are not currently monitored. Most of what’s being monitored today are very critical assets in a plant, and many of those solutions are very expensive, don’t use the cloud, and are integrated into control systems, said Bryan Siafakas, director of the product line at Sensata. . Portfolio of industrial sensors and IIoT. Our focus is on that balance of plant assets where we don’t have to tie into existing control infrastructure. It is easily adaptable to any asset you wish to monitor on the plant floor.

The learning period to build an asset baseline could take a week or two. That information is then stored and trended in Sensata IQ, leveraging AI/ML, which is used to monitor certain types of faults. Sensata sets its fault characterization accuracy at 95% based on benchmarks.

Some of the anomaly detection (via sensors) is done at the edge, and when necessary, the data is sent to the cloud for more sophisticated analysis that requires additional computational power. It depends on what data is required to be able to characterize the fault, Siafakas said.

He said ease of use begins with the implementation of Sensata’s sensors, which can be configured or installed without tools. As an example, one of their wireless vibration sensors comes with a magnetic mount, which can be attached to a motor or pump and configured with a Sensata IQ mobile app.

Answer a few easy questions and then you can view it on the platform, Siafakas said. What makes it easy the most is that AI machine learning takes the required domain experience from the customer and packages it into models that run on the system.

AI allows the system to take on that domain expertise so that the maintenance manager or plant management can interpret those signals in an easy way, alerting them that there will be a particular anomaly or failure in the system, he added.

The key point is to avoid downtime by interrupting the failure before it actually happens and shuts down a plant, Siafakas said.

These types of solutions can also help address some of the skilled labor shortages in manufacturing plants.

One of the key trends Industry 4.0 seeks to address is the skills gap as baby boomers retire and manufacturers struggle to replace skilled labor, Siafakas said. An example is maintenance personnel who have been with a plant for 30 years and may just pass by an asset and say, “That doesn’t sound good.” New members of the workforce don’t have the same level of experience or depth of knowledge, and now you can sensor those assets, leveraging AI to be able to predict those failures.

AI is a great fit for redundant tasks that can be automated, Martin said. It’s striking the right balance between software-driven automation and human-driven action insights that empower people to do what they want to do, but also empower them with the right information so they’re physically where they need to be when they need to be. being there and doing things in an efficient and optimized way. That could include not just manufacturing the physical product, but the entire design process or after-sales service and support.