Artificial Intelligence in machinery applications is currently moving from a vision to reality. Many current applications already use AI and the technology is poised to change what automation looks like in the near future.
Artificial intelligence will become an integral part of manufacturing and automation across engineering, operations, and maintenance. Many experts are convinced that AI will change everything. While many people (end users in particular) still struggle with the technology, technical constraints are less of a problem overall, with organizational and cultural constrains and the human factor emerging as the main hurdles.
For suppliers, AI provides an attractive market. For many machine builders AI is a “must have” to differentiate their products from those of competitors. But to succeed, clear uses cases are required. It’s up to technology providers to form best practices and use cases for reference, as many end users today do not have any AI guidelines in place.
As in any new field, collaborations are inevitable. Here, AI shows another potential for disruption. Most companies perceive that non-traditional suppliers to the operational technology space (and large software suppliers, such as Microsoft, Google, Amazon, and IBM, in particular) are the market leaders in AI. In this framework, traditional OT suppliers and OEMs need to be careful to not be left behind.
In the pages that follow, we summarize the results of our recent online survey on AI in machinery applications, supplemented with both one-on-one interviews with industry participants and specific use cases presented at the recent virtual ARC European Industry Forum. The author also previously produced a video for ARC Advisory Service clients that discusses these and other takeaways from both the online survey and associated ARC market research.
State of Artificial Intelligence in Machinery Applications
In recent years, there has been a lot of talk about artificial intelligence (AI) in manufacturing and machinery applications, which many companies perceive could provide them with competitive advantage. However, ARC has observed that there is still a lot of uncertainty in the market as to what AI is capable of, the most suitable applications for the technology, the right implementation strategies, and who is using AI already and to what degree.
What Will AI Look Like in the Future?
When asked how they believe AI will be used in future, more than 100 industry participants shared their responses.
Most respondents agree that machinery will have AI in the future, but there is no overall agreement whether AI will be reserved for high-end machinery. One possible explanation for this is that people have different perspectives on what constitutes “high-end” machinery. Also, we intentionally did not specify a time horizon for this question. ARC’s conclusion is that AI applications will start with more high-end machinery and then gradually migrate toward simpler machinery, such as palletizers and packaging machines.
In contrast, there is strong belief that AI will be deeply embedded. This may be in the controller, the engineering tool, or even right into the device.
Technical constraints do not seem to be a big issue among our survey participants, but cultural issues are. ARC confirmed this in our one-on-one interviews with selected industry professionals. AI will take decisions away from the well-understood controller and, especially when deeply embedded, the results of the AI techniques are not 100 percent transparent. This is a real drawback in a generally conservative industry such as industrial automation.
Another finding from our online survey is that unclear use cases are among the top inhibitors for AI in manufacturing. This lines up with ARC’s observations from other industries: adopting new technology for technology’s sake will not succeed.
Where We Are with AI in Manufacturing?
Many technology suppliers now offer AI-enabled products and many machine builders have started to evaluate the technology. However, there are several roadblocks, most prominent among these are the lack of data scientists, lack of available data, legal aspects, human factors, and – finally – unclear use cases.
ARC market research on AI in machinery applications identifies the current distribution of AI within the market. The blue line in the chart at left summarizes these. As readers can see, maintenance applications are particularly prominent. The green bubbles in the chart represent the case studies presented during a dedicated AI for Machinery workshop at our recent ARC virtual European Industry Forum.
A key challenge for machine builders and automation suppliers alike is to identify the right application for AI. The top three applications that emerged from the survey are in line with recent ARC market research, though in a slightly different order. The fact that motion planning ranks low is self-explanatory, as it only affects a small number of machines, mainly robotics, whereas quality control can be applied almost everywhere.
Maintenance (predictive and prescriptive): Maintenance is one of the main applications for AI in machinery. Growth is driven by end user needs for machine efficiency and to minimize or eliminate unplanned downtime. Most predictive maintenance solutions use simple machine learning like supervised or unsupervised learning to detect failure patterns for parts from machine data and predict when the next machine part failure will occur.
Quality control: Intelligent quality inspection is a key AI application in the machinery space. The main driver is the stringent quality requirements for consumer products. Quality control applications cut across many industrial and machinery segments in automotive, packaging, and food & beverage. Machine vision for quality inspection has already advanced. Next steps will see the growing demand for intelligent systems for quality control along all production steps. This can also be a good use case for deeply embedded AI, where the techniques work within the camera.
Operational simulation and optimization: This is another large application segment for AI in machinery. Dynamic simulation and optimization of processes allows end users to plan their machine use efficiently, plan material flow and supply dynamically, and anticipate possible shock scenarios. Key drivers of growth in this segment are the need for end users to lower overall operating cost and the rise of physics-based AI solutions. As more manufacturing lines become integrated with the supply chain and processes become more complex, demand for AI solutions that target operational simulation and optimization will grow.
Energy management: AI for energy management is a slower growing market area due to various factors, including:
Lack of transparency of energy use in plants, especially process plants
Lack of dynamic pricing for electricity, which fails to allocate real costs
Unclear responsibilities at end user plants for energy management
Difficulty in predicting energy availability at the time of use
High costs of tightly integrating various energy sources (renewable, fossil) in the production process
How Will the Market Grow?
One of the key questions typically asked about an emerging market is the anticipated growth.
According to our survey, participants expect around 20 percent market growth, which is in line with past ARC estimates. Not surprising, most participants see AI as an opportunity. Here, we need to stress that most survey participants are likely to represent companies that have already expressed interest in AI technology, so the results are likely to be skewed in a positive fashion.
With a growth rate of slightly under 20 percent, AI is growing slower than cloud application platforms, but significantly faster than overall machinery automation, which has an average growth of around 3 percent. This means that AI will strongly penetrate the control architecture of many machines over the years to come and likely to become an omni-present technology sooner, rather than later.
Technology will always have some impact on market growth. But as demonstrated by industrial PCs (IPCs), manufacturing execution systems (MES), or other automation technology, the presence of new capabilities often has impact far beyond the technology itself. The rise of IPCs in manufacturing created a vibrant market, but also dramatically transformed what programmable logic controllers (PLCs) are capable of. So much so, that the difference today is more about the look and feel than the actual performance. AI can be integrated into engineering and programming tools with embedded natural language processing (NLP) autocorrect features, or by automatically suggesting code and changing programming controllers. With the rise of low-cost AI chips, AI could become an essential part of any controller on the market within a decade or so. AI could also move even further down in the architecture to become an integral part of IO modules, sensors, and actuators. Several experts we interviewed forecast that AI will be omni-present, so we won’t even notice it is there. In this case, the market for AI-centric solutions may shift to higher performance applications using more powerful edge devices.
Over 50 percent of the survey participants think AI will fundamentally disrupt the way we manufacture products; a much smaller number of people think it will have only limited impact. Yet, automation and machinery control are slow-moving markets and it will take time for users, machine builders, and suppliers to offer and use AI on a broad scale.
Table of Contents
State of AI in Machinery Applications
Inhibitors and Roadblocks
Strategies for Success
Sample and Methodology
ARC Advisory Group clients can view the complete report at ARC Client Portal
If you would like to buy this report or obtain information about how to become a client, please Contact Us