Access Official Technical PDF

Share Summary:LinkedInEmail

Disruptive innovation in industrial automation is not just about introducing a new machine or software tool. It occurs when technology makes products or services simpler to use, more accessible and less costly—and, in turn, changes how economic activity is organized. For automation engineers and operations leaders, this perspective offers a useful way to assess technologies whose effects may reach beyond a single process or production line.

A 2019 innovation outlook identified deep learning, robotics and energy storage among the platforms with potential to drive significant change. Its forecasts were projections, not guarantees. The more durable insight is the relationship between declining costs, growing adoption and further innovation—and how technologies can reinforce one another across sectors.

Disruptive innovation in industrial automation: why cost curves matter

Technologies can become more consequential as they become less expensive to deploy. The source describes a pattern in which cost declines support increased demand, while growing production can contribute to further reductions. Once a technology crosses an economic threshold, applications that were previously impractical may become viable.

This dynamic matters to industrial decision-makers because adoption is often shaped by total operating and implementation costs, not technical potential alone. A technology may be capable of performing a task, but its business case depends on whether its cost and performance suit the operation. As costs change, the range of feasible uses can change too.

The source also emphasizes that innovation platforms have effects across industries and geographies. Electricity, for example, spread through multiple manufacturing subsectors over time. That history illustrates why automation leaders should look beyond individual product categories: a foundational technology can enable changes in many parts of an industrial system.

Deep learning extends software into physical operations

Traditional software relies on programmers to specify instructions and behavior. Deep learning takes a different approach: people provide data and a learning framework, and the system is trained to identify patterns and produce responses. In the source’s account, performance can improve as the system learns from more data.

For manufacturing, this creates a potential shift in how software is applied. Rather than limiting software to predefined business processes or calculations, deep learning can extend it into activities that involve interpreting information from the world. The source identifies manufacturing as one of the large industries that could be affected by this broader reach.

One example included in the outlook is agricultural equipment using computer vision to support targeted herbicide application. It demonstrates how a trained system can interpret visual information and inform a physical operation. The broader industrial implication is that data, learning systems and machinery can be connected in new ways. However, the example should not be treated as evidence that every industrial task is suitable for deep learning; application fit and results depend on the specific use case.

Robotics and energy storage broaden the opportunity

Robotics was identified as one of five innovation platforms, and collaborative robots were listed among the outlook’s “Big Ideas” for 2019. The text does not specify a particular cobot capability or deployment model. It does, however, place robotics alongside deep learning, suggesting that automation should be understood as an area influenced by multiple advancing technologies rather than by robotics alone.

Energy storage offers another example of how a technology platform can affect industrial operations. The outlook links declining lithium-ion battery costs to electric vehicles and utility-scale storage. It argues that demand growth in one application could contribute to lower costs, potentially making storage practical in additional contexts. The analysis also describes energy storage as a possible alternative to electricity generated by underutilized plants, subject to its assumptions and forecasts.

For plant and operations leaders, the practical point is not that a specific storage investment is automatically justified. It is that cost trends can alter the options available for energy supply and storage. Evaluating those options requires attention to the facility’s operating needs and the economics of the proposed application.

Convergence changes how technologies should be assessed

The outlook’s central framework is that innovation platforms can intersect and enable further innovation. It identifies artificial intelligence, robotics, energy storage, genome sequencing and blockchain technology as major platforms, with related technologies such as cloud computing, the Internet of Things and connected devices appearing in its convergence map.

For industrial automation teams, this is a reminder to assess technologies both individually and as parts of a broader system. A machine, learning model or energy-storage resource may have a different role when connected with other technologies. That does not mean every combination will deliver value. It means the evaluation should consider dependencies, intended outcomes and the conditions required for adoption.

A disciplined assessment can ask:

  • What operational problem is the technology meant to address? Define the task before selecting a platform.
  • What makes the application economically feasible? Consider how cost changes could affect the business case.
  • What other systems or technologies does it depend on? Identify relevant data, equipment and infrastructure.
  • What uncertainties remain? Distinguish demonstrated capabilities from projections about future adoption or market size.

Plan for opportunity without treating forecasts as certainty

The source repeatedly cautions that forecasts are limited and cannot be relied upon. That qualification is important for technical decision-makers. Predictions about adoption, costs or market potential are not substitutes for evaluating a specific operation, and a technology’s promise does not remove implementation, regulatory or competitive risks.

The practical takeaway is to monitor enabling technologies while grounding investment decisions in operational requirements. Deep learning may expand software’s role in manufacturing, robotics may contribute to new forms of automation, and energy storage may change how power is managed. Their significance will depend on cost, application fit and how they interact with the wider industrial environment. Understanding those forces can help organizations identify meaningful opportunities while remaining clear-eyed about what is known, what is projected and what still needs validation.

Access Official Technical PDF

About The Author