Industrial autonomy: How machines will perform their own maintenance

Feb. 9, 2021
In this installment of Automation Zone, learn how self-aware assets will be able to diagnose and schedule their own maintenance.

Once again, a shift is taking place in the industry. The shift is part of digital transformation as industry begins to move from industrial automation to industrial autonomy (IA2IA). Industrial autonomy transcends industrial automation by adding layers of smart sensing and machine cognition to anticipate and adapt to unforeseen circumstances – thus augmenting or removing the need for human intervention. Industrial autonomy is enabling companies to develop new and enhanced capabilities in all areas including production, planning and scheduling, distribution and supply chain management, engineering, field operations, and maintenance.

For operations and maintenance, this means transitioning from low levels of automation where humans perform all operations, to higher levels of automation where the operations are accomplished through collaboration between humans and robots, and eventually to autonomy. Industrial autonomy will not only enable a predictive maintenance application to determine whether issues exist, it will also guide humans and robots through remedial actions with instructions to accomplish them, or the application can complete the task on its own without human intervention.

Industrial autonomy will help companies achieve their goals to centrally monitor assets and, in some instances, conduct unmanned remote operations with only periodic maintenance campaigns. This is particularly attractive for complex, remote, or hazardous operations. 

What is autonomy?

Autonomy means to be independent or to be able to control and govern oneself by learning and adapting to the environment. This contrasts with automation, which encompasses a series of highly structured and preprogrammed tasks that require human supervision and intervention. This is particularly the case for abnormal plant operating and equipment conditions (see Figure 1). 

About the Authors: Penny Chen and Tom Fiske

Industrial autonomy maturity is defined by six levels. Level zero represents completely manual operations. In the semi-automated stage (level one), there are many manual operations with mostly paper-based instructions, tracking, and recording results. 

The automated stage (level two) is the state of the industry today. Automated systems conduct most production processes and aid in workflows and maintenance tasks. The limited connectivity between disciplines (silos), such as engineering, design, supply chain, manufacturing, and maintenance, severely limits collaboration and real-time decision-making.

Semi-autonomous (level three) is characterized by a mixture of autonomous components and automated assets with human orchestration. 

In autonomous orchestration (level four), most assets operate autonomously and are synchronized to optimize production, safety, maintenance, and other functions. It brings together autonomous components that perform as a system. However, in autonomous orchestration, not all disciplines are integrated, and people perform many functions. 

Autonomous operation (level five) is a highly idealized state that is difficult to attain and may not be realized in the short- to mid-term. It represents a state in which facilities operate autonomously and are integrated with multiple disciplines that also operate autonomously. This stage extends to supply chain partners and brings together multiple systems to operate as a whole.

Asset management, predictive maintenance, and field operations

The industry has a diverse mixture of old and new assets to operate and maintain. Artificial intelligence allows companies to identify anomalous asset conditions that lead to downtime and accidents. Knowing when an asset is predicted to fail allows companies to schedule maintenance or repairs before the failure. 

Autonomous systems will transcend traditional predictive maintenance. For example, control valves with embedded sensors for temperature, pressure, and sound will be able to operate autonomously, determine their maintenance needs, and coordinate service requirements to minimize production disruptions. 

What does IA2IA look like at different stages for field operators and maintenance crews? 

  • Manual – Everything including instructions and paper-based record keeping is performed manually.
  • Automated – This might include some automated functions, condition monitoring, and predictive analytics. Even in this scenario, many tasks such as opening and closing valves, reading gauges, and making visual inspection rounds, are performed manually. 
  • Semi-autonomous – There is a mixture of manual, automation, and autonomous components.  Advanced technologies and analytics are used for predictive and prescriptive condition monitoring and asset management. There are self-aware instruments and equipment capable of determining optimal outcomes. For example, an autonomous pump might have a leaky seal; however, the pump will know the context in which it operates. If the fluid is hazardous, the process must shut down immediately. If the fluid is water, it might be able to continue until it can be repaired. Perhaps the flow could be diverted or slowed down to allow the process to continue. 
  • Autonomous orchestration – The pump determines what to do and when it needs to be fixed. It could shut down the process if the material is hazardous. It could divert the flow if there is auxiliary equipment. If the pump is connected to the asset management system, it could schedule a time for maintenance to fix it. It would select a time that minimizes disruption to the process. It will be able to select a technician with the appropriate qualifications to perform the maintenance task.   
  • In autonomous operations, the pump is self-aware and will proceed through the same process as it would in autonomous orchestration. However, this time, the facility will 3D print a part and a robot will perform the maintenance task.  

Drones and robots

Many types of drones and robots are becoming popular for numerous applications in the process industries. Remotely operated aerial drones are used in many inspection operations. Robots come in a variety of form factors that are optimized for specific applications. As more and more intelligence is embedded in drones and robots, they will be able to perform an increasing number of tasks without human intervention. The data they collect and the tasks they perform must be integrated with IT and OT data.

Process plant owners are adopting robots and drones to achieve more efficient, reliable, and safer operations. In general, robots and drones better perform mundane or repetitive tasks that humans find boring and in which they are likely to make mistakes by skipping safety procedures and ignoring important warnings. Robots will not forget or skip steps and can perform tasks in remote areas that are hazardous to humans. Thus, humans can instead perform higher value-added activities. 

In the process industries, asset inspection and maintenance are the primary, early targeted applications for robotics. To achieve optimal benefit, it is necessary to deploy, maintain, integrate, and coordinate the activities of various robots (see Figure 2).

For example, a snake robot is ideal for inspection inside a pipe. Drones perform numerous upstream oil and gas inspection tasks, such as detecting leaks and fugitive gas emissions, as well as inspecting large confined spaces. Crawler robots can move around plants and facilities to perform operator tasks. Robots with “arms” can perform simple maintenance tasks, such as turning valves, pushing buttons, painting, and replacing circuit boards in cabinets. 

Depending on the task, a robot could be equipped with gas sensors to detect leaks, high-definition cameras to read gauges, infrared cameras to measure temperature, microphones to detect abnormal noises, and accelerometers to detect excess vibration. Drones use cameras to record video from elevated inspection points and to perform ultrasonic testing (UT) measurements.

Deployment, maintenance, integration, and coordination

Deploying robots in an industrial setting requires a mobile robot management system to coordinate the numerous types of robots and applications. Information integration becomes very important in creating a common graphical interface that allows operators to view the actions and alerts from each robot and drone. Adding to the complexity of the integration task is the fact that some of this information must be sent to automation platforms and some to the asset management systems. The facility requires a platform to integrate and manage all this information effectively.

Strong automation base layer

There are many benefits to achieving some level of autonomy in the process industries. More integration of automation and domain applications will provide higher levels of productivity, flexibility, efficiencies, reliability, and profitability. It will reduce or eliminate human error, provide uninterrupted operations, and remove people from remote or hazardous environments. High levels of industrial autonomy require a strong automation base layer; the use of more intelligent sensors; remote surveillance and inspection through traditional approaches and with robotics and drones; digital twins; artificial intelligence (AI); and other analytics to monitor, predict, and mitigate process and equipment failures. Robots and drones will communicate with automation and asset management systems about their missions and perform routine operator rounds, inspections, and routine maintenance tasks.

Automation Zone

This article is part of our monthly Automation Zone column. Read more from our monthly Automation Zone series.

Sponsored Recommendations

Arc Flash Prevention: What You Need to Know

March 28, 2024
Download to learn: how an arc flash forms and common causes, safety recommendations to help prevent arc flash exposure (including the use of lockout tagout and energy isolating...

Reduce engineering time by 50%

March 28, 2024
Learn how smart value chain applications are made possible by moving from manually-intensive CAD-based drafting packages to modern CAE software.

Filter Monitoring with Rittal's Blue e Air Conditioner

March 28, 2024
Steve Sullivan, Training Supervisor for Rittal North America, provides an overview of the filter monitoring capabilities of the Blue e line of industrial air conditioners.

Limitations of MERV Ratings for Dust Collector Filters

Feb. 23, 2024
It can be complicated and confusing to select the safest and most efficient dust collector filters for your facility. For the HVAC industry, MERV ratings are king. But MERV ratings...