Maintenance automation: create a new recipe for data integration

 

Process manufacturing has reaped automation’s benefits for decades. Under the guidance of ISA-95 Enterprise-Control System Integration from International Society of Automation, organizations looking to adopt automation standards-based best practices. As new technologies emerge, there are more ways to apply the automation standard.

Under Industry 4.0, digitalization has spawned countless new manufacturing technologies and systems, in departments and industries far outside process manufacturing. These include:

• Service-oriented architecture (SOA). It is considered crucial for the success of complex enterprise resource planning (ERP) systems and has many advantages for integration, extensibility, agility, and reusability.
• Industrial Internet of Things (IIoT). This technology is no longer a buzzword; it is changing how we monitor and interact with critical equipment, such as smart sensors.
• Wi-Fi and digital networks. These are becoming a primary alternative for industrial connectivity, with 5G wireless networks commonly discussed as a promising technology to enable ubiquitous and scalable connectivity to the shop floor for industrial wireless communications.
• Big Data, edge computing, robotics, artificial intelligence (AI), machine learning, application program interfaces (APIs) and augmented/virtual reality (AR/VR). These have moved from concept to pilot and will be ready soon for broader adoption.

Moving technological growth forward was challenging enough, but then the COVID-19 pandemic hit. The disruption from COVID-19 is fueling urgency among plant managers for system improvements previously viewed as “future considerations.”

Remotely monitoring asset condition, for example, has gone from a “want” to a “need” – something essential to be effective in the near term and competitive over the long run. As of late summer 2020, some 82% of maintenance organizations were considering how to effectively add or increase digitalization technologies, according to Fluke Reliability research. Meanwhile, more than 35% have experienced at least a quarter drop in production, with only 15%  operating as “normal.”

Many of these companies understand how automation could deliver a valuable step change in their day-to-day operations. Many previous IIoT pilots have failed to live up to expectations. Sometimes this was due to system conflicts, technology limitations, or, more often, gaps in communication and change management.

Today’s manufacturing environment is leaner than usual. Planners seek to optimize initiatives to their fullest extent so when people are working on-site, work is prioritized and executed based on how it fits into the big picture.

To do this effectively, teams need more than tribal knowledge or ad hoc standard operating procedures. They need a better framework for successful digitalization.

Borrow a structural framework: 5 levels of enterprise and control systems

ISA-95 and the automation pyramid have provided a critical framework for automating the interface among the five levels of enterprise and control systems:

1. Physical production processes
2. Sensors
3. Monitoring and supervision
4. Manufacturing operations management
5. Business planning and logistics.

ISA-95 was developed because the differences across industries, systems, departments and objectives made it challenging to communicate and collaborate. Businesses suffered in the absence of a common language. Now in its 35th year, ISA-95 continues to be revalidated and incorporated into more technologies. Many of these technologies and programs continue to focus on process manufacturing.

Automation system integration helps other industries

Proven systems-integration thinking is needed now outside of process manufacturing. By extending the ISA-95 framework to other industries, many can achieve the same benefits from yield improvements to cost reductions.

Automation thinkers often use a pyramid to show the interaction between levels. The automation pyramid example in Figure 2 is updated to include the IIoT at the physical layer.

What if the automation pyramid was adjusted to reflect the levels or layers found in manufacturing operations in non-process industries? It might look something like Figure 2.

The pyramid’s layers help identify the data points from various subroutines useful to aspects of the overall system. Reconfiguring the pyramid to support maintenance and reliability operations (MRO) adds a “process” layer at the bottom. It adjusts the other levels to reflect maintenance interactions with the larger plant and organizational architecture.

With the proper framework to support internal collaboration and integration, digitalization could better help pull operations through times of tight resources. But, just like in 1995 process manufacturing, ROI must be a central consideration from the beginning. For the pyramid to be useful, it has to help maintenance leaders identify where connected systems will have the most significant positive impact at their facilities.

Applying the automation pyramid framework to asset management

In most facilities, the more connected maintenance and operations teams become, the more benefits their actions generate.

Consider the process of troubleshooting a particular asset. Technicians may take a variety of measurements to get to the root cause of an issue, but they use those measurements only at that moment. If this data can be preserved in the asset’s health history, it contributes to the broader maintenance strategy for the machine – potentially adjusting how often specific parameters are checked, contributing to fewer unplanned failures, and extending an asset’s overall lifespan.

Manufacturing automation pyramid levels 

Level 1: The process

• This layer includes the physical assets, the maintenance and reliability teams (M&R) that care for them, and the machine operators. 
• Some assets are wired or wirelessly connected, while others operate out of visual management 

Level 2: Field-level tools and sensors

• Equipment–level tools and sensors provide manual as well as automated readings; not all data from un-connected measurement devices is currently saved or tracked, and not all sensor data is in a useful format. 
• Wireless sensors applied to legacy machines are bringing a more significant percentage of assets into electronic view. 

Level 3: Supervisory control and data acquisition (SCADA)

• While most manufacturers employ SCADA/programmable logic controller (PLC) systems, less than 30% can use this information for asset management. 

• Early artificial intelligence (AI) studies of SCADA data lakes are finding applicable health indicator data for triggering work orders, creating the potential for meaningful real-time, early-detection work orders.  
• Integrating level–3 SCADA data into the maintenance automation/enterprise asset management (EAM) framework improves overall equipment effectiveness (OEE) and mean time to repair (MTTR). It also reduces downtime and supplements skills shortages by helping automate work order creation and prioritization. 

Level 4: Enterprise asset management (EAM)

• The EAM or computerize maintenance management system (CMMS) is the M&R system of record. 
• Integrating the EAM with the maintenance automation framework supports shifting work from calendar– to condition-based PMs, prioritizing work based on actual asset status. 

• Tying meaningful data to the asset management system equates to a step-change in the quantity and quality of data informing daily decisions. 

Level 5: Enterprise resource planning (ERP)

• Integrating layers 1 through 4 into the company’s primary business layer feeds accurate M&R data into the overall financial accounting and reporting system. It increases the visibility of M&R contributions to the company’s bottom line and empowers data-driven decision making at the plant management level. 
• Key metrics include total cost of ownership for assets and cost comparisons between planned versus condition–based work. 

 

Optimize equipment lifecycles with P-F curve, FMEA, RCA

What if the automation pyramid is applied to best practices in asset management? Reliability engineering uses several frameworks to optimize equipment lifecycles. Those are the potential-failure (P-F) curve, failure modes and effects analysis (FMEA), and root cause analysis (RCA).

In particular, the P-F curve illustrates an asset’s progression toward failure. The X-axis represents the time to failure, beginning with an asset’s design and installation, and the Y-axis represents an asset’s resistance to failure. Reliability engineers use the curve in combination with their knowledge of an asset’s failure modes to plan predictive data gathering and condition monitoring.

If we marry the logic of the maintenance automation pyramid in Figure 2 with the asset lifecycle logic of the P-F curve in Figure 3, we might be better able to determine what kind of smart sensors and other connected reliability technology to apply where and when. Figure 4 illustrates how the two frameworks support each other.

Keep in mind: Deriving maximum benefit (ROI) requires not just extending the machine lifespan but also making the absolute best use of personnel’s time and effort. The automation pyramid could be the infrastructure that industries have needed to advance maintenance operations to be more remote and create a scalable architecture we can build off (see sidebar).

Return on investment

The short-term ROI gains from maintenance automation illustrated in Figure 5 should happen within 30 days, as mean time to repair (MTTR) and time to diagnosis reduces. This drives an increase in asset availability and unit throughout. However, the most significant short-term maintenance and repair (M&R) savings comes from a reduction of calendar-based planned maintenance (PM) events. Preventable costs include:

• Cost to carry and consume parts
• Wrench time
• Health, safety, and environmental considerations
• Downtime
• Risk of damage.

By recording, tracking and monitoring machine health data from actual operational run time to oil analysis and other condition status information, maintenance operations teams have the data they need to reduce the total number of PMs they conduct.

Long-term maintenance automation benefits increase as integration improves up and down the pyramid – and as the library of usable and accurate data grows and shifts from “Big Data” to “actionable data.” While artificial intelligence (AI) and business intelligence (BI) solutions at scale are still premature, they are making progress. Now is the time to prepare data for useful interpretation.

Augmenting the field technician

With the digital acceleration and other complications of 2020, augmenting the technician in the field has become a much higher priority. The pyramid doesn’t change the work process, so much as it accelerates when team members receive a notification and when the process begins. Integrating data from the supervisory control and data acquisition (SCADA) level to the field level allows for early detection, without someone being there to see it or someone having SCADA expertise.

A maintenance automation project of this kind should help teams in the short-term focus on what they do best – by allowing them to focus on the critical and essential nature of their jobs, prioritize work, and be immediately notified of urgent asset events.

Further tying maintenance data into enterprise resource planning (ERP) helps justify operational expenses and activity. For many, this is the only way to help other groups across the organization genuinely grasp the value maintenance operations deliver.

In the long term, a maintenance automation strategy should increase the team’s ability to rely on condition-based maintenance (CBM) and decrease the number of calendar-based PMs.

The change from scheduled maintenance to doing what actually needs to be done has many documented benefits, from labor and parts cost to safety. It can be a difficult transition for most teams to make. Clear communication, change management, and true partnership are essential.

Brian Harrison, CRL, is industry lead for IIoT at Fluke Reliability. Edited by Chris Vavra, associate editor, Control Engineering, CFE Media and Technology, cvavra@cfemedia.com.

MORE ANSWERS 

Keywords: process manufacturing, automation pyramid, maintenance automation

CONSIDER THIS 

What benefits could your company derive from the maintenance automation pyramid?

Do you have experience and expertise with the topics mentioned in this content? You should consider contributing to our CFE Media editorial team and getting the recognition you and your company deserve. Click here to start this process.