Three pillars to a reliability program
Moving toward proactive maintenance requires a foundational change.
As many maintenance managers are aware, starting a reliability program can be very tough. What many people don’t think about is that sustaining a reliability program can be even tougher. From a real-world perspective, there are foundational pillars to a healthy and sustainable reliability program.
In a survey from a recent Fluke Webcast, 56% of attendees describe their reliability journey from reactive to condition-based maintenance as "having some proactive maintenance scheduled on certain equipment." The full survey results revealed that:
- 13% of maintenance is all reactive, and they have never attempted to move beyond reactive
- 17% have abandoned proactive in the past, but are starting again
- 10% have a well-developed proactive maintenance program.
The three main causes of failure within a reliability program are poor program structure, technology selection and data management. There are also three pillars that can help any organization drive commitment to a reliability program and change culture. They are:
- New program start-up
- Technology selection
- Data management.
These three pillars to a proactive maintenance program can decrease downtime, prevent failures and reduce costs. While change can be hard, basing a reliability program on these three pillars increases success and can improve overall maintenance capability.
Pillar #1: New program start-up
Asset criticality rankings are used to help prioritize maintenance work and to identify the most critical assets for an organization. Many people find that there are so many critical items to maintain that it becomes overwhelming. Traditionally people think of their criticality list in one of four ways: binary, dynamic, every asset on its own schedule and full coverage.
However, these approaches end up being inflexible, unsustainable and miss the root cause of the dilemma of criticality: organizations have more assets than the maintenance team has capacity to manage. The first step to decreasing the impact of the criticality dilemma is to establish better classification. Consider these four classifications for assets (Fig. 1):
- Star athletes: The percent of production or compliance is correlated directly with this asset’s performance. It needs constant assessment and optimization regardless of condition, and always must be running at peak performance levels.
- Critical: The performance level of this asset is not as important as simply "running or not running," but the uptime of this asset is correlated to producing revenue.
- Semi-critical: Downtime or failure of this asset puts a strain on production or compliance. Daily processes may be able to continue sub-optimally even if the asset fails.
- Non-critical: Production or compliance are not affected by this asset. While there may be other reasons to fix this asset, it is not because of direct production loss.
Over multiple generations, medical professionals have struggled with the issue of criticality. As a result, they evolved a tiered operating approach:
- Levels of training and certification
- Levels of workers
- Volume of visits/inspections
- Amount of time spent on each person.
This tiered operating approach can be applied to the maintenance world. With a tiered maintenance program, organizations waste less time analyzing healthy machines, reduce the number of work orders flowing through an organization and avoid deploying experts on simple faults. Here is an example of tiered maintenance (Fig. 2):
Pillar #2: Technology selection
Maintenance technologies can offer basic information or advanced information depending upon the skill and experience of the user. However, different assets require a mix of technologies, such as electrical, thermal and/or mechanical. There are a few keys that all teams need when operating in a tiered structure across multiple measurement technologies:
- Real-time data entry
- Real-time issue escalation
- Additional context
- Consistency and repeatability.
Pillar #3: Data management
Finding the answers and root causes among the data can feel like finding a needle in a haystack because more data doesn’t automatically make finding problems easier. Maintenance teams need more of the right kind of data to make real change. So, what is the right kind of data? Data that passes the "ACID Test." It requires:
- Analysis: The data must be able to be analyzed.
- Context: The data must be viewed in as rich a data context as possible.
- Integrity: The data must be accurate and secure.
- Democracy: All team members should be able to contribute to and consume the data.
To fully leverage cloud technology and the Industrial Internet of Things (IIoT), the vision is to take data from hand held tools and devices and connect it to a central cloud, where a computerized maintenance management system (CMMS) will be fully integrated. With this type of integration, when equipment condition falls below a specific threshold, a work order will automatically be issued to take corrective action. It also will become possible to seamlessly order all parts needed for that repair. Assets will be managed in one central place to make the information available to all members of the maintenance team.
The end goal of the program should be all of your maintenance people working together in a tiered approach and basing the type of work on the class of asset being serviced. The technology should help maintenance personnel prioritize the tools and techniques that will prevent the most common failure modes. The effective collection and sharing of data will enable analysis in the proper context with the proper data integrity and security, and make it available to all maintenance team members when they need it.
Tyler Evans is a business unit manager for Fluke Corp. John Bernet is an application specialist for Fluke Corp.
See additional stories from the Plant Engineering May 2017 cover story below.