Is there such a thing as too much preventive maintenance?
The answer, surprisingly, is yes. It is always better to see organizations do more preventive maintenance tasks (PMs) than not enough, but to supercharge profitability related to maintenance costs, the internet of things (IoT) can actually help reduce the amount of traditional calendar-based PMs a user can do.
IoT is an acronym about as popular as a top hit on the radio, but organizations rarely understand what it is or, even more important, how it can benefit users. IoT refers to devices that are connected to each other, often through the internet, and work together or communicate. The industrial internet of things (IIoT)—a major subportion of the fourth industrial revolution (Industry 4.0)—has been making waves in the manufacturing industry, creating a new level of connectivity.
IoT’s Impact on Maintenance
In the old days, if a pump was having an issue, a user relied on human senses (sight, touch, smell, etc.)—assuming the problem was caught before catastrophic and profit-killing downtime.
For example, when personnel notices rattling noises and high vibration levels along with pitting damage to the impeller, it might be a result of cavitation.
Cavitation is when the fluid’s pressure falls below the fluid’s vapor pressure at the center of the impeller. When this happens, the liquid vaporizes, and the tiny bubbles collapse fiercely causing a water spray to hit the impeller surfaces. This results in disintegration of the impeller and formation of tiny cavities, which reduces the operating life and increases wear and tear of the seals and bearings.
Again, pre-IoT, the worst of the cavitation may not have been caught until damage occurred.
In an IoT world, sensors can detect even the slightest of vibration, pump operating temperature (is it too high?) or the inflow pressure—any of which could be contributing to the situation. An alarm in the form of a notification to appropriate staff can be dispatched to examine and correct the problem, avoiding serious damage or downtime.
In the Physical World
In any operation that relies on pumps and other assets and systems, physical sensors can be used to detect all types of metrics and data that indicate a pump’s health. Older pumps have little capability in terms of intelligence.
These assets can be retrofitted with a variety of sensors that can be attached to grab IoT data. Pressure, temperature, vibration and particulates are just a few examples of what can be monitored.
These modern sensors can broadcast the data in varying increments through the cloud/network back to staff dashboards. An operator or supervisor can easily detect issues much sooner than older manual inspection processes. More modern pumps will often have the ability to deliver this data built in. The process is similar in terms of where the data goes.
Given the wide variety (age, manufacturer, etc.) of assets and pumps in any industrial environment, the first step of an IIoT plan is to identify what data is available already and what assets will need to be retrofit with sensors—all tying back to the central dashboard.
In the Digital World
The data coming out of these assets can serve multiple purposes. Operationally, the data can help drive the following:
- overall equipment effectiveness (OEE) calculations
- real-time status of assets
- run time summaries and alarms
The computerized maintenance management system (CMMS) is an important destination for much of this data, and it can help drive superior maintenance. If a pump is giving off a high temperature alarm and shuts down, a number of critical steps can occur. The asset (pump) should be set into an offline status, possibly notifying staff.
A work order should be created, optionally auto-selecting the best person/crew/contractor to be dispatched to work on the problem. All of this can be highly automated with a CMMS interfaced with IoT systems—pumps or any asset. This could be a complement to scheduling PMs based on a calendar.
Too Much PM?
Again, it is better to do more PMs than less. However, too much preventive maintenance can be done.
For example:
- Doing PM on a major production asset every month on a regular schedule equals:
- 3 hours downtime ($2,500 in potential loss production)
- $200 in parts
- $200 in labor
- $2,900 x 12 = $34,800 annual cost
- Doing predictive maintenance (or PdM) based on usage, about every six to seven weeks, equals:
- 3 hours downtime ($2,500 in potential loss production)
- $200 in parts
- $200 in labor
- $2,900 x 8.6 per year = $25,133 annual cost ($9,667 in annual savings)
It can be just as effective to intelligently maintain pumps using a quality CMMS and IoT data with fewer actual touches on the asset.
Getting There?
On the physical side, most organizations seek an IoT partner who can assess the assets/pumps and understand what data can already be delivered (if any) and how additional retrofit sensors could deliver additional metrics, etc. A CMMS system and partner can help analyze the goals, objectives and pain points to see how the CMMS software can help. A part of this process is to investigate the relationship between that physical and digital world so that the IoT data is more than just data, but information driving superior maintenance.
Conclusion
Catching problems early can lead to asset preservation, a more efficient pump and better profitability. The power of IoT can give end users the data needed to operate smarter.