ChatGPT in Lubrication
In what has become the most buzzworthy trend of the past decade, Artificial Intelligence (AI) is making huge waves in everything from art to essays and even diagnosing machine issues. It has started a conversation around what the confines of the technology should be and how much impact this will have on people’s jobs and livelihoods.
While similar to the conversation surrounding robots taking over certain duties, the AI conversation is very interesting because the battleground isn’t necessarily what physical work could be taken over but what logical or cerebral work can be accomplished by AI.
Many commercials and fun videos have shown the harmonious union between people and technology to perform all sorts of tasks. Robot dogs dancing to music and sarcastic software giving Tony Stark a hard time are examples of how culture is beginning to anthropomorphize tools that are more sophisticated than what we are used to.
Robots aside, AI is continually being scrutinized as to whether it is sentient or able to display emotion, or in some cases, forcing the question of what emotions or feelings truly are.
The Impact of Artificial Intelligence
Outside of students using Chat GPT to write their homework (or consultants using it to write their articles), how useful is this technology? Let’s take it a step further and ask, how will this impact industry, and specifically, how will this impact a Lubrication Technician’s daily work? To best answer these questions, we must first understand the typical work of a Lube Tech.
Lubrication is a skilled trade requiring specialized training and tools to perform it correctly. Sometimes this work is shared across multiple parties, such as maintenance, reliability, and operations, and sometimes it is designated to a single team. Regardless of how it is staffed, the work has to be completed in an optimal manner to ensure that the machine operates effectively and reliably, all while extending the component’s life.
Let’s start by looking at the normal activities within a typical lubrication program.
Typical Lube Tech Activities
Greasing of Components
This is where high demand for manpower typically exists. While it depends on the type of facility, it is not uncommon for there to be thousands of grease points that must be visited multiple times over the course of a year. The Lube Tech must apply the correct grease with the correct volume at the correct time to each of these fittings.
Oiling of Components
Like greasing, multiple components must either be periodically refilled with oil, particularly total-loss systems, or the oil must be completely changed based on interval or condition. This involves draining the fluid, refilling it with new fluid, and then disposing of the used oil.
Knowing the correct fluid and performing this to get the appropriate fluid volume and minimize opportunities for contamination is key.
Inspections
Often, this activity is shared across multiple departments. Proper inspections should provide information on the equipment's external and internal conditions and operating parameters. Oil level, temperature, differential pressure, and breather condition are all examples of valuable data that should be collected and used to trigger corrective action when needed.
Some of these may result in changing a lubricant, filtering the lubricant, alerting maintenance and operations of a pending failure, or a litany of other things that might have to happen.
Change Outs
Several consumables are often attached to equipment that helps with lubrication or contamination control. Some examples would be:
- Replacing a single-point lubricator.
- Changing a filter that has hit a high differential pressure.
- Swapping out a breather that has become saturated with moisture.
All the changeouts require having the correct replacement parts on hand and are often the result of inspections.
Lube Room Duties
The lube room is the heart of the lubrication program and, as such, must be managed in the same manner as a critical asset. The Lube Tech must be involved in cleaning and organizing the room, but, more importantly, they must ensure they have the appropriate volumes of lubricants and consumables on hand. They will look at future planned activities, order these products, and work with established lead times with vendors.
Lubrication Sampling
With the desire to move to a more condition-based approach with maintenance and lubrication, sampling of equipment is beginning to be a bigger component of a Lube Tech’s work. This includes extracting representative lubricant samples, sending them to a lab for analysis, interpreting results and scheduling corrective actions.
This is just a simplified listing of activities, but you can already see how involved the daily work in a lubrication program could actually be.
Lube Tech Activities with AI
Now, let’s layer over where AI could be effective or help supplement the work that has to be completed.
Greasing and Oiling of Components
We are far from employing AI to lubricate equipment on a routine basis. There are some options out there that utilize condition monitoring to apply lubricants when the machine needs it. Still, those are typically reserved for specific applications, and using them on every point is not cost-effective.
AI could help in determining the proper frequency of application and the proper lubricant to use. Each application is unique and, as such, may require something slightly different from the lubricant and how often it needs to be applied. Subtle changes in additives and viscosity can yield significant improvements in reliability when matched correctly.
Inspections
There has been a great deal of buzz around the use of technologies such as robotics and drones to perform machine inspections; in some cases, this is already happening. The biggest impact AI could have is interpreting the results of inspections or identifying small changes in a machine’s condition that might have otherwise gone unnoticed.
However, I do not foresee a tide change replacing the human inspector as there are too many peripheral things a person might pick up on that a machine might not where a coexistence between people and machines would be incredibly beneficial in this area.
Changeouts
Proper accessory selection is important and similar to the lubricant conversation; many variables must be weighed to ensure the proper filter, breather, sight glass, etc., is selected. Giving these variables to the AI and other information, such as pricing and availability, could remove this responsibility from the technician or manager and make it more automated. This, of course, depends on the amount of information available from the manufacturers of these types of products.
Lube Room Duties
Short of robot vacuums, there has been little in the market for automated cleaning and organization. While the Lube Techs will largely manage the daily cleanup work, the procurement of lubricants could be offloaded to the AI.
Provided it can review past usage, current volumes, and future planned maintenance activities, it could interface with the purchasing system to ensure the timely delivery of lubricants and consumables to minimize expedited deliveries and product swaps due to lead times and availability issues.
Lubricant Sampling
The biggest value in this area comes from interpreting results from the sampling. Many laboratories are already using AI in some fashion to look at larger sets of data and provide better alerts to issues that may be occurring inside the machine.
With a good set of historical data, the AI might be able to track and perceive incipient failure more easily than someone just looking at the data without understanding the full scope of trending.
Conclusion
The industry is mainly on the upsweep of the “hype curve” in relation to AI, but there will definitely be significant adoption in many facets of operations, maintenance, and reliability. What all of this will look like is yet to be fully realized.
Like any new tool or technology, this will require a cultural shift in many facilities, which can cause friction and frustration. The most successful users of this technology will find a way to imbue it with the proper data so that it can make meaningful recommendations and contributions to their work.
Regardless, this technology will be impactful in many ways, and hopefully, it can be put to work, helping drive reliability into the common practice of the industry. Meanwhile, I’ll continue using it to write song lyrics and jokes about my friends and family.