Oil analysis is essential for maintaining the reliability and life span of heavy-duty equipment. In this process, data analysts assess whether samples indicate abnormal working surface wear that can likely impede performance or shorten equipment life span. However, traditional analysis occurs after the fact. As a result, signs of wear, when finally detected, threaten optimal operation and increase the possibility of downtime.
But what if the algorithms that create predictive analytics could be harnessed to detect potential abnormalities in engines, hydraulics, and other equipment from a seemingly normal oil sample beforethey occur? In fact, such algorithms have been used successfully and are growing in acceptance in the form of an artificial intelligence (AI) platform. Combined with the expertise of data analysts and the concept of machine learning, AI has become vital for maintaining heavy duty equipment and ensuring its useful life.
Used oil analysis, also referred to as oil condition monitoring (OCM), begins with a small sample provided by a fleet to a laboratory for analysis of wear, fluid condition, and contamination. If any of these conditions are present in the sample, a human analyst recommends corrective action. Analysis per sample can take up to five minutes, which may not sound like much time, but can feel like an eternity when considering the quantity of samples submitted daily. To put this in perspective, one U.S. lab reported processing 1.2 million used oil samples in 2018.
Traditional analysis is limited in scope and scale, which creates a major problem. There are only so many data points that a human analyst has time to consider. A typical figure is 100, which can be inadequate based on the seemingly endless number of data points, and, even more important, the interrelationship of those points in determining appropriate severity classifications. Some abnormalities in a sample, such as the unwanted presence of iron and lead, are obvious signs of wear, but others may not be as readily apparent.
After analysis, most samples are categorized as normal, but thatdoes not change the time involved in analyzing each one. Even normal findings can be time consuming, which is why some in the OCM industry have turned to artificial intelligence and machine learning for more-accurate and more-efficient data analysis.
To understand how all of this applies to equipment maintenance, one must first consider the amount of available data. Accessing potentially vital information buried in the reams of this data is as vital to maintenance as it is to economics and investments. For heavy duty equipment, such as engines, gears, and hydraulics, AI is now a platform that enables the identification of potentially troubling trends in a sample. It can access thousands of data points, many of which are inaccessible through traditional analysis, and then produce a report for an analyst to review.
Machine learning is a subset of AI. Here, the focus is on patterns and relationships between data. The machine learns from historical material fed by the data analyst. Information of this type is developed into a model that enables the computer to learn. The model, unlike traditional analysis of oil samples, is not rules-based, which allows for different interpretations to be easily incorporated.
There is a direct relationship between the quality of traditional oil analysis and the machine learning model. In many ways, the knowledge of an experienced analyst is as fundamental to machine learning as it is to the AI platform. Both learn from experience in much the same way as an apprentice learns a craft from a long-time practitioner.
The reality is that the machine continues to learn after the AI platform analyzes the data prior to the analyst’s review. The impact of this process on maintenance is clear… unlike traditional analysis and its after-the-fact detection of abnormal trends, oil analysis on an AI platform can identify precursorsof wear on equipment or changes in fluid conditions.
Details from each sample are recorded into a laboratory information management system and sent to the AI platform for interpretation of the level of severity. If a sample is found to be abnormal or critical, the system, along with the analyst, assesses possible corrective actions, followed by a quality control check to assure the accuracy of the findings before sending out results. This can provide a fleet maintenance department with time to act before there is either downtime or a reduction in the equipment’suseful life.
AI AND THE FUTURE OF LUBRICANT MAINTENANCE
AI and machine learning are not threats to the continuing need for OCM data analysts. Since the platform can determine normal samples instantaneously, analysts can spend their valuable time and experience on the more detailed analysis of exceptional samples.
The findings and relationships derived from data analysis are integral to continued machine learning and the growth of predictive analytics for maintenance. Both are proving to be indispensable for maintaining heavy-duty equipment by reducing unplanned downtime and lowering maintenance costs.