Presentation Profile
AI-Driven Used Oil Analysis & Tribology Interpretation: ML Algorithms for Spectrometer/Viscometer Data and Lubricant Health & Wear
Currently Scheduled: 10/14/2026 - 1:00 PM - 2:00 PM
Room: Exhibit Hall Entrance
Main Author
Raj Shah - Koehler Instrument Company, Inc.
- Alice Kim - Koehler Instrument Company, Inc.
Abstract:
Artificial intelligence, widely known as AI, and machine learning (ML) is associated with and
contributes to problem solving mechanisms, allowing for a greater productivity and ensuring
safety for humans. With the increasing adoption of AI technologies, AI and ML have known to
become positive factors in maintaining industrial equipment. The traditional method of oil
analysis involved humans measuring and recording the spectroscopy and viscosity levels of the
equipment to assess the lubricant health of the machine and possibly detect any early signs of
gear wear. However, gathering and interpreting the data can often be time-consuming and create
risks and hazards when proper safety protocols are not followed. This review will go over how
ML algorithms can be utilized to process data from spectrometers and viscometers to automate
an interpretation from the used oil analysis and tribological data. These advancements will allow
humans to identify and prevent any equipment from becoming defective or worn out, extending
the lifespan of the equipment.
