Predictive maintenance: what it is, how it works, and why It matters for modern industry

Predictive maintenance has become one of the pillars of modern industry, especially in sectors where reliability, uptime and cost control are critical. Unlike traditional approaches based on scheduled interventions or on repairing failures after they occur, predictive maintenance relies on real-time data, advanced analytics and continuous monitoring systems to anticipate issues before they lead to downtime or equipment damage.

Through smart sensors, diagnostic algorithms and condition-monitoring technologies, this approach makes it possible to understand the actual health of machinery and intervene only when necessary. The result is greater efficiency, improved safety and higher availability of production systems.

This article explores how predictive maintenance works, which technologies make it possible, and why it is becoming a strategic asset for companies aiming to achieve more reliable, sustainable and competitive industrial processes.

What predictive maintenance is and how it works

Predictive maintenance—often referred to as Condition-Based Maintenance (CBM)—is based on the continuous monitoring of the real operating conditions of machinery. Its goal is simple but strategic: to predict failures before they occur, intervening only when data shows an anomaly or a deteriorating trend.

Unlike preventive maintenance, which schedules interventions at fixed intervals regardless of equipment condition, predictive maintenance uses measurable and continuously updated information to determine the optimal moment for a maintenance action.

Its effectiveness relies on three main elements:

1. Continuous monitoring of operating conditions
   Sensors and diagnostic devices gather key parameters such as vibration, temperature, pressure, load or lubricant quality. These values are analysed over time to detect significant changes.
   
2. Analysis of physical and chemical parameters
   Many failures begin as micro-variations imperceptible to the naked eye. Tracking these fluctuations allows early detection of wear, contamination or component degradation.
   
3. Advanced diagnostics
   Techniques such as optical emission spectrometry (OES), trend analysis, machine learning models and predictive algorithms make it possible to interpret data and forecast potential failures, enabling timely and targeted interventions.
   

When implemented correctly, predictive maintenance brings concrete advantages:

• reduction of unplanned downtime,
• optimisation of maintenance costs by avoiding unnecessary or late interventions,
• extension of component lifetime,
• improvement of operational safety,
• more efficient planning of production activities.
  

Thanks to the integration of continuous monitoring and advanced diagnostics, predictive maintenance has become a standard for industries that demand reliability and operational continuity.

The importance of oil analysis in predictive maintenance

Among the various techniques used in predictive maintenance, lubricating oil analysis is one of the most effective. It allows operators to assess the real condition of components without interrupting operation. Oil works as an “internal messenger”: it circulates through every part of the mechanical system and carries with it traces of wear, residues, contaminants and degradation products.

Regular oil analysis provides a clear and accurate picture of machine health. It makes it possible to:

1. Identify wear metals or contaminants
   Elements such as iron, copper, aluminium, lead or nickel reveal abnormal wear of components like bearings, gears, bushings, pumps or hydraulic systems.
   
2. Evaluate lubricant chemistry
   Processes such as oxidation, nitration or dilution compromise the ability of the oil to protect components. Monitoring these behaviours supports timely and targeted maintenance actions.
   
3. Monitor mechanical component health
   Each machine has its own “metallic fingerprint.” Variations in elemental concentration highlight what is happening internally and help detect anomalies before they evolve into failures.
   

Spectrometric oil analysis: principles and advantages

Spectrometric analysis is one of the most advanced and widely adopted methods for evaluating mechanical wear and lubricant condition. The most common industrial technique is Rotating Disc Electrode Optical Emission Spectrometry (RDE-OES), a fast, accurate and highly repeatable method ideal for predictive monitoring.

At the core of this technology is the Rotating Disc Electrode system, which enables oil to be analysed without any sample preparation. A rotating electrode forms a thin film of oil, an electric arc vaporises a small quantity of sample, and an optical system identifies the emitted wavelengths to quantify wear metals, contaminants and additive-related elements.

The technique’s main strengths include:

• no sample preparation required,
• rapid and repeatable results,
• high sensitivity to metallic elements, even at very low concentrations,
• simultaneous detection of wear metals, contaminants and additive elements,
• seamless integration with predictive maintenance strategies.
  

For these reasons, RDE-OES has become a cornerstone of predictive diagnostics in industrial plants, vehicle fleets, energy systems and other high-criticality applications.

ASTM standards for oil analysis and RDE-OES

ASTM standards serve as the international benchmark ensuring that spectrometric oil analysis is performed consistently, accurately and comparably across laboratories and industrial environments.

The most relevant standards for RDE-OES are:

• ASTM D6595, which defines procedures for determining wear metals and metallic contaminants in used lubricating oils. It outlines how to measure elements such as Fe, Cu, Al and Pb, detect contaminants like Si or Na and identify metallic elements associated with additive packages.
  
• ASTM D6728, which covers the determination of metallic contaminants in turbine and diesel engine fuels. It is widely used in sectors such as aerospace, energy and defence, where fuel quality directly affects safety and performance.
  

It is important to note that RDE-OES detects only metallic elements and does not provide information about the organic composition of lubricants or fuels. The method is also particle-size dependent: larger particles may not be fully vaporised, potentially leading to underestimation. For this reason, results are best interpreted through historical data, trends and complementary diagnostic techniques.

Complying with ASTM standards ensures precise, reliable and internationally comparable measurements, a crucial requirement for industrial diagnostics.

GNR and the RotrOil Line: innovation in predictive diagnostics

GNR Analytical Instruments, an Italian company with over forty years of expertise in spectrometric technologies, is internationally recognised for its contributions to metal and lubricant analysis. Over decades of development, GNR has become a trusted partner for laboratories, power generation plants, steel industries, vehicle fleets and other high-criticality environments.

Within this framework, the RotrOil line represents GNR’s most advanced technological contribution to predictive maintenance. Based on RDE-OES technology, RotrOil spectrometers deliver fast, repeatable and ASTM-compliant oil analyses, enabling reliable monitoring of machinery health.

RotrOil Automatic is designed for industrial laboratories that handle high sample volumes. It features an automatic carousel, full measurement-process automation and high analytical stability, making it ideal for large plants, energy facilities, petrochemical industries and external labs.

RotrOil R2 and R3, on the other hand, offer compact, robust and portable solutions suited for in-lab or in-field analysis. They require no sample preparation, fully comply with ASTM D6595 and provide an excellent balance between performance and operating cost.

These technologies allow companies to reduce analysis time, obtain precise and repeatable results and integrate predictive diagnostics more effectively into production workflows. RotrOil instruments are used worldwide to safeguard critical assets, prevent unplanned downtime and optimise lubricant management.

• Discover RotrOil Automatic – automated spectrometer for high-volume oil analysis
  
• Explore RotrOil R2/R3 – portable, ASTM-compliant RDE-OES spectrometers
  
• Browse the full RotrOil line – complete solutions for industrial oil monitoring
  
• Visit GNR – instruments, applications and innovations in industrial diagnostics
  

Industrial applications of spectrometric oil analysis

RDE-OES oil analysis has become a key diagnostic tool across numerous industries where continuity, safety and machinery reliability are critical.

In the energy and power generation sector, turbines, generators and transformers require constant monitoring. Oil analysis helps detect wear, contamination and internal anomalies that could lead to costly downtime.

In the automotive and railway sectors, engines, transmissions and hydraulic systems produce wear particles that can be detected early through spectrometry, supporting failure prevention and fleet optimisation.

In heavy industry and steelmaking, machinery often operates under extreme conditions. Monitoring lubricant condition helps identify accelerated wear, particulate contamination and degradation, thereby reducing unplanned shutdowns and improving overall plant efficiency.

In aerospace and defence, diagnostic precision is essential. Spectrometric oil analysis helps detect micro-wear on critical components, lower operational risks and plan timely, targeted maintenance interventions.

---

FAQ