Predictive maintenance (PdM) is a general method that uses ongoing analysis of operational data to determine when equipment maintenance will be required. When applied appropriately, it can reduce maintenance expenses while improving reliability.
Maintenance of GMP facilities is necessary to ensure that they operate in a qualified and validated state and are fit for their intended use.
Traditional preventive maintenance has direct costs, may require downtime of equipment or entire production lines, and may increase the risk of contamination.
Preventive maintenance scheduling is often driven by elapsed time of use (e.g., equipment run hours) or calendar time, with maintenance intervals based on a statistical analysis of past performance and failures.
PdM, in contrast, uses ongoing measurements and metrics to calculate equipment health and the corresponding risk of failure. Its primary goal is to predict the need for maintenance, allowing it to be scheduled at convenient times.
Other potential benefits include:
- Reduced maintenance cost
- Lower frequency of failures
- Less unplanned downtime
- Reduced quality risk
- Opportunity to improve planning, personnel scheduling, and parts availability as unnecessary activities are eliminated.
PdM is an ongoing process that involves:
- Data collection and analysis
-Root cause analysis of any identified problems
- Scheduling of maintenance activities as required
PdM relies on data, which must be appropriate for its intended purpose. One way to obtain reliable data on facility and equipment performance is to establish a robust calibration program that identifies the instruments and controls that require calibration—including the frequency, range, accuracy, and precision of those instruments.
In GMP facilities, instruments and controls that produce data needed for GMP purposes also become part of the calibration program. Remaining instruments are usually not calibrated, except for the calibration provided by the vendor when first installed.
A robust PdM calibration program should identify and appropriately calibrate those instruments and controls that monitor facility and equipment performance
In this era of big data analytics, specialized software is increasingly used to detect emerging problems and is becoming part of root cause analysis.
Automated tools used to trend and analyze data for GMP decision-making must be validated.
For aging facilities, PdM can play an important role in risk management and also contribute to the cost savings.
Given a good understanding of risks that may affect a facility’s ability to perform its missions and the costs of existing preventive maintenance programs, PdM can reduce risks and maintenance costs.
Machine Learning (ML)
Industries are currently going through “The Fourth Industrial Revolution,” as professionals have called it, a term also known as “Industry 4.0.” (I4.0)
Integration amongst physical and digital systems of the production contexts is what mainly concerns Industry 4.0.
With the appearance of I4.0, the concept of prognostics and health management (PHM) has become unavoidable tendency in the framework of industrial big data and smart manufacturing; plus, at the same time, it offers a reliable solution for handling the industrial equipment health status.
Based on the collected data type machine learning algorithms can be applied for automated fault detection and diagnosis.
The industrial equipment predictive maintenance (PdM) can perceive the degradation performance because it was designed to achieve near-zero; hidden dangers, failures, pollution, and near-zero accidents in the entire environment of manufacturing processes.
These huge amounts of data collected for ML contains very useful information and valuable knowledge which can improve the whole productivity of manufacturing processes and system dynamics, and can also be applied into decision support in several areas, mainly in condition-based maintenance and health monitoring.
ML applications provide some advantages which include maintenance cost reduction, repair stop reduction, machine fault reduction, spare-part life increases and inventory reduction, operator safety enhancement, increased production, repair verification, an increase in overall profit, and many more.
Moreover, fault detection is one of the critical components of predictive maintenance; it is very much needed for industries to detect faults at very early stage.
Techniques for maintenance policies can be categorized into the following main classifications:
- Run 2 Failure (R2F): also known as corrective maintenance or unplanned maintenance. It is the simplest amongst maintenance techniques which is performed only when the equipment has failed.
- Preventive Maintenance (PvM): also known as scheduled maintenance or time-based maintenance (TBM). It sometimes leads to unnecessary maintenance which increase the operating costs. The main aim here is to improve the efficiency of the equipment by minimizing the failures in production.
- Condition based Maintenance (hCBM): The maintenance actions can only be carried out when the actions on the process are taken after one or more conditions of degradation of the process. CBM usually cannot be planned in advance.
- hPdM: known as Statistical based maintenance: maintenance schedules are only taken when needed. It utilizes prediction tools to measure when such maintenance actions are necessary, hence the maintenance can be scheduled.
PdM and Machine Learning (ML) Techniques
The aim of PdM is not only to collect process data and its parameters, but also to collect the health aspects of the equipment, machine, or component (such as pressure, vibration, viscosity, acoustics, viscosity, flow rate data) and many as such.
At the same time, this collected is now widely used for fault identification, early fault detection, equipment health assessment, and predicting the future state of the equipment.
ML is a subcategory of AI and is defined as any algorithm or a program that has the ability of learning with the smallest or no additional support. ML assists in solving many difficulties such as in vision, big data, robotics, and speech recognition.
Moreover, ML techniques are designed to derive knowledge out of existing.
ML algorithm are classified in three categories:
- Supervised
- Unsupervised
- Reinforcement learning (RL)
In unsupervised machine learning, there is no feedback from an external teacher or knowledgeable expert.
Based on the existing data, the algorithm identifies the clusters. The main aim here in supervised learning is determining the unknown classes of items by clustering, whereas classification is for supervised learning.
Unsupervised ML basically defines any ML method that attempts to learn structure in the absence of either an identified output (like supervised ML) or feedback (like Reinforcement learning (RL)).
RL is characterized by the delivery of information on training to the community. Through RL, the learner must discover which actions produce the greatest outcomes (numeric reinforcement signal) by attempting instead of being told.
Applications of ML Algorithms in PdM
ML algorithms can be used to solve several problems with the enormously available data generated from industries, thus, ML has been widely used in computer science and other areas, such as PdM of manufacturing system, tool or machine, and is one of the possible areas of use for data-driven methods (Artificial Neural Network (ANN), Reinforcement Learning (RF), Support Vector Machine (SVM), Logistics Regression (LR), and DecisionTree (DT))
Below, remarks for further researches:
- Extraction of real time data using intelligent data acquisition system can help to automate predictive maintenance.
- Combination of more than one ML models can provide better prediction compared to use of individual model.
Classification and Anomaly Detection algorithms can be combined to maintain precision of classification models without losing Anomaly detection advantages. By this way, PM can be applied to equipment or system which does not have large dataset.
Reference: Sustainability 2020, 12, 8211
Read also:
- Good Practices for Research and Development (R&D) Facilities
- Artificial Intelegence and Machine Learning in Drug Development
Resource Person: BARBARA PIROLA