Current trends: According to sources, for an average factory, inefficient maintenance policies are responsible for costs ranging from 5 to 20% of the plant’s entire productive capacity and the overall burden of unplanned downtime on industrial manufacturers across all industry segments is estimated to touch the impressive figure of $647 billion per year. Also, many companies are still not satisfied with their existing maintenance strategies. On the other hand, Predictive Maintenance (PM) is the process of modernization of the industrial world induced by the advent of the digitalization era. The health state of a machine is now constantly monitored by a network of sensors and future maintenance operations are based on the analysis of the resulting data. An increasing number of organizations, motivated by their need for reducing costs and by unlocking the potential of PM, are starting to invest significant amounts of resources on the modernization of their current maintenance strategies. Roughly 60% of the involved companies report average improvements of more than 9% of machines uptime, and further enhancements in terms of cost savings, health risks, assets lifetime.
Future Ahead: One’s deep understanding about the machine will help understand the minute spares parts and with the increasing availability of data and the potential of Artificial Intelligence and Machine Learning, one can predict, diagnose, and perform prognostic health monitoring to achieve the highest efficient machinery with minimal downtime. ML is possibly one of the technologies that is experiencing the extraordinary growth in terms of investments and interest of the industries. Prognostic and Health Management (PHM) is an engineering field whose goal is to provide users with a thorough analysis of the health condition of a machine and its components. PHM exploits tools from data science, statistics and physics in order to detect an anomaly in the system, classify it according to its specific type (diagnostic) and forecast how long the machine will be able to work in presence of this fault (prognostic). Also, efficient Prognostics decreases the probability of extreme failure events, thus improving the health of the machinery and reducing the downtime if any. Also, they drastically reduce the often-prominent costs associated with scheduled maintenance operations.
The Process Involved: PHM makes use of information extracted from data to assess the health state of an industrial component and driving maintenance operations accordingly. The very first step of the PHM process consists of selecting a suitable set of sensors and devices, setting them up in the most appropriate location and deciding on an optimal sampling frequency for data collection. Once the sensor array is in place, data can be acquired, but data are typically in forms that are not compatible with the input shape requested by AI algorithms. Therefore, a data pre-processing step must be implemented in order to clean the data, mitigate the effects induced by noise or simply reshape them so that their new format can be interpreted by data analysis techniques. The resulting data are cleaner than the original ones but can still contain a substantial amount of redundant information. This motivates the application of feature extraction techniques to reduce the dimensionality of the data and retain only the most meaningful pieces of information. Most modern AI techniques are designed to automatically extract informative features without any need for expert knowledge and manual feature engineering.
Condition-Based Maintenance: This consists of two main elements: anomaly detection and diagnosis. Both does the data extraction and data pre-processing which aims to support the decision-making with meaningful insights about the health of the system.
Anomaly Detection: Anomaly detection automatically establishes whether the input data is present or any discrepancy found from healthy machines. This is typically learned by extracting and storing internal features from data gathered.
Fault Diagnosis: Fault diagnosis moves one step forward with respect to anomaly detection since, besides detecting that an outlier is present, it also identifies the cause at the basis of that anomaly. Fault diagnosis models are based on historical data representing different faulty conditions.
Predicting Remaining Useful Life: The main difference between Condition Based Monitoring and Predictive Maintenance is that latter's algorithms deal with the problem of predicting the Remaining Useful Life (RUL) of an industrial component before breakdown of the machine.
