In the last blog, we reviewed the different types of data that maintenance departments have to work with. In this issue, we will look at technologies and tools available to get meaningful insight from all structured and unstructured data in the maintenance departments of a factory or a facility. 

For data such as vibration, temperature or current readings on a motor; there are many tools available to analyze this data and determine (or predict) a near term or long term failure. Predictive maintenance, or condition-based maintenance, is maintenance that monitors the performance and condition of equipment during normal operation to reduce the likelihood of failures. By using periodic (offline) or continuous (online) equipment condition monitoring, the ultimate goal is to perform maintenance at a scheduled point in time when the maintenance activity is most cost-effective and before the equipment loses performance within a threshold.

In order to predict these failures, it’s necessary to analyze the data that is generated by sensors, vibrations, CMMS, etc. Predictive data analytics is aimed at making predictions about future outcomes based on historical data and analytics techniques such as statistical modeling and machine learning.

Predictive maintenance relies on condition monitoring, which is the continuous monitoring of machines during process conditions in order to ensure the optimal use of machines. There are three facets of condition monitoring: online, periodic and remote.

Online condition monitoring is defined as the continuous monitoring of machines or production processes, with data collected on critical speeds and changing spindle positions. Periodic condition monitoring is achieved through vibration analysis. Remote condition monitoring allows equipment to be monitored from a remote location, with data transmitted for analysis.

The monitoring of your assets and processes through predictive analytics can 1) produce actionable responses through a CMMS program and 2) let you assess the condition of parts and the presence of defects previously impossible to detect.

With a CMMS program and monitoring system, you are able to automatically generate work orders based on the data collected and the triggered responses to that data. If a sensor picks up a defective reading on a bearing, for example, a work order can be generated and sent to the proper technician to check on the problem immediately.

There are also AI tools available, such as Proteus’ “Ask Steve” ChatBot, which allows you to converse with your maintenance data in real-time to discover patterns and trends in your data and asset condition.

Predictive analytics tools give users deep, real-time insights into an almost endless array of business activities. Tools can be used to predict various types of behavior and patterns, such as how to allocate resources at particular times, when to replenish stock or the best moment to replace a motor, basing predictions on an analysis of data collected over a period of time. 

Natural language processing (NLP) and conversational analytics

Just as search interfaces like Google made the internet accessible to everyday consumers, NLP gives maintenance people an easier way to ask questions about data and to receive an explanation of the insights. Conversational analytics takes the concept of NLP a step further by enabling such questions to be posed and answered verbally rather than through text.

According to Gartner, by 2021, NLP and conversational analytics will boost analytics and business intelligence adoption from 35% of employees to over 50%, including new classes of users, particularly front-office workers.

An AI (artificial intelligence) tool used with Eagle Technology’s CMMS program is called Ask Steve. Ask Steve works like a ChatBot, letting users “drive” through their own data history using a conversational approach, all in real-time and reproducible. It uses a controlled/engineered natural language approach in which a recommendation engine guides the user along the engineered language surface as they compose their dialogs. And as you use Ask Steve more, it learns more and can better anticipate your questions.

Unlike with standard reporting, there is no need to fully understand the data structure or tables involved in generating in-depth outcomes. In this way, Ask Steve can be used by executives, supervisors and technicians alike. It can also be used across multiple facilities or a single building.