Tag Archives: Augury

Advent of Machine Learning Into Predictive Maintenance

Over the years, numerous analytical and empirical methods have been developed to prevent machine failure, reduce costs, and increase production capacity. In the present age, companies like M2M Data Corporation and Senseye started out by developing physical models of machines based on data collected from their customers’ facilities. The data consists of parameters such as pressure, speed of motors, particle concentration in lubricants, acoustic data, temperature in the friction generating part of the equipment, and machine unique data sheets. For example, each pump has its unique “pump curves” that correlate pump rotation speed with discharge pressure and flow. A deviation of these above parameters is used to flag the plant personnel of possible machine issues, and could also suggest a specific malfunction. Continue reading

Sensing Opportunities in Predictive Maintenance Technology Innovation

We have expressed previously that predictive maintenance (PdM) is one of the most promising areas of the industrial internet of things (IIoT) and that there are a number of startups developing innovative sensor-based and/or software solutions. To identify where these technology developers are innovating, we took a look at the patent activity on PdM. During the past 20 years, there have been approximately 4,400 patents focused on PdM applications, and companies are filing more patent applications every year to differentiate themselves and protect their models, algorithms, or hardware-based solutions. In our “Predictive Maintenance: The Art of Uptime” report, we mentioned that innovations in connected sensing technologies and analytics are driving better operations, enabling users to gather and process real-time data on machine health to decrease downtime. For example, a study related to the oil and gas (O&G) industry revealed that performing PdM using a data-driven approach with sensors experienced 36% less unplanned downtime. There is still significant space for companies to decrease their unplanned downtime, and one method to do that is by adding more data streams from sensors. Therefore, we filtered our search to focus only on developments in PdM using sensing technologies, and we can see in Figure 1 that the IP space has been increasingly more active since 2014. Nonetheless, the patents aiming to protect sensor-based PdM are roughly 30% of the total amount of IP seeking protection for PdM applications.

Continue reading