Fertilizer producer sees value of Cordant™ Machine Health
The Challenge
With a mature culture of reliability, the company had a sound program for monitoring machine health. However, the program was reliant on walkarounds using portable devices and the company wanted to explore the use of emerging technologies to enhance the program and improve outcomes. The company specifically sought to harness AI and Machine Learning to uncover hidden issues within its data and free up personnel to focus on value-added activities, ultimately enabling more proactive maintenance strategies. The company also recognized the opportunity to integrate its machine health data and process diagnostics data for a more holistic view of assets.
The Solution
Baker Hughes worked with the company to better understand its challenges and build out a business case for Cordant™ Machine Health. This involved supporting the selection of which machines to be included in the initial scope based on the company’s short- and long-term goals. The company subsequently deployed Cordant™ Machine Health and Ranger Pro™ wireless condition monitoring sensors on 25 of its critical machines, including pumps and blowers.
Early Detection
In the first two and a half months of operation, the solution identified a change in operating condition for a blower. While the change was subtle and may have gone unnoticed at that point with traditional asset health monitoring methods, the machine’s condition in Cordant™ Machine Health shifted from ‘acceptable’ to a ‘monitor’ status. The early detection initiated an investigation by the facility team, who were able to let the blower run until a more detailed inspection could be scheduled at a convenient time. The planned inspection revealed that three of the blower's large, 8-foot-diameter propeller blades had developed small cracks. The AI's ability to detect this anomaly before it escalated enabled the company to avoid a full overhaul of the machine.
Results
This single early-detection event helped the company prove the value of its investment with estimated savings of 40 hours of downtime and $130,000 in maintenance costs. These estimates are based on a previous overhaul of the machine and do not include costs such as spare parts replacement. The savings related to this event account for almost 83% of the total $170.1K saved by the company in the first five months of deployment. The savings also contributed to a 117% ROI achieved in just 13.8 weeks. These results have led the company to scale its deployment to encompass 59 machines.