Are conflicts between operations and maintenance still the norm in your organization? A quick poll of the audience attending the Maintenance & Operations: Better Together session at the 2019 ARC Industry Forum in Orlando, Florida, indicated that the answer is YES. Operations are focused on short-term production goals while maintenance takes a longer-term view to help ensure asset reliability. This often leads to conflict between the two groups. But new IIoT-enabled tools can help improve collaboration between the two in pursuit of common organizational goals.
Human and Machine Learning at Covestro
Covestro, a global provider of premium polymers, is undertaking a digital transformation across the entire company. According to Jane Arnold, Head of Global Process Control Technology, the focus at Covestro is to embrace a digital culture. A digital innovator, Covestro is exploring digitalization in operations, customer experience, and business models. For example, the company relies on fully integrated digital communication to strengthen its collaboration with its customers. Bidding chemicals on the Alibaba B2B sales platform, as Covestro currently does, is a vast departure from the company’s traditional go-to-market model.
Covestro is using digitalization to increase efficiencies. Production tar-gets include no unplanned shutdowns due to process/equipment failures, optimized scoping for turnarounds, and improved standard operation. At technology centers, the goal is easy access to data, efficient monitoring, and continuous evaluation of process and equipment conditions.
Maintenance efficiency focuses on getting the right information at the right time for maintenance and outage planning purposes. Another focus is predictive process and self-service analytics with greater reliance on machine learning for predictive maintenance. Covestro’s culture embraces humans and machines working together. The ultimate goal is to have machines make recommendations and the humans do final interpretation based on the information available.
Ms. Arnold shared that a Covestro plant suffered major corrosion dam-age in July 2018 due to HCL formation caused by water leaking from a steam generator water side to the process side. After significant dam-age to critical equipment and a 20-day outage, Covestro invited the predictive analytics team at AVEVA to perform a retrospective case study. The team used Prism software on historical data seeking possible early warning signs prior to the major failure. The study proved that 10 days prior to the event, something had happened. Had the data model been available to view the spikes, the problem would have been recognized and acted upon quickly to minimize the damage and lost production time. By sharing such information, operations and maintenance are linked to create a more sustainable path forward. Correlating what the operator sees on his screen with an overall model based on years of operating history enables a better decision.
Nova Scotia Power Takes It Further
According to Rob MacNeil, Senior Technical Advisor, at Nova Scotia Power (NSP) the changing landscape of generation is driving NSP’s strategic goals and objectives. In such a dynamic environment, effectively managing, maintaining, and sustaining generating assets through to retirement without stranding investment is the company’s greatest challenge. NSP’s holistic asset management program takes a long-term view. This encompasses operations and maintenance as well as planning and engineering, because these activities are interrelated and linked. Maintenance strategy is a significant piece of NSP’s asset management aspects as it considers all areas of work that contribute to the health and care of its assets. Predix Asset Performance Management from GE Digital touches every aspect of NSP’s asset management pyramid, enabling the company to optimize its maintenance strategies as well as other resources, such as labor.
Technology alone does not drive NSP’s strategy, but rather something it deploys to support broader goals and objectives. NSP is currently examining digital approaches that can displace traditional maintenance and surveillance activities. Moving to digital strategies to obtain earlier detection of impending problems enables better planning and resolution before problems can advance in the failure mechanism, according to Mr. MacNeil. He also noted that, once deployed, digital strategies reduce operations costs and could also mean moving dollars from OpEx to CapEx, which is important for a utility such as NSP.
ARC Advisory Group clients can view the complete report at ARC Client Portal
If you would like to buy this report or obtain information about how to become a client, please Contact Us
Keywords: Asset Performance Management (APM), Machine Learning, Predictive Maintenance, Operations, ARC Advisory Group.