We’ve detected that you are using an outdated browser. This will prevent you from accessing certain features. Update browser

Combining operational AI and risk to make smarter decisions.

A significant part of achieving operational excellence for any business is mitigating threats to employee safety, environment, product or share price, to name a few. Risk should not be treated as a static concept. It changes constantly as operation conditions do, however, given industry statistics on downtime, emissions leaks and safety incidents, it’s clear that more could be done to identify and mitigate risk ahead of time. We estimate that over a trillion dollars is lost in the process industry alone every year by unplanned downtime, while 40% of methane emissions are as a result of leaks.

Asset integrity and performance sits at the very heart of improving these aspects of operational excellence. While most operators monitor equipment behaviour through operational data (MES etc.), less are able to gain deep and actionable insights from it that would allow them to minimise risk and improve overall performance.

As such, there is a substantial opportunity for operators to use artificial intelligence (AI), real-time digital twin technologies and modern asset performance management techniques to support operational excellence through a better understanding of the impact critical assets have on safety, environment and the bottom-line. However, to do so, they must first bridge the gap between operational technology and operational management.

Combining operational AI and risk to make smarter decisions

Lloyd’s Register has been developing a solution that empowers operators to make smarter decisions in pursuit of operational excellence. In collaboration with Falkonry Clue, the solution combines operational awareness from real-time data allowing users to predict possible issues based on operational conditions, with the ability to prioritise risk and prescribe the right activities for the corresponding failure modes using Lloyd’s Register’s AllAssets. Coupling operational AI with risk gives operators real-time visibility and a depth of insight into their operations that will not only have an immediate positive impact but will also create longer-term benefits such as reduced downtime, improved sustainability, and enhanced management of risk and safety.

It is thought that siloed data can cost companies as much as 20% in lost revenue every year. To address this, AllAssets creates a centralised point for an operator’s engineering data, using its integration capabilities to break down data siloes and brings information from other data sources to consolidate those in a single view. Asset risk is then qualified using our decision models. This produces a series of metrics which help operators to prioritise and optimise their maintenance activities based on availability, reliability and commercial impact. AllAssets aggregates over 20 years of digital engineering content as well as feedback from its deployment across millions of assets to recommend the best maintenance and mitigation activities for the equipment based on static data.

Falkonry Clue connects with real-time operational data (from a data historian, for example). Its automated condition discovery function creates a digital fingerprint of each asset to monitor its health and productivity in real-time. This allows it to learn the correct behaviour of the equipment and alert on any deviations. After the behaviour has been categorised as a possible issue, or a normal condition, the engine will utilise this knowledge to improve its predictive capabilities.

So, what happens when a possible failure has been detected? Well, that is when AllAssets comes into play: linking pattern recognition technology with digital engineering content means to go from early detection of possible failures to prescriptive actions to mitigate the equipment’s risk. Aligned with corresponding failure modes, it provides the necessary insights to act swiftly on the issues and supplies a potential root cause explanation to provide context. Through a process of continuous learning, a model is built to highlight when the algorithm might suggest the incident will happen again, known as an event horizon estimation.

Introducing operational AI in this way creates a predictive digital twin and provides the operator with a seamless view of risk and real-time conditions based on operational data. From it, potential impacts to reliability and safety can be easily inferred at every level helping all departments of the business strive for operational excellence.

Combining operational AI and risk to make smarter decisions

In summary, the solution allows operators to be much more dynamic in the way that they approach risk and performance simultaneously, providing an opportunity to understand risks in that moment, and also how they may evolve over time. By bringing together operational and risk data, analysing it in context with the process and/or plant and then creating actionable insights, the solution offers a holistic, end-to-end approach that can positively impact operational excellence. The solution is designed as such that the operational management level will also have much greater visibility and understanding of risk, allowing them to better inform long-term planning and impacts on changes on operational conditions, which is key for planning and forecasting.

Ultimately, through combining risk and operational AI, operators can make faster, more accurate and more confident decisions helping them to achieve a wide range of safety, reliability, sustainability and economic goals in pursuit of operational excellence.

INSIGHTS

What we think

LR's experts regularly share their research and insights.

Need to get in touch?

Hit enter or the arrow to search Hit enter to search

Search icon

Are you looking for?