Industrial Analytics in a Post-Pandemic world

Embracing AI Industrial Analytics leads to higher productivity, safer workplaces and asset longevity

• Organisations need to embrace technology and find a new method of working in a Post-Pandemic world
• We look at three types of industrial businesses: Data-adverse, Data-centric and Data-illiterate
• AI Industrial Analytics can be used to predict maintenance failures and optimise production by harnessing existing industrial data. Working smarter not harder.

As some economies slowly start to re-open following COVID-19 shut-downs, focus is shifting to how in this new post-pandemic world industries can work smarter by embracing technology.  Many experts are predicting that our way of working may never return to pre-COVID conditions, and things like remote working are here to stay.

As we look to smarter working we have to consider the increasing role technology can play, especially as we way up the predicted skills shortage and the ever increasing demand from our share holders to cut costs.

Embracing technology for many industries will however require a step-change to how they’ve typically operated, the re-engineering of process to allow technology to fill gaps and take the place of traditional labour-intensive methods.

Smarter does not necessarily mean harder, and smarter can be achieved through the application of analytics to industrial data.

Many companies will look to how industrial analytics, and in-particular predictive analytics can be used in their operations. Asset-heavy and reliant industries such as power, oil and gas & mining can benefit hugely from Predictive Analytics however many are still using out-dated costly and labour-intensive maintenance methodologies, such as Run-to-Failure and Preventative Maintenance. 

Typically, organisations sit in one of three categories when it comes to embracing their Big Data and harnessing Predictive Analytics

• Data-adverse
• Data-centric
• Data-illiterate

Data-adverse companies, are companies that choose not to embrace the data that they have from their equipment, possibly choosing to rely on more traditional methods of maintenance. This may come from a belief that the data isn’t available to them or of value to them. Data-adverse companies typically believe the technology is too expensive to apply.

Data-centric companies know all about their data, they possibly have an in-house data-science team and are trying to use their data to get insights into the root cause of failures and how to optimise their production.  Often these teams are drowning in ‘too much data’ with problem statements taking hundreds if not thousands of man-hours to solve.  These companies are already on their digital transformation journey with significant top-down support.

Data-illiterate companies are struggling to understand their data, not having the in-house experts to analyse and translate findings.  These companies know there is value in their data but are struggling to harness it. They are searching for the right platform and partner to help them on their digital journey.

Artificial Intelligence & Machine Learning technology has advanced significantly in the last 2-3years, assisting to bring about the 4^th Industrial Revolution.   These technologies are now being applied to Industrial Data giving companies the ability to get valuable real time insights from their data.

So what does this technology mean for our three types of businesses?

Data-adverse companies may be please to know two things. Firstly, the technology has advanced and costs to implement have reduced significantly, secondly nearly all plants and platforms are already producing sufficient amounts of data that there is no need for any changes to infrastructure. Yes, this is true even if your equipment is thirty years old!  See this example of a refinery that suffered from premature filter clogging for 20years.

Data-centric businesses will be pleased to hear that there are tools for your data science team that will solve 80% of the problems, leaving them to focus on the complex 20%. The tools multiply the efforts of the team and result in faster insights that can have real-time benefits. See this case study of an off-shore platform where analytics was conducted 2000x faster than traditional methods.

Data-illiterate organisations will find the Do It Yourself AI Modelling and customisable dashboards transforming. Maintenance and Reliability teams will able to use their data to obtain valuable insights to make informed decisions about their productivity and maintenance.  See this example of an FPSO that used AI Predictive Analytics to predict the first ever failure on a Produced Water Pump.

As industrial businesses move from reactive to proactive, preparing to operate in a post-pandemic world, it is critical that we challenge traditional methods of operating and look to embrace technology to bring about necessary cost savings.  Predictive Analytics is one of the technologies that all industrial companies, no matter how data-driven the organisation is, can implement to bring about real benefits to production levels, staff safety, asset longevity as well as on-going maintenance savings.

Read the latest issue of the OGV Energy magazine HERE.