Updated: Oct 9, 2020
More important than ever before…
The pressure to operate at the highest levels of efficiency while increasing productivity & lowering costs has never been higher. For professionals operating in remote environments, this has become challenging to achieve. Real-time visibility to plant
operations & process equipment in order to avoid costly unscheduled maintenance and reduce downtime is now the need of the hour across the industry.
Over the last decade, the significant increase of IIoT enabled sensors (generating vast amounts of data) & the use of machine learning-based predictive analytics has enabled companies to cut operational expenditure by optimizing maintenance schedules, predicting critical asset replacement & driving up productivity. (We are also hosting a webinar on the same topic on 29th April, 2020. More details at the bottom of the article)
Business case for critical assets
There are several hundred thousand critical assets (pumps, motors, compressors, turbines, etc.) deployed across the industry. Over 50% of them are not instrumented due to a fair majority being legacy equipment. The current manual reading based predictive maintenance techniques are tedious & are not able to keep up with the ground reality of the health of assets across the field/plant. For assets in remote or hazardous locations, the frequency of inspection is even lower.
The cost of a single asset’s failure ranges from a few $100,000 to a few $100million due to production downtime, repair and in the worst-case – accidents due to catastrophic failure. All of this can now be avoided with easy to integrate, reliable & affordable sensors that can wirelessly transmit data to the edge/cloud for machine learning processing, in turn providing plant personal with real time usable information like asset health indices, etc.
Predictive analytics for critical assets – Technology & solution providers
Our analysis at WhatNext tells us that while there are quite a number of sector agnostic solution providers, the number of predictive analytics technology providers catering to the oil & gas industry & specifically for critical assets is very low. There is immense potential for many other incumbents to enter this space given the vastness of assets to the sensorised and digitalised.
Existing solution providers include some of the large established players like BakerHughesC3.ai, OneStim from Schlumberger, GE’s Predix and Prism by Schneider & Aveva.
Other companies that have entered this space over the last eight years & made a mark in the industry for the solutions they’ve implemented include Spark Cognition, Flutura, Presenso (now acquired by ZF) & Petasense. Oil field service giants like Halliburton, Aker Solutions, Weatherford, and National Oilwell Varco also offer predictive maintenance technologies to monitor the equipment health and predict the failure in advance.
Implementation of predictive analytics for critical assets across the industry
In Nov’2019, ADNOC partnered with Honeywell to use Honeywell’s AI-powered asset monitoring and analytics platform to maximize asset efficiency and integrity across ADNOC’s upstream & downstream operations.
ExxonMobil partnered with Microsoft in Feb’19, to use its Microsoft Azure cloud computing platform & data analytics tools to deploy predictive maintenance technologies at Permian shale assets in west Texas and south-east New Mexico.
Shell has been using AI and machine learning in predictive maintenance to predict asset failures & reduced efficiencies for the past several years.
In April 2018, Total partnered with Google Cloud to develop AI-driven software for geophysical data analysis, besides delivering equipment monitoring capabilities.
In September 2018, Chevron adopted the cloud-based data analytics approach for predictive equipment failure in its refinery operations. Working with Microsoft, the company aims to install sensors on thousands of pieces of equipment by 2024, enabling them to predict exactly when equipment will need to be serviced.