Strategy by End User to implement IoT in Process Industry in Oil & Gas Sector.

Pandemic has changed a lot our ways of doing business and work . Industry 4.0 was in vogue since past decade. A quite developments happened in IoT application, Automation, Smart manufacturing, Digital transformation due to inexpensive and fast Data computing.

Before Pandemic projects were slow as happens in adoption of any Technology .However now Ecosystem of Industry Post Pandemic are making End users and other Stakeholders to rethink about approach and strategy for implementation of new Technology at shop floor.

Following developments could be noticed by Technology providers and End users Post Pandemic.

1.Covid appropriate behavior demands less physical interaction and less number of people at workplace.

2.Remote work is being encouraged for observation, meetings, analysis and decision making.

3.Remote monitoring , OEM support, troubleshooting operation and maintenance, arranging spare-parts and expert advice is more IoT based using cloud computing.

4.End user is realizing importance of digital transformation and utility of Artificial intelligence and Machine learning in Asset Performance Management.

5.IT sector already has seen changing of habits of people in all walks of life in home delivery of product and services and mix of work from home and work from anywhere.

In such scenario what strategy should be adopted by Process industry in investing for automation and digital transformation?

Based on my experience of implementation of Artificial Intelligence -Machine Learning in Oil and Gas sector particularly few projects in Refineries and Oil Producing fields, following observation and challenges have been noticed remarkably.

1.Digital Transformation requires quality data from sound sensors with available Architecture.

2.Operation and Maintenance language is different for Data scientists and AI-ML Engineers. OEMs have apprehension about data use from their Machines in predicting Machinery behavior.

3.Issues of Data Security and whether Edge computing or Cloud computing are main concerns of end User .

4.Extensive understanding is required among subject matter experts about Operation and maintenance practices to extract right data and features from different pain points.

5.Tremoundous benefits appear in application of machine learning as technology is already matured and proven by proof of concept success in Quality and Maintenance prediction practical cases.

6. A significant positive change has been noticed in adoption and credibility of new technologies in realization of benefits by some End Users.

Looking to above best strategy by end users would be to not wait for Proof of Concept. They can have Hybrid model of best of both from first principle based model and Machine Learning based Model keeping in mind user stories. They can adopt immediately and see ROI not in years but in months. They should monetize available data rather putting new sensors until and unless it is mandatory to getting relevant and quality Data for business to business case .

S.K.Singh

Oil and Gas Sector Specialist.