Objective:

One of the esteemed Oil and Gas companies of India was unable to properly monitor and reconcile product losses during tanker loading operations of Motor Spirit (Gasoline), HSD (High-Speed Diesel), and HFHSD (High Flash High-Speed Diesel) stored in its tank farm. The company possessed abundant data of products and processes; however, the lack of a competent analytical tool to process the large volumes of data, was resulting in the absence of measures for countering the losses. 

Solution: 

Eugenie’s AI-based anomaly detection framework analyzed the diverse product volume data of the affected fuels. The data collected during the unloading of the fuels were scrutinized with Eugenie. The discrepancies in the product volumes were segregated into technical and non-technical determinants. 

With Eugenie’s machine learning algorithms, the following technical determinants were calculated on a real-time basis 

• Evaporation
• Density variation
• Ambient & fluid temperature variation
• Rainwater influx in floating roof tank
• Bottom sediment & Water (BS&W)
• Tank ID profile

Impact:

Eugenie’s digital ecosystem offered digital twins of the storage tanks that presented a holistic view of operational monitoring. With Eugenie’s easy-to-use control panel, operations teams of the company were able to track, diagnose, and act on any discrepancies in real-time. 

Eugenie’s accurate insights resulted in faster actions, thereby reducing data-to-action time from several days to a few hours. Data-based predictive and preventive insights of Eugenie enabled the operations teams to discard the manual data processing. Eugenie's operational intelligence led to greatly reducing costs associated with the fuel losses.