Future of Oil and Gas Supply Chain With AI

Multiple advances in technology over the past decade have unlocked digital solutions that can improve growth and efficiency of the Oil and Gas value chain. Augmented Artificial Intelligence has the power to further transform businesses and enable large improvements in performance metrics. Tech Mahindra’s AI based innovative solutions are focused on improving production while plugging leakages at every stage; for the benefit of the entire Oil and Gas ecosystem.

• It’s difficult for the upstream sector to collate the huge volume, variety and high frequency of data generated in a usable form
• Present refinery planning process contingency does not take historical market data, events, and predictive models of plant condition into account, which leads to a huge gap between planned and actual GRM
• Excel-based manual monitoring does not predict system performance degradation from high energy cost due to fouling
• Optimisation of revenue is a challenge with traditional pricing mechanism which lacks consumer centric data

AI solutions for the Oil and Gas Value chain

• Drilling performance optimization – Multivariate, Multidimensional, Stochastic analytics
• NPT Reduction: ROP Optimization using well-to-well correlations
• Well Advisor based on NLP
• Well Cost Optimizer
• Predicting the ‘Time to Failure’ of the Progressive Cavity Pump
• Early warning system for hurricane and evacuation of offshore crew
• Digitization of well log
• Drones for offshore platform inspection
• Indexing seismic data files and seismic section images for improving search ability

• Neural Networks & Monte Carlo Simulation to move Refinery Planning from deterministic GRM to probabilistic GRM
• Maximum Possible Production– Refinery and Petchem
• Heat Exchanger Fouling Prediction – Example Crude Preheater
• Defining the Operating envelope of Cracked gas Compressor / Centrifugal Compressor failure prediction - Cracked Gas Compressor for Ethylene Plant
• Crude Assay Prediction using AI
• Ethylene Furnace Yield Optimization
• Desalter Efficiency Improvement
• Image / Video Analytics for Unsafe acts and practices

• Retail Marketing Analytics
• Retail planning and demand management
• Retail Dynamic Pricing
• Churn Analysis
• Affinity Analysis

• Lubricant Traceability & Product Authentication
• Additives Optimization AI Tool
• AI-enabled Lubes Scheduling
• Lubricant business model disruption from Product sale to Services
• AI-enabled Social Media Analytics

Solutions

While the upstream sector generates huge volume, variety and high frequency of data about production, it was difficult to factor in such data to optimise performance. Our model for drilling performance optimisation leverages the power of AI to reduce NPT and increase production from optimal well placement, through efficient and standardised reproduction of the scattered data.

The present refinery planning process does not input relevant historical market data, which presents a huge gap in planned and actual GRM. Our AI first solution helps take into account relevant factors like contingency, events and refinery condition for input to Aspen PIMS (trusted planning software solution for optimizing operations). This helps refineries achieve a probabilistic GRM with confidence level and predict cash flow better.

Heat exchanger fouling increases energy cost and tube failure risk while dependency on manual monitoring hampers availability. Our AI based fouling prediction model records heat exchanger cleaning records and other variables to minimise energy loss and product slippage through timely alerts. The intervention also helps improve availability.

Traditional pricing mechanism lacks a customer centric approach and ignores applicable consumer buying behaviour. Our Retail Dynamic Pricing Model uses AI to take demand into the model and compute optimal fuel cost for retail. This presents the customer with an incentive to save on fuel bills while businesses get insights to plan for the future and increase profits.

Benefits

• Optimise NPT by 8-10%
• Optimise drilling cost variance by 5-10%
• Increase production with optimal well placement

• Leverage contingency, past pattern and refinery condition for input to Aspen PIMS
• Gain probabilistic GRM
• Help predict cash flow better
• Improve GRM potential

• Improve availability by intervention in advance
• Minimise energy cost in refinery by about 30%
• Avoid tube ruptures from localised heating by timely intervention
• Reduce product slippage/loss

• Optimise revenue by 10-12%
• Reduce cost of operation and improve profitability by 1-2%
• Customer can save on fuel bills and added incentives
• Business gains insights to plan ahead