A web-based tool has provided a company with more than 85% accuracy when predicting the blend compatibility of different crudes.
The refining sector is going through a difficult period, with decreased demand for oil products and increased regulatory pressure. To stay competitive, companies need to embed digital capabilities in all aspects of their operations in order to improve efficiency, reduce costs, and protect revenues and margins.
Digital transformation presents a game-changing opportunity to improve yield productivity, asset reliability, and workforce effectiveness. Refiners need to sharpen their digital capabilities in three critical areas: analytics in production, field force effectiveness, and asset management.
Yield productivity can be improved in two important ways. First, advanced analytics with real-time characterization of assets, feedstock, and output can be used to define optimal plant settings. Second, machine learning in supply chain management can support portfolio programming and scheduling accuracy.
BCG helped a company optimize its production planning and test a web-based tool that predicts crude compatibility. The objective was to extract value by blending the selected crudes while avoiding asset-processing issues. The new insights are being used to guide campaign planning and optimize margins.
Companies can improve field force effectiveness through the use of wearable devices and digital tools, including augmented-reality systems. Wearable devices can also connect to an enhanced safety-control room to detect and mitigate critical risks, such as man down or gas leakages.
BCG worked with a leading refiner to adopt portable devices so it could simplify critical tasks and improve safety and field force productivity. For example, software allowed electricians in the field to check the status of equipment, visualize standard operating procedures and work instructions, and communicate the completion of tasks in real time to immediately trigger the next interventions.
Companies can improve asset reliability through advanced maintenance management, using big data and machine-learning algorithms to optimize maintenance cycles and predict critical equipment failures.
BCG helped a company develop a machine-learning algorithm to predict failure and optimize the maintenance cycle of a large conversion unit. The big data algorithm is able to predict a unit’s critical equipment fouling level. Machine-learning technology helps engineers identify important factors that can lead to future problems, allowing the company to avoid unplanned maintenance events.