David Schaub

Creating a centralized emissions data model can enhance transparency and comparability, but involves significant implementation challenges including data standardization, quality, migration, and CAPEX investment. In this session our panelists will debate if the benefits really outweigh the challenges. 

• While using AI and ML can help standardize diverse operational data sets, considering the implementation issues, do they really improve the efficiency of data migration? 
• Limited and conflicting data hinders the ability to measure emissions reduction impact. Does a centralized view provide the insights needed, or risk overlooking key areas? 
• Data-agnostic solutions will undoubtedly streamline the process of converting raw data into actionable insights, thereby boosting efficiency and profitability. But can these benefits be achieved when working on a facility-by-facility basis? 

Once data scientists and engineers have a proof of concept for utilizing AI to reduce emissions on an individual facility/ asset, and the value has been realized, the next logical step is to scale the approach. In this session Shell will explore how they scaled their use of AI across business units, considering both technical and operational complexities.  

• Understanding how a centralized and standardized data model allows for the efficient roll out of AI solutions across business units and teams  
• Explore how Shell has used modelling capabilities from proof of concepts to demonstrate the benefits of large-scale rollouts in reducing company-wide emissions 
• Different facilities, teams and operations have established specific data models, collection methods and ways of working. How does this affect the capabilities of AI and ML for producing accurate emission reduction insights?