Sreekanth Rajagopalan
Sreekanth Rajagopalan
Associate Research Scientist at Dow Chemical Company
Sreekanth Rajagopalan is an Associate Research Scientist in the Machine Learning, Optimization, and Statistics team in Dow’s Research & Development organization. In his role, Sree develops purpose-built analytics solutions in the areas of manufacturing operations, supply chain, and sustainability with a focus on systems thinking and enterprise-wide optimization. Prior to joining Dow, Sree obtained a PhD in Chemical Engineering from Carnegie Mellon University in 2018 with a background in stochastic programming and global optimization. Sree is passionate about problem solving using math modeling and software development. He is active in peer-to-peer learning forums and governance teams on MLOps and DevOps for Data Science at Dow.
Track: Manufacturing
An Enterprise Decision-making Platform for Managing Maintenance Turnarounds in Process Industries
Maintenance turnarounds of manufacturing facilities are planned shutdowns of the facilities that can be several weeks long. Besides enabling safety and regulatory compliance activities and asset renewal to be conducted, a turnaround provides a window to complete reliability improvement and capital projects before resuming production. The petroleum refining industry is estimated to spend $4 billion across 550 turnaround projects in 2022 in the US. In addition to direct spending on turnarounds, a facility incurs an indirect cost from lost revenue during the shutdown time that can further be amplified in a dynamic market. For a company the size of Dow in terms of asset base and integration, turnaround planning is a complex problem offering several levers and trade-offs. In this talk, we will present how Dow has developed an ecosystem of mixed-integer linear programming models over the years to enable data-driven decision-making for planning turnarounds across the enterprise.
