Dow Builds Carbon Footprint Ledger on Databricks for Sustainability

The CFL provides two core capabilities: dynamic optimization of a product's carbon attributes by assigning the lowest-carbon inputs, and verifiable certification by issuing Product Carbon Footprint certificates. These capabilities are assured against ISO 14067 and the GHG Protocol Product Standard.

Dow's Enterprise Data & AI team built an end-to-end pipeline within their Integrated Data Hub (IDH) on Databricks, ingesting, transforming, and unifying data from across the organization. Unity Catalog provides governance for sensitive data, while Apache Spark handles large-scale data processing.

The implementation utilizes Apache Spark, Delta Lake, Unity Catalog, and MLflow to collapse end-to-end PCF processing time from weeks to a fraction of that. This includes full lineage and audit-grade governance, enabling rapid generation of PCF information for all Dow products.

An advanced optimization model, developed and deployed as a production-grade service using Databricks' MLOps approach with MLflow, dynamically identifies the lowest-greenhouse-gas production pathway for products across Dow's global manufacturing network.

Dow has developed a Carbon Footprint Ledger (CFL) on the Databricks Data Intelligence Platform to unify enterprise data and calculate cradle-to-gate Product Carbon Footprints (PCFs) for its entire product portfolio. This system aids in optimizing products for lower carbon impact and issuing verifiable certifications for low-carbon offerings.

The CFL on Databricks is a foundational step towards a vision where product carbon footprint calculations are automated, commercial teams can match sustainability content to customer needs, and external stakeholders receive sustainability reporting with audit assurance similar to financial reporting.