Case Study
7% increased efficiency by using ML to optimize the purchase order process for a Fortune Global 500 chemical company
- Industry: Chemicals​
A Fortune Global 500, one of the largest chemical companies in the world. This $47 billion multinational chemical company employs around 19,000 people and is a global leader in innovation, consistently developing high-quality chemicals, polymers, fuels, and technologies.Â
Challenge
A leading chemical company needed a new way to analyze and audit vendor invoices in order to make better cost-saving decisions in procurement.
Invoices were saved as images in SAP; purchase order lines were in free-form text.
- Lack of analysed data and missed opportunities
- Time consuming
- Error prone
Solution
We developed an ML & ICR-powered solution to extract information from diverse vendor invoice images, built an integrated data model, and provided actionable intelligence to optimize free form text (FFT) PO process.
Tech stack
Outcome
7%
Savings across undefined spend category.
50%
Workforce efficiency gains.
80%
Model accuracy.
Sustained savings driven by actionable intelligence.