Case Study: Venator improves pigment quality insights with TekLink’s Azure machine learning solution

Identifying Product Quality Drivers with Machine Learning

TekLink worked with Venator, a global manufacturer and marketer of chemical products, to explore how machine learning and AI could be used to collect, collate, and analyze IoT data from manufacturing sites. The goal was to identify the conditions that produce the highest quality pigment and spot outliers early enough for engineers to take corrective action.

TekLink implemented an MS Azure-based solution using Databricks, Azure Data Factory, and Power BI to build automated data pipelines, support self-service analysis, and develop machine learning models with XGBoost and Random Forest. The solution continuously consolidated sensor data in the cloud, improved understanding of the drivers of dispersion quality, and helped guide future process and data-collection improvements that can reduce production costs, increase yield, and improve product quality.

Venator