Prescriptive Analytics

This project is ongoing. The technical materials for this project are confidential.

Background: Panel gap and flush defects are a significant contributor to rework in this company’s manufacturing process, leading to increased cost and reduced efficiency. Inconsistent measurements and delayed detection make it challenging to maintain quality at scale.

Aim: The goal is to reduce defect-related rework through the development of a prescriptive machine learning model that accurately predicts panel gap issues and improves measurement precision.

Action: We are developing a machine learning solution that analyzes real-time measurement data to detect patterns, raise alerts when gaps are consistently out of spec, and identify root causes. Dashboards are being created to help engineers monitor trends and receive prescriptive recommendations – not only on what adjustment to make, but also on which specific machine in the production line should be adjusted to correct the issue.

Result: This system will enable proactive quality control, reduce measurement errors, and empower engineers to take targeted corrective actions, cutting rework and improving manufacturing efficiency.