Certificate of Analysis Automation

We receive Ceritificate of Analysis (COA) on a daily basis for all raw materials we receive from suppliers both locally and internationally. These COAs comes in various formats, benchmark results criteria and layout - they're vital in quality assurance process to ensure raw materials meet international standards and internal control criteria before acceptance is performed. Currently, these documents are either received via email or physical copies and are analysed by QA team manually to detect out of specifications and anomalies for quality rejection process. This is painstakingly difficult due to the vast amount of lab COA formats and benchmark results for a large portoflio of raw materials from commodities like flour to small batches of flavouring additives.

Fully automate incoming COA reading via OCR with AI capabilities to detect anomaly by learning benchmark results of each raw material groups or test results category. With these data, to also predict non-conformace rate or performance of a supplier based on historical trend and propose next steps for QA team to act on. Example: if Supplier A always fails on Microbes test, to automatically focus on this test result for next incoming COA.

Currently entire workflow is done manually without any system in place. SAP comes at the tail end for usage decision and stock release only.

Currently entire workflow is done manually without any system in place. SAP comes at the tail end for usage decision and stock release only.

Quality assurance team, incoming warehouse receiving team, R&D team

Unstructured COAs in PDFs, emails, and hardcopies occasionally for imported raw materials.

1. Cloud platform preferred 2. Immediate integration to analytics platform for analysis (if platform doesn't have own capabilities) 3. Potential future integration to SAP S/4HANA for usage decision