Making pharmaceutical manufacturing data ready for AI

Naz, Tabbasum and Johnston, Blair and Robertson, Murray and Vassileiou, Antony and Bailes, Sophie and Dawson, Neil and Zomer, Simone and Lai, Tiffany and Findlay, Rachel and Reynolds, Gavin and Robertson, Amy (2022) Making pharmaceutical manufacturing data ready for AI. In: CMAC Annual Open Day 2022, 2022-05-16 - 2022-05-18.

[thumbnail of Naz-etal-CMAC-2022-Making-pharmaceutical-manufacturing-data-ready-for-AI]
Preview
Text. Filename: Naz_etal_CMAC_2022_Making_pharmaceutical_manufacturing_data_ready_for_AI.pdf
Final Published Version
License: Strathprints license 1.0

Download (1MB)| Preview

Abstract

Large volumes of pharmaceutical manufacturing data have been generated in recent years. A lot of time and effort has been spent producing data but they are, for the most part, scattered, unstructured, not machine readable and in heterogeneous formats. The work presented here provides integrated management and access to these valuable datasets. The Digital Design Accelerator Platform (DDAP) Extract- Transform-Load (ETL) tool has been developed to derive maximum value from the data acquisition effort to date and to allow future data to be integrated easily. DDAP ETL with multiple components can be used for automatic extraction, transformation and loading of heterogeneous pharmaceutical manufacturing data from multiple instruments. It is a collaborative effort to digitalise and make data Findable, Accessible, Interoperable and Reusable (FAIR). It also provides an opportunity to explore semantic heterogeneity across partners for standardisation efforts and ontology development in the medicine manufacturing domain. DDAP ETL can help domain experts to reap the benefits of the digital age and extract more value from organised data. It provides a foundation for future analytics and data-driven projects across the sector. In future AI, predictive analysis, statistical analysis, data visualization, data mining and machine learning techniques can be applied on the extracted data.

Persistent Identifier

https://doi.org/10.17868/strath.00081662