Executive Summary : | The biopharmaceutical industry is gradually embracing Industry 4.0 by incorporating advanced computational technologies such as AI-ML, deep learning, Arduino, and IoT. However, the complex nature of biomanufacturing processes poses challenges for achieving Industry 4.0 implementation. This project aims to address these challenges by leveraging continuous biomanufacturing and process analytical technology (PAT) coupled with artificial intelligence (AI) and deep learning (DL) tools. The project objectives include acquiring data from various spectroscopic techniques, integrating the data using a Kalman filter, designing and training a denoising autoencoder for data pre-processing, monitoring and controlling Critical Quality Attributes (CQAs), implementing PAT tools for process optimization and control, and evaluating the performance of the regression model and PAT tools. The methodology involves data acquisition and integration, autoencoder-based pre-processing, CQA monitoring and control, process optimization using PAT tools, integration of the regression model with PAT, and evaluation and analysis. The expected deliverables include an integrated spectroscopy dataset, a trained autoencoder model, a one-dimensional CNN regression model, PAT tools for process optimization and control, a comprehensive project report, and documentation for practical implementation. This project aims to enhance accuracy, process control, and monitoring in biopharmaceutical manufacturing, contributing to the digitalization of the industry. |