Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource PHD PhD PhD Thesis Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Relation A related resource 61000427 Title A name given to the resource Pharmaceutical Tablet Uniformity Prediction Using Spectroscopy-Based Data Fusion and Machine Learning Approaches Subject The topic of the resource Data Science Description An account of the resource The pharmaceutical industry is highly regulated, and every manufacturer must demonstrate the drug product's quality, safety, and efficacy before market release. Quality control plays a vital role in ensuring drug products' consistency, purity, and potency through rigorous testing of raw materials in the process and the finished stages of manufacturing. Quality risk management and process understanding are critical to maintaining quality throughout manufacturing. Quality by Design (QbD) offers a structured approach, while Process Analytical Technology (PAT) facilitates real-time monitoring to control risk of product quality. Process analyzers, multivariate methods, process control, and continuous improvement tools are part of the PAT framework that enhances process understanding and aids risk mitigation strategies. Near-infrared (NIR) spectroscopy is a commonly used analytical technique in PAT environments for both qualitative and quantitative measurements; these are real-time and non-destructive process analyzers. Chemometrics helps extract information from this chemical data using mathematical and statistical methods. With the advent of Industry 4.0, machine learning models have gained popularity in spectroscopy due to their ability to handle complex, high-dimensional data and adapt to various applications. This research introduces a systematic approach to implementing machine learning models as an alternative to traditional chemical testing in predicting the content uniformity of pharmaceutical tablets. The objective is to improve the quality of data analysis and its predictive performance. This study also outlines the importance of using manufacturing information as stratified variables in predictive modeling and spectroscopic data, or sensor fusion data. To demonstrate the effectiveness of this approach, a real-world NIR dataset developed based on various characteristics such as manufacturing scale, tablet strength, dose proportion, and coating is utilized. This real-world application of the research makes the content more relatable and interesting to the reader. This allows the seamless use of the model across different known environments as the model is trained using sensor data fusion. A comparison of Partial Least Squares regression models and machine learning Neural Network models is evaluated for the model predictability. The work also delves into selecting and optimizing appropriate hyperparameters for a chosen optimal model. It explores the impact of model performance to ensure successful implementation in the production environment and discusses various approaches in monitoring during life cycle management. Creator An entity primarily responsible for making the resource M, Hussain Ali Source A related resource from which the described resource is derived Author's Submission Publisher An entity responsible for making the resource available Christ(Deemed to be University) Date A point or period of time associated with an event in the lifecycle of the resource 2024-01-01 Contributor An entity responsible for making contributions to the resource Gobi, R. Rights Information about rights held in and over the resource Open Access Format The file format, physical medium, or dimensions of the resource PDF Language A language of the resource English Type The nature or genre of the resource PhD Identifier An unambiguous reference to the resource within a given context <a href="http://hdl.handle.net/10603/615728" target="_blank" rel="noreferrer noopener">http://hdl.handle.net/10603/615728</a>