Digital Design for Crystallisation in Advanced Pharmaceutical Manufacturing: Uncertainty of information and how to find data

先进的制药方法，包括连续制药（C-PM）涉及复杂的操作，需要高度的过程控制，理解和经验。我们正在创新开发智能、人工智能驱动和机器人支持的开发平台，即“数据工厂”，以加速产品和流程开发，并实现数字孪生的开发和应用。CMAC的愿景是实现化学、制造和控制（CMC）业务的数字化转型，作为新药批准的一部分，其关键目标是开发结晶分类系统（CCS），该系统将允许从有针对性的、节省材料的实验方法中快速预测结晶结果。CMAC的重大投资建立了新的DataFactories，用于加速结晶开发，使用模型驱动实验设计和协作机器人技术来驱动初始实验，以估计热力学（溶解度）和动力学（成核，生长）参数以及将指导工艺设计的关键颗粒属性（固体形式，大小和形状）。该平台由多个传感器组成，具有离线测量和数据分析功能。目前，不确定性的测量及其在整个数字设计过程中的传播还没有得到很好的理解。在CCS DataFactory中的所有测量点对这些进行评估，将提高对过程的理解，确定改进路线，并建立对数字化工具的信心。目前，对于仪器的（Meta）数据的存储或这种大的异质数据的处理没有共同的标准。虽然控制软件工具试图捕获调节工艺所需的信息，但它们并不便于捕获来自工艺相关仪器的分析结果。因此，需要改进数据存储、治理和访问有关复杂多阶段过程的大量信息，以提高药物开发和制造的质量和效率，并充分发挥数字化转型的潜力。博士将专注于使用DataFactory获取全面的数据，并开发新的方法，用于通过新的结晶参数数据库传播信息和不确定性，该数据库是CMAC愿景的一部分，成为制药产品和过程的数据中心，并提供CCS。

Advanced pharmaceutical manufacturing approaches including continuous pharmaceutical manufacturing (C-PM) involve complex operations requiring high degrees of process control, understanding and experience. We are innovating the development of smart, AI-driven and robotically enabled development platforms, or 'DataFactories' to accelerate product and process development and enable the development and application of digital twins. CMAC's vision is to deliver the digital transformation of Chemistry, Manufacturing and Control (CMC) operations as part of the approval of new medicines and a key aim of this is the development of a Crystallisation Classification System (CCS) that will allow rapid prediction of crystallisation outcomes from targeted, material sparing experimental approaches. Significant investment by CMAC has established new DataFactories for accelerated crystallisation development using model driven experimental design and collaborative robotics to drive initial experiments to estimate thermodynamic (solubility) and kinetic (nucleation, growth) parameters as well as key particle attributes (solid form, size and shape) that will direct process design. The platform's comprise multiple sensors with offline measurements and data analysis to achieve this. Currently, the measurement of uncertainties and their propagation throughout the digital design process are not well understood. Evaluation of these at all measurement points in the CCS DataFactory will improve process understanding, identify routes for improvement and build confidence in digital tools. Presently, there are no common standards for the storage of (meta)data of instruments or the handling of such large heterogenous data. While control software tools attempt to capture the information required to regulate the processes ad-hoc, they do not facilitate capturing analytical results from process-related instruments. Thus, there is a need for improved data storage, governance and access to the myriad of information about complex mulit-phase processes to enhance the quality and efficiency of medicines development and manufacturing and realise the full potential of digital transformation. The PhD will focus on the acquisition of comprehensive data using the DataFactory and develop novel methods for the propagation of information and uncertainty delivered through the novel Crystallisation Parameter database being established as part of CMAC's vision to become the data centre for pharmaceutical products and processes and deliver the CCS.