Autonomous MicroScale Manufacture of Active Pharmaceutical Ingredients (APIs)

该项目将通过创建独特的自主微型原料药制造和测试系统，专注于开发灵活的小规模生产设施。该系统将进行工艺开发，以生产具有所需关键质量属性(CQA)的稳定活性物质，以供后续二次制造使用。基础研究将解决复杂、多长度和多维的材料工艺-产品-性能关系。此外，集成几种工业数字技术(IDT)将降低风险并加快药品制造，减少实验，并大幅减少60%的开发时间和原材料/溶剂的使用。而CQA目标是通过自我优化的结晶和工艺条件来实现的。该项目将与专注于1)结晶和2)球形团聚的支持行业合作伙伴共同开发。该项目的目标包括：任务1建立自我优化、多模式结晶/粒子工程和测试平台-耦合结晶器；过滤和测试-选定的传感器和执行器，以探索选定的API/溶剂系统的知识空间。离线测试需要机器人技术吗？什么设备是可行的？目标=实现广泛的初级颗粒可达区域，通过添加剂、外场/磨机等扩展。输出=工艺条件和颗粒属性任务2开发自主颗粒形成数字自学习结晶人工智能，通过贝叶斯优化从平台内可访问条件内的饲料中提取的数据加上执行器和结晶模式。有多少参数？如何配置型号？有选择地探索仅使用溶剂、外场、添加剂以达到工程颗粒要求。识别和选择直接压缩、流动或其他目标属性的可访问粒子属性。输出=具有优化性能的材料；数据；预测设计模型和结构属性过程关系。交付成果：1)自动化结晶和粒子工程制造和测试平台，2)自主IDT驱动的制造演示器，以预测性地设计用于快速口服固体剂量/胶囊配方的原料药颗粒的最佳性能3)使用案例/第一代和第二代IDT制造演示器。

This studentship will focus on developing agile, small-scale production facilities via creation of a unique autonomous microscale API manufacturing and testing system. The system will undertake process development to produce a stable active with desired critical quality attributes (CQAs) for subsequent secondary manufacture. Fundamental research will resolve complex, multi-length, and dimensional material process-product-performance relationships. Plus, integrating several industrial digital technologies (IDTs) will de-risk and accelerate drug manufacture, reducing experiments and dramatically reducing development time and raw material/solvents use by 60% . While CQA objectives are achieved by self-optimised crystallisation and process conditions. This will be co-developed with supporting industry partners focusing on 1) crystallisation and 2) spherical agglomeration.Objectives of the project include:Task 1 Build the Self Optimizing, MultiMode Crystallization/Particle Engineering and Testing Platform - Couple crystalliser; filtration and testing - selected sensors and actuators to explore knowledge space for selected API/solvent systems. Robotics required for offline tesing? What equipment is feasible?Objective = enable wide range of primary particle attainable region, extended by additives, external fields/mill, etc.Outputs = process conditions and particle attributes Task 2 Develop the Autonomous Particle Formation Digital Twin - Self-learning crystallisation AI via Bayesian optimisation from data extracted from feeds within accessible conditions within platform plus actuators and crystallisation modes. How many parameters? How to configure model? Selectively explore solvent only, external fields, additives to achieve engineered particle requirements. Identify and select accessible particle attributes for direct compression or flow or other targeted attribute.Outputs = materials with optimised properties; data; predictive design models and structure property process relationships.Deliverables: 1) Automated crystallisation and particle engineering manufacture and testing platform, 2) Autonomous IDT-driven manufacturing demonstrator to predictively design API particulates for optimum performance for rapid oral solid dose/capsule formulation 3) Use cases / 1st and 2nd generation IDT manufacturing demonstrators.