Design of Process Monitoring and Analysis Systems, using a Model-based Computer Aided Framework
Systematic methods and tools for managing the complexity
Process Analytical Technology - PAT (T4-6)
Keywords: PAT, process monitoring, process analysis, design, framework
In the manufacturing industry, for example, the pharmaceutical industry, obtaining a consistent end-product quality is mandatory. A thorough understanding of the process as well as a properly designed monitoring and analysis system are required to obtain the desired end-product properties. This work focuses on developing a model-based computer-aided framework that will provide the methods and tools through which methodology for the design of the monitoring and analysis systems related to product quality control as an add-on to traditional chemical process design methods, can be developed. The design of a process monitoring and analysis system is a step-wise procedure consisting of the selection of critical process variables, followed by the proper selection and placement of suitable monitoring and analysis equipments to keep track on the behavior of these critical process variables, and finally, also including the coupling of the monitoring and analysis system to a control system to ensure that critical process variables can be controlled.
The starting point for the design methodology is the problem definition in terms of data related to the process specifications and the desired product quality, which are provided by the product manufacturer (or PAT system designer). A knowledge base and a model library are two important supporting tools within the framework. The knowledge base provides the necessary data during the design. It contains information related to process monitoring and analysis systems and includes for each unit process, a list of critical process variables that should be monitored, a list of the corresponding manipulating variables (actuators), and the corresponding equipments typically used for their measurement (including data related to accuracy, robustness, time, etc.). The model library contains the models which relate the process variables. These models will support the analysis, for example by predicting process variables which are not measurable, thus generating the additional or missing data needed to obtain the final design. The framework guides the user through a systematic analysis (design methodology) related to the process-product and the important process-product variables to obtain the desired design for the product monitoring and analysis system. Optimization of the design can be achieved in terms of product data analysis time, cost of monitoring equipment subject to the maintenance of the desired product quality. The design methodology together with the model-based computer-aided framework has a potentially wide industrial application range, has the ability to assist in faster product-process development, and to allow identification of potential problems in the early stages of process-product development.
The objective of this paper is to highlight the design methodology together with the model-based computer-aided framework. First, the structure of the knowledge base has been developed. Later on, information was introduced in this knowledge base on the basis of an extensive literature search combined with an industrial survey and data collection. A collection of the necessary process operational models have been created through a computer-aided modeling tool (ICAS-MOT, developed at CAPEC). Process analysis, identification of critical points and process variables and selection of the appropriate monitoring and analysis system for the product-process will be illustrated through two case studies: a first case study, a tablet manufacturing process, is mainly focused on the pharmaceutical industry; a second case study, a fermentation process demonstrates the applicability of the framework for the bio-based process industries.
* Corresponding author (rag@kt.dtu.dk) / (kvg@kt.dtu.dk)/ (rs@kt.dtu.dk)
Presented Wednesday 19, 12:20 to 12:40, in session Process Analytical Technology - PAT (T4-6).
