Data-driven Process Systems Engineering Lab
Data-driven Process Systems Engineering Lab
2011 AIChE Annual Meeting in Minneapolis
October 17, 2011
The main purpose of this work is to develop the necessary tools required to integrate, simulate and design continuous manufacturing systems for pharmaceutical powder based products. The development of a detailed simulation will enable (a) the optimization of the integrated process (steady-state and dynamic), (b) the quantification of uncertainty propagation from early stages of the manufacturing line down to the final product properties, (c) the evaluation of different control strategies for maintaining the process within its design space, and (d) the identification of possible integration bottlenecks and conflicting objectives. One of the initial challenges of this effort was the identification of the appropriate modeling platform and the integration of different types of modeling techniques describing each unit operation. Currently, even though the tools and methods to simulate integrated processes involving fluids is achieved through commercially available and validated flowsheet program packages , the current progress of dynamic simulation of solid processing systems is not as advanced. The recently developed gPROMs-SOLIDS  simulation package is evaluated in this work to perform dynamic flowsheet simulation of two different scenarios of a multi-component tablet manufacturing system, namely direct compaction and roller compaction. In addition, a dynamic sensitivity analysis technique is developed to investigate the effects of different forms of uncertainty inherent in the material properties or the model parameters of the process during operation. Pharmaceutical products are in their vast majority powder based and must satisfy very strict quality specifications in order to meet the regulatory standards. Historically, manufacturing in the pharmaceutical industry has been carried out in batch mode which potentially results in expensive, inefficient and poorly controlled processes[3-4]. However, recently the potential for improving product quality and advantages of continuous manufacturing has been recognized by both pharmaceutical industries and regulatory authorities . In addition, recent powder characterization studies have provided significant insight into how material properties of active ingredients, and excipients and their compositions in a mixture affect the behavior of the powders in different apparatus. Therefore, a great opportunity arises for merging knowledge, experience, experimental results and modeling tools for developing continuous manufacturing platform that will enable the safe implementation of this transition. In this work we focus on the manufacturing of oral solid dosage form drugs through two different case studies. The first case study is a very common manufacturing strategy for powder based products, namely direct compaction, which involves three basic processing steps: powder feeding, blending, and compaction. The second scenario is a more complex processing line which is used for overcoming problems such as poor flowability and it involves powder feeding, mixing, roller compaction, milling and compaction. The integrated design of such continuous systems requires the detailed characterization of the unit operations involved in order to quantify the functional relationship of key quality attributes with process parameters and material properties. The models used for each unit operation range from simple data-driven models to more complex first- principle based models, depending on the available knowledge for each process operation. Specifically, the feeding system is modeled as a first order transfer function with time delay based on dynamic experimental data performed to characterize the dynamics of the process under different operating conditions. Powder blending has been studied extensively in the literature which result in a number of different models describing its performance [6-7]. In this work, two models are developed to describe the performance of the mixer, a data-based residence time distribution model and a two dimensional population balance model. The roller compaction dynamics are described by the work proposed in , where the ribbon thickness and density are correlated with the input material properties and operating conditions of the process. Results obtained from this integrated simulation verify experimental observations regarding the effect of possible feeder refilling perturbations to the performance of downstream processes. In addition, the use of recycling streams is investigated and is found to be advantageous for a specific range of recycle ratios. The second aspect of this study is associated with the uncertainty inherent in the design and synthesis step of a continuous pharmaceutical manufacturing process and how this affects the performance of different unit operations as well as final product properties. In a flowsheet simulation every unit operation interacts with other process operations and a variation of a parameter in one unit might affect the outcome of the entire process. Thus there is a need to use Global Sensitivity Analysis techniques. The uncertain parameters are assigned a probability distribution which is the then sampled to generate a set of scenarios. These scenarios are evaluated using Monte Carlo simulations and sampling-based (Partial Ranked Correlation Coefficients) and variance-based (Sobol’ first and total order) sensitivity measures are calculated to provide a Global Sensitivity Analysis [9-10]. To assess the importance of various model parameters in process dynamics, time-varying sensitivity indexes are calculated for specific time points of interest . This allows assessing whether the parameters are important over the entire simulation and to quantify which of the output variables and units are more sensitive to perturbations.
In summary, the main objective of this study is to build a dynamic flowsheet simulation of an integrated continuous downstream pharmaceutical process, using a combination of fundamental and empirical models. Using two cases, the results elucidate (a) the evolution of key particle properties during the transient state and possible perturbations, (b) the effect of changes in process parameters and/or material properties, which typically can vary during continuous manufacturing and (c) the dynamic response and recycle dynamics of an integrated blender and a recirculation tank. To complement this study, a dynamic sensitivity analysis framework can identify significant uncertain model parameters and material properties during different phases of dynamic operation of the integrated process. This work aims to provide important insight in the synthesis, design and optimization of continuous manufacturing of pharmaceutics.References: 1. Biegler T L, Grossmann E I, and Westerberg W.A, Systematic Methods of Chemica Process Design. International Series in the Physical and Chemical Engineering Sciences. 1997, New Jersey: Prentice Hall. 2. ProcessSystemsEnterprise, gPROMS Advanced User Guide. 2003: London, UK. 3. Gorsek, A. and P. Glavic, Design of Batch Versus Continuous Processes: Part I: Single-Purpose Equipment. Chemical Engineering Research and Design, 1997. 75(7): p. 709-717. 4. Leuenberger, H., New trends in the production of pharmaceutical granules: batch versus continuous processing. European Journal of Pharmaceutics and Biopharmaceutics, 2001. 52(3): p. 289-296. 5. Plumb, K., Continuous Processing in the Pharmaceutical Industry: Changing the Mind Set.Chemical Engineering Research and Design, 2005. 83(6): p. 730-738. 6. Boukouvala, F., et al., Computational approaches for studying the granular dynamics of continuous blending processes II: Population balance and data-based methods, Manuscript under review. 7. Dubey, A., et al., Computational Approaches for the study of granular dynamics of continuous blending process- I : DEM based methods. Macromolecular Materials and Engineering, 2010(Accepted for Publication). 8. Hsu, S.-H., G. Reklaitis, and V. Venkatasubramanian, Modeling and Control of Roller Compaction for Pharmaceutical Manufacturing. Part I: Process Dynamics and Control Framework.Journal of Pharmaceutical Innovation, 2010. 5(1): p. 14-23. 9. Saltelli, A., K. Chan, and E.M. Scott, Sensitivity analysis. Wiley series in probability and statistics. 2000, Chichester ; New York: Wiley. xv, 475 p. 10. Saltelli, A., Sensitivity analysis in practice : a guide to assessing scientific models. 2004, Hoboken, NJ: Wiley. xi, 219 p. 11. Marino, S., et al., A methodology for performing global uncertainty and sensitivity analysis in systems biology. J Theor Biol, 2008. 254(1): p. 178-96.