Issue link: https://www.e-digitaleditions.com/i/1420709
48 Pharmaceutical Technology APIs, EXCIPIENTS, AND MANUFACTURING eBOOK 2021 P h a r mTe c h . c o m Manufacturing for diversion decisions. The purpose here has been to verify the functionality of the developed tool. The toolbox fits the model well with the data. This module of the toolbox is implemented based on tank-in-series model. However, this module has the flexibility to use the axial dispersion model as well. Results and discussions Figure 4 shows the pulse response of the RTD model with real feed frame RTD parameters. Figure 5 shows the convolution prediction result with the model and measured inlet concentration data. The red dashed line are limits of the concentration and the black line is the reject signal. When outlet con- centration falls in limits, the signal goes to 0, other- wise the signal goes to 1. Conclusion Python implementation of the RTD-based control toolbox can finish not only the task of fitting the RTD tank-in-series model, but also use the model and an input signal to predict the outlet concentra- tion of a continuous pharmaceutical manufactur- ing plant. The toolbox has two integrated mod- ules to complete each task. These modules can be used together in a continuous process or be used separately. The data transmission between the toolbox and control platform, and also among the toolbox itself, is based on .csv files. These provides the f lexibility to use this tool with any commer- cially available complementary tools. This feature guarantees the openness of the toolbox and its overall convenience. Acknowledgements This work is supported by FDA. References 1. A. Bhaskar, F.N. Barros, and R. Singh, Int J Pharm 534(1–2) 159–78 (2017). 2. A. Bhaskar, R. Singh, J Pharm Innov) 14 (3) 316–331 (2019). PT Figure 5. Prediction of outlet concentration and reject signal. 1: accept, 0: reject. The modular toolbox can be integrated with any commercially available tools to control tablet potency.