(442f) Computationally Assisted Cleaning Assessment – Feasibility Study

In the process industry, there has been considerable emphasis in recent years on process intensification to optimize manufacturing processes to become smaller, cleaner, and more energy-efficient. A review of these efforts shows that the focus of process intensification has been mainly on the steps in the manufacturing process that directly contribute to the product, and only little attention was given to the supporting process, e.g., cleaning procedures that are necessary to sustain the process.

While the cleaning procedures typically contribute only a small amount to the product's cost, they nevertheless strain the environment. Cleaning requires energy to pump, heat and evaporate cleaning fluids, and the solvents and detergents used for cleaning can be harmful to the environment. Establishing optimized cleaning procedures would thus not only improve the output from pharmaceutical production units but also contribute to making pharmaceutical production more sustainable.

One way of realizing an optimal cleaning procedure without experimentation is to use mechanistic modeling to calculate or predict the cleaning process. These models can help understand where and for how long the cleaning solution stays and what forces are acting on the devices over. One such model is the smoothed particle hydrodynamics (SPH). SPH uses a Lagrangian approach to simulate the fluid behavior. Simulations allow for the evaluation and testing of various process conditions in silico and the finding of optimal conditions. This approach will contribute to a more sustainable manufacturing process, remove bottlenecks, and ensure that expensive equipment can be utilized optimally.

In this work, a tool for evaluating cleaning processes in silico was developed and tested on different bin blender geometries. Experiments were performed to validate the developed model. The focus of this study was to demonstrate how such a model can be used to test different cleaning systems and potential routes for optimization.