Modelling the effect of base component properties and processing conditions on mixture products using probabilistic, knowledge-guided neural networks.

Development of materials by mixing different base components is a widespread methodology to create materials with improved properties compared to those of its base components. However, efficient determination of the properties of mixture-based materials during design remains challenging without prior knowledge of the underlying physical phenomena. In this work a new data-based methodology is proposed involving the use of probabilistic, knowledge-guided artificial neural networks to jointly model the properties of the base components, the proportions in which they are mixed, and the processing conditions used during manufacture to predict properties of final products.

A framework for the in silico assessment of the robustness of an MPC in a CDC line in function of process variability.

The shift from batch manufacturing towards continuous manufacturing for the production of oral solid dosages requires the development and implementation of process models and process control. Previous work focused mainly on developing deterministic models for the investigated system. Furthermore, the in silico tuning and analysis of a control strategy are mostly done based on deterministic models.

Validation of model-based design of experiments for continuous wet granulation and drying.

This paper presents an application case of model-based design of experiments for the continuous twin-screw wet granulation and fluid-bed drying sequence. The proposed framework consists of three previously developed models. Here, we are testing the applicability of previously published unit operation models in this specific part of the production line to a new active pharmaceutical ingredient.

T-shaped partial least squares for high-dosed new active pharmaceutical ingredients in continuous twin-screw wet granulation: Granule size prediction with limited material information.

This work presents a granule size prediction approach applicable to diverse formulations containing new active pharmaceutical ingredients (APIs) in continuous twin-screw wet granulation. The approach consists of a surrogate selection method to identify similar materials with new APIs and a T-shaped partial least squares (T-PLS) model for granule size prediction across varying formulations and process conditions. We devised a surrogate material selection method, employing a combination of linear pre-processing and nonlinear classification algorithms, which effectively identified suitable surrogates for new materials.

Mechanistic modeling of semicontinuous fluidized bed drying of pharmaceutical granules by incorporating single particle and bulk drying kinetics.

In this work, a mechanistic fluidized bed drying model computing the granule moisture content in function of granule size, drying time, process settings and formulation properties is developed. Modeling the moisture content distribution concerning the granule size is essential for tabletability and drug product quality. This work combines a mechanistic bulk model and a single-particle drying kinetics model in a semicontinuous mode.

Effect of process parameters and formulation properties on the lead-lag between in-line NIR tablet press feed frame and off-line NIR tablet measurements.

The use of in-line near-infrared (NIR) measurements for tablet potency monitoring and diversion was studied. First, the optimal sample size for in-line NIR measurements inside the feed chute and the dosing and filling chamber of the tablet press feed frame was determined to allow proper comparison between these different measurement positions. Because of the considerably longer measurement time needed to obtain the same sample size inside the feed chute compared to the feed frame, the possibility of powder segregation inside the feed chute and the additional powder mixing inside the feed frame, the latter is preferred over the feed chute for in-line blend potency monitoring.

Partial least squares regression to calculate population balance model parameters from material properties in continuous twin-screw wet granulation.

In the pharmaceutical industry, twin-screw wet granulation has become a realistic option for the continuous manufacturing of solid drug products. Towards the efficient design, population balance models (PBMs) have been recognized as a tool to compute granule size distribution and understand physical phenomena. However, the missing link between material properties and the model parameters limits the swift applicability and generalization of new active pharmaceutical ingredients (APIs).

Elucidation of the powder flow pattern in a twin-screw LIW-feeder for various refill regimes.

The importance of residence time distribution modeling is acknowledged as a tool for enabling material tracking and control within a continuous manufacturing line in order to safeguard both product quality and production efficiency. One of the first unit-operations into a continuous direct compression line (i.e.

An extended 3-compartment model for describing step change experiments in pharmaceutical twin-screw feeders at different refill regimes.

Residence time distributions (RTDs) are a valuable tool for product tracking in the unit operations of a continuous line for manufacturing pharmaceutical oral solid dosage (OSD) and the integrated system itself. The first unit operation in such a continuous line in which extended intermixing can occur, is typically a feeder. The RTD of a feeder can be obtained by performing tracer experiments with a tracer material.

A multivariate formulation and process development platform for direct compression.

The efficient development of robust tableting processes is challenging due to the lack of mechanistic understanding on the impact of raw material properties and process parameters on tablet quality. The experimental determination of the effect of process and formulation parameters on tablet properties and subsequent optimization is labor-intensive, expensive and time-consuming. The combined use of an extensive raw material property database, process simulation tools and multivariate modeling allows more efficient and more optimized development of the direct compression (DC) process.

Determination and understanding of lead-lag between in-line NIR tablet press feed frame and off-line NIR tablet measurements.

The influence of different tableting process parameters on lead-lag was studied by collecting in-line near-infrared (NIR) spectra in the filling chamber of the tablet press feed frame and off-line NIR tablet data. Lead-lag is defined as the difference in time and API concentration between the measured in-line feed frame NIR response and the off-line NIR tablet data. Lead-lag results from the product formulation blend undergoing additional mixing after passing the NIR probe inside the feed frame, before being filled into the dies of the tablet press.

Development of a tablet press feed frame lead lag determination model using in-line and off-line NIR measurements.

For continuous pharmaceutical manufacturing of oral solid dosages, it is essential that product quality is measured inline. In this application, a continuous rotary tablet press is used. The goal is a model-based assessment of the quality of the blend in the feed frame to determine whether the concentration of the active pharmaceutical ingredient (API) will be within the prescribed limits.

TPLS as predictive platform for twin-screw wet granulation process and formulation development.

In recent years, the interest in continuous manufacturing techniques, such as twin-screw wet granulation, has increased. However, the understanding of the influence of the combination of raw material properties and process settings upon the granule quality attributes is still limited. In this study, a T-shaped partial least squares (TPLS) model was developed to link raw material properties, the ratios in which these raw materials were combined and the applied process parameters for the twin-screw wet granulation process with the granule quality attributes.

Improvement of a 1D Population Balance Model for Twin-Screw Wet Granulation by Using Identifiability Analysis.

Recently, the pharmaceutical industry has undergone changes in the production of solid oral dosages from traditional inefficient and expensive batch production to continuous manufacturing. The latest advancements include increased use of continuous twin-screw wet granulation and application of advanced modeling tools such as Population Balance Models (PBMs). However, improved understanding of the physical process within the granulator and improvement of current population balance models are necessary for the continuous production process to be successful in practice.

In-line temperature measurement to improve the understanding of the wetting phase in twin-screw wet granulation and its use in process development.

Wetting is the initial stage of wet granulation processes during which the first contact between the powder and the liquid occurs. Wetting is a critical step to allow granule growth and consolidation, but also to ensure uniform active pharmaceutical ingredient (API) distribution over all granule size fractions. A physical understanding of the wetting stage is therefore crucial to design a robust granulation process.

Model-based NIR spectroscopy implementation for in-line assay monitoring during a pharmaceutical suspension manufacturing process.

The implementation of Process Analytical Technology (PAT) instruments is generally achieved stochastically. Sub-optimal PAT locations could introduce variation in the measurements which is not related to the analyte of interest. For this reason, rational approaches should be considered to establish an optimal sensor placement where relevant measurements are possible and the impact of disturbances is minimized.

In-depth experimental analysis of pharmaceutical twin-screw wet granulation in view of detailed process understanding.

Twin-screw wet granulation is gaining increasing interest within the pharmaceutical industry for the continuous manufacturing of solid oral dosage forms. However, limited prior fundamental physical understanding has been generated relating to the granule formation mechanisms and kinetics along the internal compartmental length of a twin-screw granulator barrel, and about how process settings, barrel screw configuration and formulation properties such as particle size, density and surface properties influence these mechanisms. One of the main reasons for this limited understanding is that experimental data is generally only collected at the exit of the twin-screw granulator barrel although the granule formation occurs spatially along the internal length of the barrel.

Application of iterative robust model-based optimal experimental design for the calibration of biocatalytic models.

The aim of model calibration is to estimate unique parameter values from available experimental data, here applied to a biocatalytic process. The traditional approach of first gathering data followed by performing a model calibration is inefficient, since the information gathered during experimentation is not actively used to optimize the experimental design. By applying an iterative robust model-based optimal experimental design, the limited amount of data collected is used to design additional informative experiments.