Massimiliano Barolo

A general model-based design of experiments approach to achieve practical identifiability of pharmacokinetic and pharmacodynamic models.

The use of pharmacokinetic (PK) and pharmacodynamic (PD) models is a common and widespread practice in the preliminary stages of drug development. However, PK–PD models may be affected by structural identifiability issues intrinsically related to their mathematical formulation. A preliminary structural identifiability analysis is usually carried out to check if the set of model parameters can be uniquely determined from experimental observations under the ideal assumptions of noise-free data and no model uncertainty. However, even for structurally identifiable models, real-life experimental conditions and model uncertainty may strongly affect the practical possibility to estimate the model parameters in a statistically sound way. A systematic procedure coupling the numerical assessment of structural identifiability with advanced model-based design of experiments formulations is presented in this paper. The objective is to propose a general approach to design experiments in an optimal way, detecting a proper set of experimental settings that ensure the practical identifiability of PK–PD models. Two simulated case studies based on in vitro bacterial growth and killing models are presented to demonstrate the applicability and generality of the methodology to tackle model identifiability issues effectively, through the design of feasible and highly informative experiments.

Use of Near-Infrared Spectroscopy for Fast Fraud Detection in Seafood: Application to the Authentication of Wild European Sea Bass (Dicentrarchus labrax)

The possibility of using near-infrared spectroscopy (NIRS) for the authentication of wild European sea bass ( Dicentrarchus labrax ) was investigated in this study. Three different chemometric techniques to process the NIR spectra were developed, and their ability to discriminate between wild and farmed sea bass samples was evaluated. One approach used spectral information to directly build the discrimination model in a latent variable space; the second approach first used wavelets to transform the spectral information and subsequently derived the discrimination model using the transformed spectra; in the third approach a cascaded arrangement was proposed whereby very limited chemical information was first estimated from spectra using a regression model, and this estimated information was then used to build the discrimination model in a latent variable space. All techniques showed that NIRS can be used to reliably discriminate between wild and farmed sea bass, achieving the same classification performance as classification methods that use chemical properties and morphometric traits. However, compared to methods based on chemical analysis, NIRS-based classification methods do not require reagents and are simpler, faster, more economical, and environmentally safer. All proposed techniques indicated that the most predictive spectral regions were those related to the absorbance of groups CH, CH(2), CH(3), and H(2)O, which are related to fat, fatty acids, and water content.

Some issues in the design and operation of a batch distillation column with a middle vessel

Some issues in the design and operation of a batch distillation column with a middle vessel are addressed. Simulation results indicate that operation at infinite reflux and reboil ratios may be more profitable than conventionally considered finite reflux and reboil ratios policies. From an experimental point of view, the possibility of running a pre-existing continuous pilot plant column in a complex batch mode is investigated. Practical suggestions for column design and process control are given.

Neural-network approach to dynamic optimization of batch distillation: Application to a middle- vessel column

A framework is proposed to optimize the operation of batch columns with substantial reduction of the computational power needed to carry out the optimization calculations. The proposed framework relies on the use of an artificial neural network (ANN) based process model to be employed by the optimizer. To test the viability of the framework, the optimization of a pilot-plant middle-vessel batch column (MVBC) is considered. The maximum-product problem is formulated and solved by optimizing the column operating parameters, such as the reflux and reboil ratios and the batch time. It is shown that the ANN based model is capable of reproducing the actual plant dynamics with good accuracy, and that the proposed framework allows a large number of optimization studies to be carried out with little computational effort.

Understanding the dynamics of a batch distillation column with a middle vessel

The dynamic behavior of a batch distillation column with a middle vessel is studied. A detailed mathematical model of a pilot-plant column is developed and validated against experimental data on a highly non-ideal system. The simulated operation includes the column startup phase, during which the empty trays are sequentially filled with liquid from the top down. Then, the model is used to investigate the effect of different operating and design parameters on the column operability and productivity. In particular, it is shown that restrictions exist in the choice of the column operating variables. These restrictions are related to both the total material balance and the component balance. Interactions between the operating parameters may result in unexpected responses of the plant. All the results make clear that the dynamic behavior of a complex batch column is not so easily understandable as the one of conventional batch rectifiers and batch strippers.

Using Process Simulators for Steady-State and Dynamic Plant Analysis: An Industrial Case Study

Process simulation tools are widely adopted for the design and optimization of chemical processes. However, for quite a long time their use has been confined within research centres and highly specialized technical groups. This is especially true for dynamic simulation software, long regarded as a very specific tool requiring considerable expertise. In this work we intend to demonstrate the benefits that process engineers working on the plant may receive from an appropriate use of commercial software currently available for steady-state and dynamic simulation. A case-study concerning the purification section of an industrial plant for vinyl chloride monomer production will be considered. First of all, a steady-state simulation will be considered. Primarily, the simulation will allow a better judgement of the plant operating conditions; then it will be illustrated that sensitivity studies may produce great benefits in the general economy and productivity of the plant. Secondly, it will be shown how a dynamic model suitable for practical needs can be derived from the steady-state model. This model can be used as a powerful tool to assess the performance of the control system in handling standard operational disturbances as well as abnormal events. Simple improvements of the control system design will be also simulated and commented on.

General procedure to aid the development of continuous pharmaceutical processes using multivariate statistical modeling - an industrial case study.

Streamlining the manufacturing process has been recognized as a key issue to reduce production costs and improve safety in pharmaceutical manufacturing. Although data available from earlier developmental stages are often sparse and unstructured, they can be very useful to improve the understanding about the process under development. In this paper, a general procedure is proposed for the application of latent variable statistical methods to support the development of new continuous processes in the presence of limited experimental data. The proposed procedure is tested on an industrial case study concerning the development of a continuous line for the manufacturing of paracetamol tablets. The main driving forces acting on the process are identified and ranked according to their importance in explaining the variability in the available data. This improves the understanding about the process by elucidating how different active pharmaceutical ingredient pretreatments, different formulation modes and different settings on the processing units affect the overall operation as well as the properties of the intermediate and final products. The results can be used as a starting point to perform a comprehensive and science-based quality risk assessment that help to define a robust control strategy, possibly enhanced with the integration of a design space for the continuous process at a later stage.

Optimal design of clinical tests for the identification of physiological models of type 1 diabetes in the presence of model mismatch

How to design a clinical test aimed at identifying in the safest, most precise and quickest way the subject-specific parameters of a detailed model of glucose homeostasis in type 1 diabetes is the topic of this article. Recently, standard techniques of model-based design of experiments (MBDoE) for parameter identification have been proposed to design clinical tests for the identification of the model parameters for a single type 1 diabetic individual. However, standard MBDoE is affected by some limitations. In particular, the existence of a structural mismatch between the responses of the subject and that of the model to be identified, together with initial uncertainty in the model parameters may lead to design clinical tests that are sub-optimal (scarcely informative) or even unsafe (the actual response of the subject might be hypoglycaemic or strongly hyperglycaemic). The integrated use of two advanced MBDoE techniques (online model-based redesign of experiments and backoff-based MBDoE) is proposed in this article as a way to effectively tackle the above issue. Online model-based experiment redesign is utilised to exploit the information embedded in the experimental data as soon as the data become available, and to adjust the clinical test accordingly whilst the test is running. Backoff-based MBDoE explicitly accounts for model parameter uncertainty, and allows one to plan a test that is both optimally informative and safe by design. The effectiveness and features of the proposed approach are assessed and critically discussed via a simulated case study based on state-of-the-art detailed models of glucose homeostasis. It is shown that the proposed approach based on advanced MBDoE techniques allows defining safe, informative and subject-tailored clinical tests for model identification, with limited experimental effort.

Closed-loop optimal operation of batch distillation columns

Abstract Often, the main source of disturbance for a batch distillation system is an upset in the feed to the process. If the operation of a batch column is carried out on the basis of the nominal value of the feed composition, a high degree of uncertainty in the initial conditions to the batch may lead to run the column suboptimally, with a possibly large economic penalty. In this paper, a three-step strategy is proposed for the closed-loop implementation of optimal operating policies for batch rectifiers. First, the optimal reflux rate is calculated off-line for several feed compositions. Then, a correlation is developed off-line between the optimal reflux rate and the composition profile in the column at the end of the startup phase. Finally, the detection of the composition profile is performed on-line during the startup phase, so that the optimal reflux rate can be calculated and implemented in a closed-loop fashion. This allows operating the column optimally even though the actual feed composition is not known. Since the calculations to be performed on-line are straightforward, the computational demand is kept to a minimum. Results for binary and ternary systems indicate that, by using the proposed procedure, the column performance can be improved by as much as 30% with respect to a conventional open-loop optimal strategy.

Composition Estimations in a Middle-Vessel Batch Distillation Column Using Artificial Neural Networks

A virtual sensor that estimates product compositions in a middle-vessel batch distillation column has been developed. The sensor is based on a recurrent artificial neural network, and uses information available from secondary measurements (such as temperatures and flow rates). The criteria adopted for selecting the most suitable training data set and the benefits deriving from pre-processing these data by means of principal component analysis are demonstrated by simulation. The effects of sensor location, model initialization, and noisy temperature measurements on the performance of the soft sensor are also investigated. It is shown that the estimated compositions are in good agreement with the actual values.