ESCAPE. European Symposium on Computer Aided Process Engineering最新文献

We develop a machine learning framework that integrates high content/high throughput image analysis and artificial neural networks (ANNs) to model the separation between chemical compounds based on their estrogenic receptor activity. Natural and man-made chemicals have the potential to disrupt the endocrine system by interfering with hormone actions in people and wildlife. Although numerous studies have revealed new knowledge on the mechanism through which these compounds interfere with various hormone receptors, it is still a very challenging task to comprehensively evaluate the endocrine disrupting potential of all existing chemicals and their mixtures by pure in vitro or in vivo approaches. Machine learning offers a unique advantage in the rapid evaluation of chemical toxicity through learning the underlying patterns in the experimental biological activity data. Motivated by this, we train and test ANN classifiers for modeling the activity of estrogen receptor-α agonists and antagonists at the single-cell level by using high throughput/high content microscopy descriptors. Our framework preprocesses the experimental data by cleaning, scaling, and feature engineering where only the middle 50% of the values from each sample with detectable receptor-DNA binding is considered in the dataset. Principal component analysis is also used to minimize the effects of experimental noise in modeling where these projected features are used in classification model building. The results show that our ANN-based nonlinear data-driven framework classifies the benchmark agonist and antagonist chemicals with 98.41% accuracy.

This work addresses the control optimization of time-varying systems without the full discretization of the underlying high-fidelity models and derives optimal control trajectories using surrogate modeling and data-driven optimization. Time-varying systems are ubiquitous in the chemical process industry and their systematic control is essential for ensuring each system to be operated at the desired settings. To this end, we postulate nonlinear continuous-time control action trajectories using time-varying surrogate models and derive the parameters of these functional forms using data-driven optimization. Data-driven optimization allows us to collect data from the high-fidelity model without pursuing any discretization and fine-tune candidate control trajectories based on the retrieved input-output information from the nonlinear system. We test exponential and polynomial surrogate forms for the control trajectories and explore various data-driven optimization strategies (local vs. global and sample-based vs. model-based) to test the consistency of each approach for controlling dynamic systems. The applicability of our approach is demonstrated on a motivating example and a CSTR control case study with favorable results.

We report progress of an ongoing work to develop a virtual sensor for flowability, which is a critical tool for enabling real time process monitoring in a granulation line. The sensor is based on camera imaging to measure the size and shape distribution of granules produced by wet granulation. Then, statistical methods were used to correlate them with flowability measurements such as ring shear tests, drained angle of repose, dynamic angle of repose, and tapped density. The virtual sensor addresses the issue with these flowability measurements, which are based on off-line characterization methods that can take hours to perform. With a virtual sensor based on real-time measurement methods, the prediction of granule flowability become faster, allowing for timely decisions regarding process control and the supply chain.

Controllers are often tuned during plant commissioning, with a fixed process model. However, over time degradation can occur in the process, the process model and the controller, making it necessary to either re-tune the controller or re-identify the process model. Authors have proposed a variety of approaches to identify plant-model mismatch (PMM) and control performance degradation (CPD). While each approach may have its own advantages and disadvantages, they are generally designed to function on different timescales. The differing timescales result in the need for a multi-level hierarchical approach to monitor, detect, and manage PMM and CPD, as illustrated through a continuous pharmaceutical manufacturing application, i.e., a direct compression tablet manufacturing process. This work also highlights the requirement for index-based metrics, that enable the impact of PMM and CPD to be quantified and assessed from a control performance monitoring perspective, to aid fault diagnosis through root cause analysis to guide maintenance decisions for continuous manufacturing applications.

As the pharmaceutical industry transitions from batch to continuous manufacturing, real-time monitoring, and mechanistic model-based control are essential to conform to FDA quality standards. Glidants and lubricants are known to affect the Critical Quality Attributes (CQAs) of a tablet such as tensile strength, tablet porosity, and dissolution profile (Razavi et al., 2018; Apeji and Olowosulu, 2020). Quantitative models for predicting these effects are essential for enabling centralized control strategies of lubricant and glidant feeding and blending in direct compression tableting lines. This work presents the development of mechanistic reduced order models to capture the effects of lubricant (magnesium stearate) and glidant (silica) on CQAs and Critical Process Parameters (CPPs). A Latin Hypercube experimental campaign with thirty different mixing conditions of silica with MCC (Avicel PH200) and APAP (Acetaminophen) was carried out using a Natoli NP400 tablet press and a SOTAX AT4 tablet tester. Experiments show that the tensile strength and blend bulk density are significantly affected by the mixing conditions of silica. Similarly, adding magnesium stearate (MgSt) changes the bulk density of the blend, compaction force required to form a tablet, and tensile strength of the tablet, depending on the lubrication conditions (Mehrotra et al., 2007; Razavi et al., 2018).

A comprehensive evaluation of toxic chemicals and understanding their potential harm to human physiology is vital in mitigating their adverse effects following exposure from environmental emergencies. In this work, we develop data-driven classification models to facilitate rapid decision making in such catastrophic events and predict the estrogenic activity of environmental toxicants as estrogen receptor-α (ERα) agonists or antagonists. By combining high-content analysis, big-data analytics, and machine learning algorithms, we demonstrate that highly accurate classifiers can be constructed for evaluating the estrogenic potential of many chemicals. We follow a rigorous, high throughput microscopy-based high-content analysis pipeline to measure the single cell-level response of benchmark compounds with known in vivo effects on the ERα pathway. The resulting high-dimensional dataset is then pre-processed by fitting a non-central gamma probability distribution function to each feature, compound, and concentration. The characteristic parameters of the distribution, which represent the mean and the shape of the distribution, are used as features for the classification analysis via Random Forest (RF) and Support Vector Machine (SVM) algorithms. The results show that the SVM classifier can predict the estrogenic potential of benchmark chemicals with higher accuracy than the RF algorithm, which misclassifies two antagonist compounds.

Supply chain management is an interconnected problem that requires the coordination of various decisions and elements across long-term (i.e., supply chain structure), medium-term (i.e., production planning), and short-term (i.e., production scheduling) operations. Traditionally, decision-making strategies for such problems follow a sequential approach where longer-term decisions are made first and implemented at lower levels, accordingly. However, there are shared variables across different decision layers of the supply chain that are dictating the feasibility and optimality of the overall supply chain performance. Multi-level programming offers a holistic approach that explicitly accounts for this inherent hierarchy and interconnectivity between supply chain elements, however, requires more rigorous solution strategies as they are strongly NP-hard. In this work, we use the DOMINO framework, a data-driven optimization algorithm initially developed to solve single-leader single-follower bi-level mixed-integer optimization problems, and further develop it to address integrated planning and scheduling formulations with multiple follower lower-level problems, which has not received extensive attention in the open literature. By sampling for the production targets over a pre-specified planning horizon, DOMINO deterministically solves the scheduling problem at each planning period per sample, while accounting for the total cost of planning, inventories, and demand satisfaction. This input-output data is then passed onto a data-driven optimizer to recover a guaranteed feasible, near-optimal solution to the integrated planning and scheduling problem. We show the applicability of the proposed approach for the solution of a two-product planning and scheduling case study.

The pharmaceutical industry has been undergoing a paradigm shift towards continuous manufacturing, under which novel approaches to real-time product quality assurance have been investigated. A new perspective, entitled Quality-by-Control (QbC), has recently been proposed as an important extension and complementary approach to enable comprehensive Quality-by-Design (QbD) implementation. In this study, a QbC approach was demonstrated for a commercial scale tablet press in a continuous direct compaction process. First, the necessary understanding of the compressibility of a model formulation was obtained under QbD guidance using a pilot scale tablet press, Natoli BLP-16. Second, a data reconciliation strategy was used to reconcile the tablet weight measurement based on this understanding on a commercial scale tablet press, Natoli NP-400. Parameter estimation to monitor and update the material property variance was also considered. Third, a hierarchical three-level control strategy, which addressed the fast process dynamics of the commercial scale tablet press was designed. The strategy consisted of the Level 0 built-in machine control, Level 1 decoupled Proportional Integral Derivative (PID) control loops for tablet weight, pre-compression force, main compression force, and production rate control, and Level 2 data reconciliation of sensor measurements. The effective and reliable performance, which could be demonstrated on the rotary tablet press, confirmed that a QbC approach, based on product and process knowledge and advanced model-based techniques, can ensure robustness and efficiency in pharmaceutical continuous manufacturing.

While the importance of the Food-Energy-Water Nexus (FEW-N) has been widely accepted, a holistic approach to facilitate decision making in FEW-N systems, along with a quantitative index assessing the integrated FEW-N performance is rather lacking. In this work, we propose a FEW-N metric along with a framework to facilitate decision making for FEW-N process systems through a FEW-N integrated approach. The framework and metric are illustrated through a case study on a dairy production and processing plant. The dairy industry is a significant user of water and energy, with water being a top issue for most dairy industries and organizations worldwide. Following the framework, we develop a mixed-integer scheduling model, with alternative pathways, that faithfully replicated the major food, energy, and water aspects of a real cottage-cheese production plant. Using the developed FEW-N metric we were able to optimize the cottage-cheese plant process and observe different trade-offs between the FEW-N elements.