Our research focus is in the area of Process Systems Engineering and in
particular the development and analysis of efficient understanding, modeling and solution approaches for
product and process design and process operations.
In particular the following three research areas are of interest:
(1) Process operations involving scheduling, planning and supply chain management comprising of deterministic and stochastic approaches,
(2) Modeling of reactive flow processes involving reduction of complex reaction systems appearing in combustion and environmental systems, integration of detailed chemistry with complex flow, simulation and analysis of reactor systems and metabolic engineering and
(3) Design and synthesis of flexible manufacturing systems that includes uncertainty analysis, process synthesis and optimization algorithmic development. Some of the highlights of our research are described below:
Short-term Scheduling in a Dynamic Environment
The main goal of this project is to develop an integrated methodology to handle uncertainty in short-term scheduling problem. Recognizing the fact that real plants exist in a dynamic environment where the scheduling parameters change as the schedule is being executed, a schedule developed before hand may become inefficient or even infeasible. The ability to react to unexpected events during the execution of a schedule is an important part of plant operating strategy. The uncertainty considered in this work can be viewed as (a) fluctuations in product demands, prices and processing times, and (b) unexpected deviations in unit availability (machine break-down). First, the expected range of parameter uncertainty is incorporated within the short-term scheduling model so as to determine a robust schedule capable of meeting the expected range of uncertain parameter values. This is achieved by exploiting and further analyzing the sensitivity information provided by the solution of the original Mixed Integer Linear Programming (MILP) problem. The second objective is to develop efficient models that systematically incorporate different alternatives concerning the rescheduling decisions. The goal is to determine the optimal rescheduling policy that minimizes the deviations from the old schedule while taking into account the satisfaction of other production decisions.
Design of Non-continuous Multi-product Chemical Plants: Optimization of Long-Term Plant Performance
The manufacturing of bulk drugs, intermediates and fine chemicals at large is characterized by a high level of uncertainty regarding the product mix (which products, when and at what volumes) over the intermediate and long-terms. These uncertainties also apply to the makers of the fine chemical intermediates who must ramp up their capacities even sooner sometimes before the target drug can be projected to be clinically useful or safe. The objective of this project is to develop a systematic methodology for the design of very agile, large-scale pilot plants capable of consecutively running any of a large number of unplanned chemical steps without significant modifications. It is proposed that systematic process design methods can be developed to impact the key pilot plant attributes to a large capacity commercial plant for the production of fine chemicals/intermediate and bulk drugs. This target will be reached without excessive capital outlays, resulting in a manufacturing capability that is agile, productive and sparing of capital investment over its useful life. Specifically, it is proposed to bring together a large assembly of bulk drug processing skills and advanced mathematical methods that have shown success in process design under uncertainty.
Modeling of Reactive Flow Processes
Reactor models based on first principles have been proved to provide accurate results over a wide range of operating conditions, reactor types, charges and transport dynamics. Unfortunately though a complex kinetic network is most commonly needed that results in excessive computational expense even with the present increased computer power.
The challenging problem lies on the interaction of the various physical processes with complex reaction kinetics in turbulent reacting flows. The main limitation is the existence of a wide spectrum of time and length scales in turbulent flows that makes Direct Numerical Simulation (DNS) of Navier-Stokes equations infeasible in terms of computational requirements. Thus simplified mixing and reaction models are needed as tools for the description of complex reaction systems.
In this work we are proposing first to decouple these two important issues by developing:
(a) an efficient approach for reduction of kinetic models, and
(b) a sufficiently detailed mixing model and
then investigate ways of taking into account mixing effects at the stage of reduction.
Moreover, uncertainty considerations are of great interest in this project. In kinetic modeling some of the potential sources of uncertainty include reaction rate parameters, thermodynamic parameters, such as species heat of formation and entropy, initial conditions and transport properties. The objective of this work is to investigate the effects of uncertainty in complex reaction networks. More specifically the questions that are addressed are the following. First, what is the range of validity where the kinetic model is valid, second, how uncertainty information can be incorporated in the mechanism generation so as to determine a kinetic model feasible for the whole range of uncertainty and finally how uncertainty propagates through the system. Particular emphasis is given to environmental and combustion systems.
Domain Decomposition Approach for Complex Multi-scale Dynamic Systems
A major challenge in modeling macroscopic chemical and separation systems is that elementary processes encompass a wide range of length and time scales. The disparity of scales necessitates the use of different approaches at each scale for solving such problems. Simultaneous consideration within a single uniform framework requires excessive computing times and moreover there might exist more exact solution techniques for addressing each problem separately. Our work focuses on the development of large-scale computational methods that allow the analysis of chemical and separation processes with appropriate spatial and temporal resolution at multiple length and time scales. Special attention is devoted to the coupling with other physical processes and the inherent multi-dimensionality of the problems. The computational methods developed in our research are novel domain decomposition techniques based on modified block-waveform relaxation coupled with subspace-iteration methods.
Issues of Uncertainty in a Supply Chain Management
This project focuses first on the development of new ways to quantify the performance of the system under uncertainty, and techniques of uncertainty propagation, and second the application of such uncertainty analysis tools within the plant operations. This encompasses scheduling, planning and supply chain management for which modeling efforts are being currently developed in my group.
Existing approaches for uncertainty analysis either are two expensive computationally (e.g. Monte Carlo techniques) or are too limited in terms of the amount of information that one can obtain (e.g. flexibility analysis). A well-recognized technique that targets on a similar objective is the Six Sigma analysis tool. What this tool achieves is to measure not only the nominal process performance but also its whole distribution that can be quantified by its variance (sigma). It is well known that moving towards the tail of the distribution the probability of fault behavior is getting smaller, thus process that are characterized by a large range of distribution have very small probability for "bad" performance and consequently they are much more profitable.
This work targets the development of efficient and systematic techniques to quantify the range of variability of systems given uncertainty in process parameters. A comprehensive approach will take into account specific system's behavior and model predictive techniques so as to identify a critical set of required experiments. The work will be extending on developing efficient propagation techniques for complex dynamic systems. Stochastic Surface Response methods will be used as an alternative to traditional Monte Carlo to reduce the enormous computational complexity. Extensions of these techniques to include time and spatial variability explicitly will be also investigated.
Optimization of Hepatocyte Function
Extracorporeal bioartificial liver devices (BAL) are perhaps among the most promising technologies for the treatment of liver failure, but significant technical challenges remain in order to develop systems with sufficient processing capacity and of manageable size. One key limitation is that during BAL operation, when the device is exposed to plasma from the patient, hepatocytes are prone to accumulate intracellular lipids and exhibit poor liver-specific functions.
Based on hepatic intermediary metabolism, we have utilized mathematical programming techniques to optimize the biochemical environment of hepatocyte cultures towards the desired effect of increased albumin and urea synthesis. To investigate the feasible range of optimal hepatic function, we have obtained a Pareto optimal set of solutions corresponding to liver specific functions of urea and albumin secretion in the metabolic framework using multiobjective optimization.
The importance of amino acids in the supplementation and the criticality of the metabolic pathways have been investigated using logic based programming techniques. Since the metabolite measurements are bound to be patient specific, and hence subject to variability, uncertainty has to be integrated with system analysis to improve the prediction of hepatic function.
We have used the concept of two-stage stochastic programming to obtain robust solutions by considering extracellular variability. The proposed analysis represents a new systematic approach to analyze behavior of hepatocyte cultures and optimize different operating parameters for an extracorporeal device based on real-time conditions. This work has been accepted for publication recently in Biotechnology and Bioengineering (Sharma et al. 2005).
Pharmaceutical Process Optimization
In this project the problem we are trying to solve is the optimization of process design for a variety of pharmaceutical manufacturing processing units. The project involves both experimental and computational components. Experiments are used to build the necessary knowledge database for the behavior of the process under a variety of different operating conditions as well as the effect of different design parameters on the system's performance. Predictive models to represent the system's performance are developed using hybrid schemes involving first principle models and data based methodologies.
The first stage of the project addresses single units, whereas at the second stage the interactions between processing units will be analyzed and considered in an integrative modeling framework in order to identify the additional bottlenecks that arise due to process connectivity. Finally the last stage will involve the optimization of the overall production plant alternatives. This will be of particular importance in the design of the optimal control scheme.
|
Recent Publications
Portillo, P.M., F.J. Muzzio, and M.G. Ierapetritou. Hybrid DEM-compartment modeling approach for granular mixing. AIChE J. 53 , 119, 2007.
Jia, Z. and M.G. Ierapetritou. Generate Pareto Optimal Solutions of Scheduling Problems using Normal Boundary Intersection Technique. Comp. Chem. Eng. 31, 268, 2007.
Goyal, V., and M.G. Ierapetritou. Stochastic MINLP Optimization using Simplicial Approximation. Comp. Chem. Eng. 31,1081, 2007.
Davis, E., M.G. Ierapetritou. Adaptive Optimization of Noisy Black-Box Functions Inherent In Microscopic Models. Comp. Chem. Eng. 31 , 466, 2007.
Portillo, P. F. Muzzio and M.G. Ierapetritou. Characterizing Powder Mixing Processes utilizing Compartment Models. Int. Jl. Of Pharm. 320 , 14, 2006.
Jia, Z., and M.G. Ierapetritou. Uncertainty Analysis on the Right-Hand-Side for MILP problems AIChE J., 52, 2486, 2006.
Wu, D., and M.G. Ierapetritou. Lagrangean Decomposition Using an Improved Nelder-Mead Approach for Lagrangean Multiplier Update. Comp. Chem. Eng. 30,778, 2006.
Banerjee, I., and M.G. Ierapetritou. An Adaptive Reduction Scheme to Model Reactive Flow. Comb. & Flame 144, 219, 2006.
Bindal, A., M.G. Ierapetritou, S. Balakrishnan, A. Makeev, I. Kevrekidis and A. Armaou Equation-free, coarse-grained computational optimization using timesteppers. Chem. Eng. Sci. 61, 279, 2006.
Sharma, N., M.G. Ierapetritou and M.L. Yarmush. Novel Quantitative Tools for Engineering Analysis of Hepatocyte Cultures in Bioartificial Liver Systems. Biotech. & Bioeng. 92(3) , 321, 2005.
Balakrishnan S., Roy A., Ierapetritou M.G., Flach G.P. and Georgopoulos P.G.A Comparative Assessment of Efficient Uncertainty Analysis Techniques for Environmental Fate and Transport Models: Application to the FACT Model. Journal of Hydrology 307(1-4) , 204, 2005.
Goyal, V. and M.G. Ierapetritou. Multiobjective Framework for Modular Design Generation Incorporating Demand Uncertainty. Ind. Eng. Chem. Res. 44 , 3594, 2005.
Banerjee, I., and M.G. Ierapetritou. Feasibility evaluation of nonconvex systems using shape reconstruction techniques. Ind. Eng. Chem. Res. 44 , 3638, 2005.
Sirdeshpande, A.R., M.G. Ierapetritou, M.J. Andrecovich, J.P. Naumovitz. Process synthesis optimization and flexibility evaluation of air separation cycles. AIChE J. 51 , 1190, 2005.
|
