Multi-bucket optimization for integrated planning and scheduling in the perishable dairy supply chain

This paper considers a dairy industry problem on integrated planning and scheduling of set yoghurt production. A mixed integer linear programming formulation is introduced to integrate tactical and operational decisions and a heuristic approach is proposed to decompose time buckets of the decisions. The decomposition heuristic improves computational efficiency by solving big bucket planning and small bucket scheduling problems. Further, mixed integer linear programming and constraint programming methodologies are combined with the algorithm to show their complementary strengths. Numerical studies using illustrative data with high demand granularity (i.e., a large number of small-sized customer orders) demonstrate that the proposed decomposition heuristic has consistent results minimizing the total cost (i.e., on average 8.75% gap with the best lower bound value found by MILP) and, the developed hybrid approach is capable of solving real sized instances within a reasonable amount of time (i.e., on average 92% faster than MILP in CPU time).     

Some efficient approaches for multi-objective constrained optimization of computationally expensive black-box model problems

Multi-objective constrained optimization problems which arise in many engineering fields often involve computationally expensive black-box model simulators of industrial processes which have to be solved with limited computational time budget, and hence limited number of simulator calls. This paper proposes two heuristic approaches aiming to build proxy problem models, solvable by computationally efficient optimization methods, in order to quickly provide a sufficiently accurate approximation of the Pareto front. The first approach builds a multi-objective mixed-integer linear programming (MO-MILP) surrogate model of the optimization problem relying on piece-wise linear approximations of objectives and constraints obtained through brute-force sensitivity computation. The second approach builds a multi-objective nonlinear programming (MO-NLP) surrogate model using curve fitting of objectives and constraints. In both approaches the desired number of approximated solutions of the Pareto front are generated by applying the ɛ-constraint method to the multi-objective surrogate problems. The proposed approaches are tested for the cost vs. life cycle assessment (LCA)-based environmental optimization of drinking water production plants. The results obtained with both approaches show that a good quality approximation of Pareto front can be obtained with a significantly smaller computational time than with a state-of-the-art metaheuristic algorithm.     

Simultaneous MINLP synthesis of heat exchanger networks comprising different exchanger types

This paper describes the simultaneous MINLP synthesis of heat integrated heat exchanger networks comprising different heat exchanger types. The stage-wise superstructure of heat exchanger networks (HEN) by Yee and Grossmann (Comput. Chem. Eng. 14 (1990) p. 1165) is extended to alternative exchanger types. The selection of the types is modeled by disjunctions based on operating limitations and the required heat transfer area. Since different types of heat exchangers involve different design geometries, which influences the inlet and outlet temperatures of heat exchangers, additional constraints are specified to provide a feasible temperature distribution in HEN. The consideration of different exchanger types drastically increases the combinatorics, size and computation effort needed to solve the problem. The integer-infeasible path MINLP approach has been applied to perform an efficient initialization scheme and to halve CPU times for solving MILP master problem of the modified OA/ER algorithm. A special multilevel MINLP procedure in reduced integer space has been proposed to solve medium size HEN problems (20 streams) comprising 10³ and more binary variables.     

Exploring flux representations of complex kinetics networks

To extract meaningful information from complex kinetic models involving a large number of species and reactions, advanced computational techniques are required. In this work, new approaches have been proposed based on element flux calculations for systematic kinetic analysis of complex reaction models. These approaches quantify element transformation flux between species to determine a metric that accurately captures the production and consumption of species. Furthermore, a graph searching procedure is employed to retrieve all possible reaction pathways from the highly complex reaction networks. Element fluxes involved in these pathways provide an indicator to quantitatively evaluate pathway activities. Based on pathway activities, a novel approach is proposed to project the totality of the information contained in pathway weights onto a single scalar, reactivity status indicator, which enables a compact representation of local chemistry. The proposed approaches are illustrated with highly complex kinetic mechanisms describing oxidation of n‐pentane, n‐heptane, and a biodiesel surrogate methyl‐butanoate. © 2011 American Institute of Chemical Engineers AIChE J, 2012     

Highly Accurate Prediction of Protein-Protein Interactions via Incorporating Evolutionary Information and Physicochemical Characteristics

Protein-protein interactions (PPIs) occur at almost all levels of cell functions and play crucial roles in various cellular processes. Thus, identification of PPIs is critical for deciphering the molecular mechanisms and further providing insight into biological processes. Although a variety of high-throughput experimental techniques have been developed to identify PPIs, existing PPI pairs by experimental approaches only cover a small fraction of the whole PPI networks, and further, those approaches hold inherent disadvantages, such as being time-consuming, expensive, and having high false positive rate. Therefore, it is urgent and imperative to develop automatic in silico approaches to predict PPIs efficiently and accurately. In this article, we propose a novel mixture of physicochemical and evolutionary-based feature extraction method for predicting PPIs using our newly developed discriminative vector machine (DVM) classifier. The improvements of the proposed method mainly consist in introducing an effective feature extraction method that can capture discriminative features from the evolutionary-based information and physicochemical characteristics, and then a powerful and robust DVM classifier is employed. To the best of our knowledge, it is the first time that DVM model is applied to the field of bioinformatics. When applying the proposed method to the Yeast and Helicobacter pylori (H. pylori) datasets, we obtain excellent prediction accuracies of 94.35% and 90.61%, respectively. The computational results indicate that our method is effective and robust for predicting PPIs, and can be taken as a useful supplementary tool to the traditional experimental methods for future proteomics research.     

A systematic approach for synthesizing combined mass and heat exchange networks

Mass and heat are very important resources in the process industry. Numerous approaches have been proposed for the optimization of either mass or heat exchange networks. Since the process usage of mass and heat is typically intertwined, it is important to account for such interactions. The objective of this paper is to introduce a systematic method for the simultaneous synthesis of combined mass- and heat-exchange networks (CMAHENs). The proposed method is based on a novel approach that incorporates the mass pinch technology (MPT) for mass exchange networks (MENs) synthesis and the pseudo-T-H diagram approach (PTHDA) for the heat exchange networks (HENs) synthesis. New bypass streams are included in the structural representation of the problem to expand the search space. A combined optimization approach is applied to minimize the total annualized cost of the CMAHEN. Finally, two cases are solved to illustrate the application of the proposed method.     

Global optimization of large‐scale mixed‐integer linear fractional programming problems: A reformulation‐linearization method and process scheduling applications

Mixed‐integer linear fractional program (MILFP) is a class of mixed‐integer nonlinear programs (MINLP) where the objective function is the ratio of two linear functions and all constraints are linear. Global optimization of large‐scale MILFPs can be computationally intractable due to the presence of discrete variables and the pseudoconvex/pseudoconcave objective function. We propose a novel and efficient reformulation–linearization method, which integrates Charnes–Cooper transformation and Glover's linearization scheme, to transform general MILFPs into their equivalent mixed‐integer linear programs (MILP), allowing MILFPs to be globally optimized effectively with MILP methods. Extensive computational studies are performed to demonstrate the efficiency of this method. To illustrate its applications, we consider two batch scheduling problems, which are modeled as MILFPs based on the continuous‐time formulations. Computational results show that the proposed approach requires significantly shorter CPU times than various general‐purpose MINLP methods and shows similar performance than the tailored parametric algorithm for solving large‐scale MILFP problems. Specifically, it performs with respect to the CPU time roughly a half of the parametric algorithm for the scheduling applications. © 2013 American Institute of Chemical Engineers AIChE J, 59: 4255–4272, 2013     

Optimal management of logistic activities in multi-site environments

The new emerging area of Enterprise Wide Optimization (EWO) has focused the attention in effectively solving the combined production/distribution scheduling problem. The importance of logistic activities performed in multi-site environments comes from the relative magnitude of the associated transportation costs and the good chance of getting large savings on such expenses. This paper first develops an exact MILP mathematical formulation for the multiple vehicle time-window-constrained pickup and delivery (MVPDPTW) problem. The approach is able to account for many-to-many transportation requests, pure pickup and delivery tasks, heterogeneous vehicles and multiple depots. Optimal solutions for a variety of benchmark problems with cluster/random distributions of pickup and delivery locations and limited sizes in terms of customer requests and vehicles have been discovered. However, the computational cost exponentially grows with the number of requests. For large-scale m-PDPTW problems, a local search improvement algorithm steadily providing a better solution through two evolutionary steps is also presented. A neighborhood structure around the starting solution is generated by first allowing multiple request exchanges among nearby trips and then permitting the reordering of nodes on every individual route. If a better set of routes is found, both steps are repeated until no improved solution is discovered. Compact MILP mathematical formulations for both sub-problems have been developed and solved through an efficient branch-and-bound algorithm. A significant number of large-scale m-PDPTW benchmark problems, some of them including up to 100 transportation requests, were successfully solved in reasonable CPU times.     

Synthesis of multipass heat exchanger networks using genetic algorithms

In this work, a methodology based on genetic algorithms (GAs) is developed for the optimal synthesis of multipass heat exchanger networks (HENs). The network model is based on a stagewise superstructure, and the problem of finding the optimum number of 1–2 shells in series of multipass heat exchangers is aided by an efficient optimization model that uses the standard F_T design method. The proposed methodology allows for proper handling of the trade-offs involving energy consumption, number of units, number of 1–2 shells and network area to provide a network with the minimum total annual cost. The results of the examples show that the new approach is able to find more economical networks than those generated by other methods.     

带显著误差集的改进MILP数据协调方法

Mixed integer linear programming (MILP) approach for simultaneous gross error detection and data reconciliation has been proved as an efficient way to adjust process data with material, energy, and other balance constrains. But the efficiency will decrease significantly when this method is applied in a large-scale problem because there are too many binary variables involved. In this article, an improved method is proposed in order to generate gross error candidates with reliability factors before data rectification. Candidates are used in the MILP objective function to improve the efficiency and accuracy by reducing the number of binary variables and giving accurate weights for suspected gross errors candidates. Performance of this improved method is compared and discussed by applying the algorithm in a widely used industrial example.     

Computational Protein Engineering: Bridging the Gap between Rational Design and Laboratory Evolution

Enzymes are tremendously proficient catalysts, which can be used as extracellular catalysts for a whole host of processes, from chemical synthesis to the generation of novel biofuels. For them to be more amenable to the needs of biotechnology, however, it is often necessary to be able to manipulate their physico-chemical properties in an efficient and streamlined manner, and, ideally, to be able to train them to catalyze completely new reactions. Recent years have seen an explosion of interest in different approaches to achieve this, both in the laboratory, and in silico. There remains, however, a gap between current approaches to computational enzyme design, which have primarily focused on the early stages of the design process, and laboratory evolution, which is an extremely powerful tool for enzyme redesign, but will always be limited by the vastness of sequence space combined with the low frequency for desirable mutations. This review discusses different approaches towards computational enzyme design and demonstrates how combining newly developed screening approaches that can rapidly predict potential mutation “hotspots” with approaches that can quantitatively and reliably dissect the catalytic step can bridge the gap that currently exists between computational enzyme design and laboratory evolution studies.     

Particle velocities and their residence time distribution in the riser of a CFB

The riser of a Circulating Fluidised Bed (CFB) is the key-component where gas-solid or gas-catalytic reactions occur. Both types of reactions require different conditions of operating velocities (U), solids circulation fluxes (G), overall hydrodynamics and residence times of solids and gas. The solids hydrodynamics and their residence time distribution in the riser are the focal points of this paper. The riser of a CFB can operate in different hydrodynamic regimes, each with a pronounced impact on the solids motion. These regimes are firstly reviewed to define their distinct characteristics as a function of the combined parameters, U and G.Experiments were carried out, using Positron Emission Particle Tracking of single radio-actively labelled tracer particles. Results on the particle velocity are assessed for operation in the different regimes. Design equations are proposed.The particle velocities and overall solids mixing are closely linked. The solid mixing has been previously studied by mostly tracer response techniques, and different approaches have been proposed. None of the previous approaches unambiguously fits the mixing patterns throughout the different operating regimes of the riser. The measured average particle velocity and the velocity distribution offer an alternative approach to determine the solids residence time distribution (RTD) for a given riser geometry. Findings are transformed into design equations.The overall approach is finally illustrated for a riser of known geometry and operating within the different hydrodynamic regimes.     

Quantitative analysis of cellular metabolic dissipative, self-organized structures

One of the most important goals of the postgenomic era is understanding the metabolic dynamic processes and the functional structures generated by them. Extensive studies during the last three decades have shown that the dissipative self-organization of the functional enzymatic associations, the catalytic reactions produced during the metabolite channeling, the microcompartmentalization of these metabolic processes and the emergence of dissipative networks are the fundamental elements of the dynamical organization of cell metabolism. Here we present an overview of how mathematical models can be used to address the properties of dissipative metabolic structures at different organizational levels, both for individual enzymatic associations and for enzymatic networks. Recent analyses performed with dissipative metabolic networks have shown that unicellular organisms display a singular global enzymatic structure common to all living cellular organisms, which seems to be an intrinsic property of the functional metabolism as a whole. Mathematical models firmly based on experiments and their corresponding computational approaches are needed to fully grasp the molecular mechanisms of metabolic dynamical processes. They are necessary to enable the quantitative and qualitative analysis of the cellular catalytic reactions and also to help comprehend the conditions under which the structural dynamical phenomena and biological rhythms arise. Understanding the molecular mechanisms responsible for the metabolic dissipative structures is crucial for unraveling the dynamics of cellular life.     

A rule-based genetic algorithm for the scheduling of single-stage multi-product batch plants with parallel units

This paper presents a heuristic rule-based genetic algorithm (GA) for large-size single-stage multi-product scheduling problems (SMSP) in batch plants with parallel units. SMSP have been widely studied by the researchers. Most of them used mixed-integer linear programming (MILP) formulation to solve the problems. With the problem size increasing, the computational effort of MILP increases greatly. Therefore, it is very difficult for MILP to obtain acceptable solutions to large-size problems within reasonable time. To solve large-size problems, the preferred method in industry is the use of scheduling rules. However, due to the constraints in SMSP, the simple rule-based method may not guarantee the feasibility and quality of the solution. In this study, a random search based on heuristic rules was proposed first. Through exploring a set of random solutions, better feasible solutions can be achieved. To improve the quality of the random solutions, a genetic algorithm-based on heuristic rules has been proposed. The heuristic rules play a very important role in cutting down the solution space and reducing the search time. Through comparative study, the proposed method demonstrates promising performance in solving large-size SMSP.     

Mathematical modeling for simultaneous design of plants and supply chain in the batch process industry

Most supply chain design models have focused on the integration problem, where links among nodes must be settled in order to allow an efficient operation of the whole system. At this level, all the problem elements are modeled like black boxes, and the optimal solution determines the nodes allocation and their capacity, and links among nodes. In this work, a new approach is proposed where decisions about plant design are simultaneously made with operational and planning decisions on the supply chain. Thus, tradeoffs between the plant structure and the network design are assessed. The model considers unit duplications and the allocation of storage tanks for plant design. Using different sets of discrete sizes for batch units and tanks, a mixed integer linear programming model (MILP) is attained. The proposed formulation is compared with other non-integrated approaches in order to illustrate the advantages of the presented simultaneous approach.     

Computer Simulation of Enzymatic Reactions

Our strategy in modeling ezymatic reactions is demonstrated by considering two complimentary approaches, which are based on the powerful EVB method. The first approach uses the X-ray coordinates at their face value and estimates the reaction potential surface by using a simplified and efficient microscopic dipolar model for the solvent. The second approach (which became practical only upon the emergence of supercomputers) uses a surface constrained all atom solvent model, molecular dynamics (MD) and the umbrella sampling method to evaluate the actual free energy barrier. Both approaches indicate that the correct evaluation of the “solvation energy” of the ionic resonance structures is the key for understanding enzyme catalysis. The MD version of the EVB method is used in a preliminary study of the validity of linear free-energy relationships in enzymatic reactions. This study supports the ad hoc assumptions introduced in the simplified method.     

Systematic optimisation technology for high-level screening and scoping of complex distillation systems

Established procedures for complex distillation synthesis employ detailed models that are unable to function for high-level screening. In this article, a new approach is presented in the form of Conceptual Programming for the preliminary screening of complex distillation systems. Conceptual Programming employs a generic supertask representation that replaces the need of superstructures. Tasks stand for simple distillation columns and hybrids represent complex distillation arrangements. A supertask accounts for all possible configurations without introducing representation and modelling difficulties. The basic tasks are then modelled using appropriate shortcut or semi-rigorous methods. A mixed integer linear programming (MILP) problem is formulated to obtain the optimum structure and performance targets. The strengths of the approach are highlighted with a light alcohol separation problem.     

Computational strategies for large-scale MILP transshipment models for heat exchanger network synthesis

Determining the minimum number of units is an important step in heat exchanger network synthesis (HENS). The MILP transshipment model (Papoulias and Grossmann, 1983) and transportation model (Cerda and Westerberg, 1983) were developed for this purpose. However, they are computationally expensive when solving for large-scale problems. Several approaches are studied in this paper to enable the fast solution of large-scale MILP transshipment models. Model reformulation techniques are developed for tighter formulations with reduced LP relaxation gaps. Solution strategies are also proposed for improving the efficiency of the branch and bound method. Both approaches aim at finding the exact global optimal solution with reduced solution times. Several approximation approaches are also developed for finding good approximate solutions in relatively short times. Case study results show that the MILP transshipment model can be solved for relatively large-scale problems in reasonable times by applying the approaches proposed in this paper.