Structures mathematical optimization

Mathematical optimization models that explicitly consider such a multi-stage structure belong to the class of multi-stage stochastic programs. A deterministic optimization model with uncertain parameters is extended to a multi-stage model by three measures ... 

In the mathematical optimization based approaches first a superstructure is created which has embedded a large number of alternative designs. Then mathematical techniques like MINLP are used to find the optimum process within the specified superstructure. For the products considered here there are two big hurdles preventing the large scale use of these techniques (Hill, 2004). Firstly a lot of the physico-chemical phenomena occurring are not completely understood. This makes rigorous modeling difficult. Secondly there is a lack of relevant property models for structured products. 

When the number and volume of the polyhedral compartments are given, the optimal structure of the foam is the one that creates the smallest total film area. This condition constitutes a formidable but straightforward mathematical optimization problem. Solution as an average, the polyhedra consist of 13.4 sides. Experimentally it was indeed found that the polyhedra most commonly found in foams have 14 sides, followed by 12 sides as a second choice. 

Multiple Objective Decision Analysis (MODA) designs the most preferred alternative within a (usually continuous) solution space using a mathematical programming structure to optimize the level of a set of quantifiable objectives. 

The objective of using a mathematical optimization model is to identify network design alternatives that best exploit structural cost differences between various locations and resolve trade-offs between different cost elements such as production cost advantages and additional transporta-tion/tariff costs. Therefore, relationships expressing the costs that will be incurred as a function of cost drivers (decision variables of the model) have to be established. A proper creation of these cost functions is a critical success factor of the overall analysis (cf. Shapiro 2001, p. 234 Vidal and Goetschalckx 1996, p. 13). 

There appears to be three fundamental approaches to the synthesis of chemical process flowsheets. The first, systematic generation, builds the flowsheet from smaller, more basic components strung together in such a way that raw materials eventually become transformed into the desired product. The second, evolutionary modification, starts with an existing flowsheet for the same or a similar product and then makes modifications as necessary to adopt the design to meet the objectives of the specific case at hand. The third, superstructure optimization, views synthesis as a mathematical optimization over structure this approach starts with a larger superflowsheet that contains embedded within it many redundant alternatives and interconnections and then systematically strips the less desirable parts of the superstructure away. 

Simultaneous Mathematical Optimization of both Structural and Design Parameters... 

Complex thermal systems cannot usually be optimized using mathematical optimization techniques. The reasons include system complexity opportunities for structural changes not identified during model development incomplete cost models and inability to consider in the model additional important factors such as plant availability, maintenability, and operability. Even if mathematical techniques are applied, the process designer gains no insight into the real thermodynamic losses, the cost formation process within the thermal system, or on how the solution was obtained. 

A number of direct ways for linking atomistic and meso-scale melt simulations have been proposed more recently. The idea behind these direct methods is to reproduce structure or thermodynamics of the atomistic simulation on the meso-scale self-consistently. As this approach is an optimization problem, mathematical optimization techniques are applicable. One of the most robust (but not very efficient) multidimensional optimizers is the simplex optimizer, which has the advantage of not needing derivatives, which are difficult to obtain in the simulation. The simplex method was first applied to optimizing atomistic simulation models to experimental data. We can formally write any observable, like, for example, the density p, as a function of the parameters of the simulation model Bj. In Eq. [2], the density is a function of the Lennard-Jones parameters. 

Empirical energy functions can fulfill the demands required by computational studies of biochemical and biophysical systems. The mathematical equations in empirical energy functions include relatively simple terms to describe the physical interactions that dictate the structure and dynamic properties of biological molecules. In addition, empirical force fields use atomistic models, in which atoms are the smallest particles in the system rather than the electrons and nuclei used in quantum mechanics. These two simplifications allow for the computational speed required to perform the required number of energy calculations on biomolecules in their environments to be attained, and, more important, via the use of properly optimized parameters in the mathematical models the required chemical accuracy can be achieved. The use of empirical energy functions was initially applied to small organic molecules, where it was referred to as molecular mechanics [4], and more recently to biological systems [2,3]. 

How many organizations do what we could really call direct rational solution of the composite structure design problem — very, very few. Perhaps only in some very restricted design areas do people feel that they can use a mathematically oriented optimization approach. That situation is unfortunate, but changing. 

DG was primarily developed as a mathematical tool for obtaining spahal structures when pairwise distance information is given [118]. The DG method does not use any classical force fields. Thus, the conformational energy of a molecule is neglected and all 3D structures which are compatible with the distance restraints are presented. Nowadays, it is often used in the determination of 3D structures of small and medium-sized organic molecules. Gompared to force field-based methods, DG is a fast computational technique in order to scan the global conformational space. To get optimized structures, DG mostly has to be followed by various molecular dynamic simulation. 

Once the selectivity is optimized, a system optimization can be performed to Improve resolution or to minimize the separation time. Unlike selectivity optimization, system cqptimization is usually highly predictable, since only kinetic parameters are generally considered (see section 1.7). Typical experimental variables include column length, particle size, flow rate, instrument configuration, sample injection size, etc. Hany of these parameters can be. Interrelated mathematically and, therefore, computer simulation and e]q>ert systems have been successful in providing a structured approach to this problem (480,482,491-493). 

---