Optimisation method

In this book the discussion of optimisation methods will, of necessity, be limited to a brief review of the main techniques used in process and equipment design. The extensive literature on the subject should be consulted for full details of the methods available, and their application and limitations see Beightler and Wilde (1967), Beveridge and Schechter (1970), Stoecker (1989), Rudd and Watson (1968), Edgar and Himmelblau (2001). The books by Rudd and Watson (1968) and Edgar and Himmelblau (2001) are particularly recommended to students. 

Table 2. Flipping ratios of strong and weak reflections measured starting from different time proportions (three steps of 60s). When using the optimised method, the standard deviation c(R) does not depend on the initial conditions. 

T. Herges, H. Merlitz, and W. Wenzel. Stochastic optimisation methods for biomolecular structure prediction. J. Ass. Lab. Autom., 7 98-104, 2002. 

Pareto Optimality, which makes provisions about mixtures in the whole factor space, therefore cannot be used in combination with a sequential optimisation method. 

So far this approach is analogous to most of the simultaneous optimisation methods. However, the optimisation is not continued by preselecting desired values for any criterion to construct contour plots (Figure 4.13 and 4.14), or to search for acceptable solutions [29]. 

OPTIMISED METHOD CONDITIONS USING RESULTS FROM THE... 

Although application of chemometrics in sample preparation is very uncommon, several optimisation techniques may be used to optimise sample preparation systematically. Those techniques can roughly be divided into simultaneous and sequential methods. The main restrictions of a sequential simplex optimisation [6,7] find their origin in the complexity of the optimisation function needed. This function is a predefined function, often composed of several criteria. Such a composite criterion may lead to ambiguous results [8]. Other important disadvantages of simplex optimisation methods are that not seldom local optima are selected instead of global optima and that the number of experiments needed is not known beforehand. 

As the first test case, we selected 20 aliphatic primary amines from a set of 493 commercially available ones by three different methods random reagent selection, entropy-optimised ProSAR selection and an occupancy-optimised method which purely maximises the occupancy of pharmacophore bins (56), i.e. ensures that as many bins as possible are covered by the reagent selection regardless of the pharmacophore distribution. As the greedy algorithm... 

Analytical chemists are by nature innovators and seekers of improvement. In the development area these qualities are invaluable in optimising method performance. Alas far too often, this desire for continuous improvement spills over into the interpretation of methods for quality control. Here we require consistency of application and rigorous control of processes and procedures. These aspects are anathema for many practitioners of the art of chemical analysis . 

Optimisation methods may also be used to maximise key parameters, e.g. resolution, but are beyond the scope of this handbook. Miller and Miller s book on Statistics for Analytical Chemistry provides a gentle introduction to the topic of optimisation methods and response surfaces as well as digestible background reading for most of the statistical topics covered in this handbook. For those wishing to delve deeply into the subject of chemometric methods, the Handbook of Chemometrics and Qualimetrics in two volumes by Massart et al., is a detailed source of information. 

Equation 8.18 presented the general, non-linear optimisation method for calorimetric measurements based on a reaction model, and Equation 8.20 presented the general, non-linear optimisation method for spectroscopic measurements based on a reaction model. A comparison of these two equations reveals that the non-linear parameters dit...tNp, which are defined by the reaction model, are common to both equations. Only the linear parameters in these equations (E and Ar H. ..v,..) are different. As the linear parameters in both... 

There should also be continued research to optimise methods to detoxify contaminated agricultural crops to produce safe products. 

Cuthrell and Biegler (1987) and Renfro et al. (1987) developed dynamic optimisation methods based on the infeasible path approach. The main advantage of this approach is that it avoids repetitive simulations during iteration of the... 

Mujtaba and Macchietto (1994) presented an industrial case study in which dynamic optimisation method of Mujtaba and Macchietto (1993) is utilised for the development of the optimal operation of an entire batch distillation campaign where 100 batches of fresh charge have to be processed with secondary reprocessing of intermediate off-cuts. The process involved a complex separation of a five-component mixture of industrial interest, described using non-ideal thermodynamic models. In addition, the operation of the whole production campaign was subject to a number of resource constraints, for example -... 

In this chapter first, the optimisation method of Al-Tuwaim and Luyben (1991) for single separation duty is presented. Then the optimisation problem formulation and solution considered by Mujtaba and Macchietto (1996) is explained. Finally, the optimisation problem formulations considered by Logsdon et al. (1990) and Bonny et al. (1996) are presented. 

Note that the results of the maximum profit problem obtained using the techniques presented above will be close to those determined by rigorous optimisation method (using the techniques presented in Mujtaba and Macchietto, 1993, 1996) only if the polynomial approximations are very good as were the case for the example presented here. Mujtaba et al. (2004) presented Neural Network based approximations of these functions. 

Neural Network based hybrid dynamic modelling and optimisation methods for conventional and unconventional column configurations... 

C. K. Bayne and I. B. Rubin, Practical Experimental Designs and Optimisation Methods for Chemists, VCH, Deerfield Beach, FL, 1986. 

This volume discusses some iterative optimisation methods drawn from within AI. Iterative methods of the sort discussed here are often computationally intensive, and as long as computers were of modest power these algorithms struggled to compete with other methods. Indeed, their computational demands had the effect of severely limiting the interests of scientists in them until the last decade of the twentieth century. However, evolutionary methods have, as we shall see, special advantages and they have become increasingly attractive as computer power has grown. 

Competition unavoidably requires a population size greater than 1 - a single individual cannot compete with itself. Since EAs show evolutionary behaviour, it is reasonable to anticipate that they too will normally need to work upon a group of individuals. This requires that these algorithms must operate on many potential solutions simultaneously, so that selection pressure can be applied to cull the poorer solutions and drive the search towards those of higher quality. This manipulation of a group, or population of solutions, is a fundamental difference with most other optimisation methods, which typically create and then refine a single solution. 

The GA is the most widely used EA within chemistry. There is a substantial, and growing, literature which demonstrates the power of this method as an optimisation tool within not just the chemical sciences, but across a range of scientific disciplines. Provided that the problem is of the appropriate structure, it is a very powerful optimisation method. It is also tolerant, in other words it is capable of providing useful solutions even when values of variables governing its operation, such as the population size, are not chosen optimally. 

There are limits to the applicability of virtually every optimisation method. A GA may only be applied to a problem if it is possible to express the solution as a sequence of values. This sequence is referred to as a chromosome or a string, and each parameter within the chromosome is a gene. The entire set of genes constitutes the genotype, and the solution to which this genotype corresponds is known as the phenotype (Fig. 8). The GA works to refine strings with the help of the evolutionary operators outlined previously. 

MOGP is based on the more traditional optimisation method genetic programming (GP), which is a type of GA [53,54]. The main difference between GP and a GA is in the chromosome representation in a GA an individual is usually represented by a fixed-length linear string, whereas in GP individuals are represented by treelike structures hence, they can vary in shape and size as the population undergoes evolution. The internal nodes of the tree, typically represent mathematical operators, and the terminal nodes, typically represent variables and constant values thus, the chromosome can represent a mathematical expression as shown in Fig. 4. 

There is already an extensive literature relating to compound-selection methods, from which it is possible to identify four major classes of method although, as we shall see, there is some degree of overlap between these four classes, viz cluster-based methods, dissimilarity-based methods, partition-based methods and optimisation methods. The next four sections of this chapter present the various algorithms that have been suggested for each approach we then discuss comparisons and applications of these algorithms, and the chapter concludes with some thoughts on further developments in the field. 

For an in-depth discussion of the many aspects involved in choosing an empirical potential for the purpose of structure prediction using global optimisation methods, see ref. 51. 

---