Unified modelling in support of organization design and change

The aim of this research is to provide a unified modelling-based method to help with the evaluation of organization design and change decisions. Relevant literature regarding model-driven organization design and change is described. This helps identify the requirements for a new modelling methodology. Such a methodology is developed and described. The three phases of the developed method include the following. First, the use of CIMOSA-based multi-perspective enterprise modelling to understand and capture the most enduring characteristics of process-oriented organizations and externalize various types of requirement knowledge about any target organization. Second, the use of causal loop diagrams to identify dynamic causal impacts and effects related to the issues and constraints on the organization under study. Third, the use of simulation modelling to quantify the effects of each issue in terms of organizational performance. The design and case study application of a unified modelling method based on CIMOSA (computer integrated manufacturing open systems architecture) enterprise modelling, causal loop diagrams, and simulation modelling, is explored to illustrate its potential to support systematic organization design and change. Further application of the proposed methodology in various company and industry sectors, especially in manufacturing sectors, would be helpful to illustrate complementary uses and relative benefits and drawbacks of the methodology in different types of organization. The proposed unified modelling-based method provides a systematic way of enabling key aspects of organization design and change. The case company, its relevant data, and developed models help to explore and validate the proposed method. The application of CIMOSA-based unified modelling method and integrated application of these three modelling techniques within a single solution space constitutes an advance on previous best practice. Also, the purpose and application domain of the proposed method offers an addition to knowledge. © IMechE 2009.

Sequential decision making in general state space models

Many problems in control and signal processing can be formulated as sequential decision problems for general state space models. However, except for some simple models one cannot obtain analytical solutions and has to resort to approximation. In this thesis, we have investigated problems where Sequential Monte Carlo (SMC) methods can be combined with a gradient based search to provide solutions to online optimisation problems. We summarise the main contributions of the thesis as follows. Chapter 4 focuses on solving the sensor scheduling problem when cast as a controlled Hidden Markov Model. We consider the case in which the state, observation and action spaces are continuous. This general case is important as it is the natural framework for many applications. In sensor scheduling, our aim is to minimise the variance of the estimation error of the hidden state with respect to the action sequence. We present a novel SMC method that uses a stochastic gradient algorithm to find optimal actions. This is in contrast to existing works in the literature that only solve approximations to the original problem. In Chapter 5 we presented how an SMC can be used to solve a risk sensitive control problem. We adopt the use of the Feynman-Kac representation of a controlled Markov chain flow and exploit the properties of the logarithmic Lyapunov exponent, which lead to a policy gradient solution for the parameterised problem. The resulting SMC algorithm follows a similar structure with the Recursive Maximum Likelihood(RML) algorithm for online parameter estimation. In Chapters 6, 7 and 8, dynamic Graphical models were combined with with state space models for the purpose of online decentralised inference. We have concentrated more on the distributed parameter estimation problem using two Maximum Likelihood techniques, namely Recursive Maximum Likelihood (RML) and Expectation Maximization (EM). The resulting algorithms can be interpreted as an extension of the Belief Propagation (BP) algorithm to compute likelihood gradients. In order to design an SMC algorithm, in Chapter 8 uses a nonparametric approximations for Belief Propagation. The algorithms were successfully applied to solve the sensor localisation problem for sensor networks of small and medium size.

Muitiobjective pressurized water reactor reload core design by nondominated genetic algorithm search

The design of pressurized water reactor reload cores is not only a formidable optimization problem but also, in many instances, a multiobjective problem. A genetic algorithm (GA) designed to perform true multiobjective optimization on such problems is described. Genetic algorithms simulate natural evolution. They differ from most optimization techniques by searching from one group of solutions to another, rather than from one solution to another. New solutions are generated by breeding from existing solutions. By selecting better (in a multiobjective sense) solutions as parents more often, the population can be evolved to reveal the trade-off surface between the competing objectives. An example illustrating the effectiveness of this novel method is presented and analyzed. It is found that in solving a reload design problem the algorithm evaluates a similar number of loading patterns to other state-of-the-art methods, but in the process reveals much more information about the nature of the problem being solved. The actual computational cost incurred depends: on the core simulator used; the GA itself is code independent.

Optimal H2 order-one reduction by solving eigenproblems for polynomial equations

A method is given for solving an optimal H2 approximation problem for SISO linear time-invariant stable systems. The method, based on constructive algebra, guarantees that the global optimum is found; it does not involve any gradient-based search, and hence avoids the usual problems of local minima. We examine mostly the case when the model order is reduced by one, and when the original system has distinct poles. This case exhibits special structure which allows us to provide a complete solution. The problem is converted into linear algebra by exhibiting a finite-dimensional basis for a certain space, and can then be solved by eigenvalue calculations, following the methods developed by Stetter and Moeller. The use of Buchberger's algorithm is avoided by writing the first-order optimality conditions in a special form, from which a Groebner basis is immediately available. Compared with our previous work the method presented here has much smaller time and memory requirements, and can therefore be applied to systems of significantly higher McMillan degree. In addition, some hypotheses which were required in the previous work have been removed. Some examples are included.

The modified Kanerva model for automatic speech recognition

A parallel processing network derived from Kanerva's associative memory theory Kanerva 1984 is shown to be able to train rapidly on connected speech data and recognize further speech data with a label error rate of 0·68%. This modified Kanerva model can be trained substantially faster than other networks with comparable pattern discrimination properties. Kanerva presented his theory of a self-propagating search in 1984, and showed theoretically that large-scale versions of his model would have powerful pattern matching properties. This paper describes how the design for the modified Kanerva model is derived from Kanerva's original theory. Several designs are tested to discover which form may be implemented fastest while still maintaining versatile recognition performance. A method is developed to deal with the time varying nature of the speech signal by recognizing static patterns together with a fixed quantity of contextual information. In order to recognize speech features in different contexts it is necessary for a network to be able to model disjoint pattern classes. This type of modelling cannot be performed by a single layer of links. Network research was once held back by the inability of single-layer networks to solve this sort of problem, and the lack of a training algorithm for multi-layer networks. Rumelhart, Hinton & Williams 1985 provided one solution by demonstrating the "back propagation" training algorithm for multi-layer networks. A second alternative is used in the modified Kanerva model. A non-linear fixed transformation maps the pattern space into a space of higher dimensionality in which the speech features are linearly separable. A single-layer network may then be used to perform the recognition. The advantage of this solution over the other using multi-layer networks lies in the greater power and speed of the single-layer network training algorithm. © 1989.

Finite volume distance field solution applied to medial axis transform

Accurate and efficient computation of the nearest wall distance d (or level set) is important for many areas of computational science/engineering. Differential equation-based distance/ level set algorithms, such as the hyperbolic-natured Eikonal equation, have demonstrated valuable computational efficiency. Here, in the context, as an 'auxiliary' equation to the main flow equations, the Eikonal equation is solved efficiently with two different finite volume approaches (the cell vertex and cell-centered). Application of the distance solution is studied for various geometries. Moreover, a procedure using the differential field to obtain the medial axis transform (MAT) for different geometries is presented. The latter provides a skeleton representation of geometric models that has many useful analysis properties. As an alternative approach to the pure geometric methods (e.g. the Voronoi approach), the current d-MAT procedure bypasses many difficulties that are usually encountered by pure geometric methods, especially in three dimensional space. It is also shown that the d-MAT approach provides the potential to sculpt/control the MAT form for specialized solution purposes. Copyright © 2010 by the American Institute of Aeronautics and Astronautics, Inc.

Structure Discovery in Nonparametric Regression through Compositional Kernel Search

Despite its importance, choosing the structural form of the kernel in nonparametric regression remains a black art. We define a space of kernel structures which are built compositionally by adding and multiplying a small number of base kernels. We present a method for searching over this space of structures which mirrors the scientific discovery process. The learned structures can often decompose functions into interpretable components and enable long-range extrapolation on time-series datasets. Our structure search method outperforms many widely used kernels and kernel combination methods on a variety of prediction tasks.

Generalized power method for sparse principal component analysis

In this paper we develop a new approach to sparse principal component analysis (sparse PCA). We propose two single-unit and two block optimization formulations of the sparse PCA problem, aimed at extracting a single sparse dominant principal component of a data matrix, or more components at once, respectively. While the initial formulations involve nonconvex functions, and are therefore computationally intractable, we rewrite them into the form of an optimization program involving maximization of a convex function on a compact set. The dimension of the search space is decreased enormously if the data matrix has many more columns (variables) than rows. We then propose and analyze a simple gradient method suited for the task. It appears that our algorithm has best convergence properties in the case when either the objective function or the feasible set are strongly convex, which is the case with our single-unit formulations and can be enforced in the block case. Finally, we demonstrate numerically on a set of random and gene expression test problems that our approach outperforms existing algorithms both in quality of the obtained solution and in computational speed. © 2010 Michel Journée, Yurii Nesterov, Peter Richtárik and Rodolphe Sepulchre.

Computational engineering design for micro-scale combustion devices: A thermally improved configuration

A multi-objective design optimisation study has been carried out with the objectives to improve the overall efficiency of the device and to reduce the fuel consumption for the proposed micro-scale combustor design configuration. In a previous study we identified the topology of the combustion chamber that produced improved behaviour of the device in terms of the above design criteria. We now extend our design approach, and we propose a new configuration by the addition of a micro-cooling channel that will improve the thermal behaviour of the design as previously suggested in literature. Our initial numerical results revealed an improvement of 2.6% in the combustion efficiency when we applied the micro-cooling channel to an optimum design configuration we identified from our earlier multi-objective optimisation study, and under the same operating conditions. The computational modelling of the combustion process is implemented in the commercial computational fluid dynamics package ANSYS-CFX using Finite Rate Chemistry and a single step hydrogen-air reaction. With this model we try to balance good accuracy of the combustion solution and at the same time practicality within the context of an optimisation process. The whole design system comprises also the ANSYS-ICEM CFD package for the automatic geometry and mesh generation and the Multi-Objective Tabu Search algorithm for the design space exploration. We model the design problem with 5 geometrical parameters and 3 operational parameters subject to 5 design constraints that secure practicality and feasibility of the new optimum design configurations. The final results demonstrate the reliability and efficiency of the developed computational design system and most importantly we assess the practicality and manufacturability of the revealed optimum design configurations of micro-combustor devices. Copyright © 2013 by ASME.

Low-rank optimization with trace norm penalty

The paper addresses the problem of low-rank trace norm minimization. We propose an algorithm that alternates between fixed-rank optimization and rank-one updates. The fixed-rank optimization is characterized by an efficient factorization that makes the trace norm differentiable in the search space and the computation of duality gap numerically tractable. The search space is nonlinear but is equipped with a Riemannian structure that leads to efficient computations. We present a second-order trust-region algorithm with a guaranteed quadratic rate of convergence. Overall, the proposed optimization scheme converges superlinearly to the global solution while maintaining complexity that is linear in the number of rows and columns of the matrix. To compute a set of solutions efficiently for a grid of regularization parameters we propose a predictor-corrector approach that outperforms the naive warm-restart approach on the fixed-rank quotient manifold. The performance of the proposed algorithm is illustrated on problems of low-rank matrix completion and multivariate linear regression. © 2013 Society for Industrial and Applied Mathematics.