Diffusion of Administrative Innovation: TQM Implementation and Effectiveness in a Global

The main objective of this dissertation is to create new knowledge on an administrative innovation, its adoption, diffusion and finally its effectiveness. In this dissertation the administrative innovation is approached through a widely utilized management philosophy, namely the total quality management (TQM) strategy. TQM operationalizes a self-assessment procedure, which is based on continual improvement principles and measuring the improvements. This dissertation also captures the theme of change management as it analyzes the adoption and diffusion of the administrative innovation. It identifies innovation characteristics as well as organisational and individual factors explaining the adoption and implementation. As a special feature, this study also explores the effectiveness of the innovation based on objective data. For studying the administrative innovation (TQM model), a multinational Case Company provides a versatile ground for a deep, longitudinal analysis. The Case Company started the adoption systematically in the mid 1980s in some of its units. As part of their strategic planning today, the procedure is in use throughout the entire global company. The empirical story begins from the innovation adoption decision that was made in the Case Company over 22 years ago. In order to be able to capture the right atmosphere and backgrounds leading to the adoption decision, key informants from that time were interviewed, since the main target was to clarify the dynamics of how an administrative innovation develops. In addition, archival material was collected and studied, available memos and data relating to the innovation, innovation adoption and later to the implementation contained altogether 20500 pages of documents. A survey was furthermore conducted at the end of 2006 focusing on questions related to the innovation, organization and leadership characteristics and the response rate totalled up to 54%. For measuring the effectiveness of the innovation implementation, the needed longitudinal objective performance data was collected. This data included the profit unit level experience of TQM, the development of the self assessment scores per profit unit and performance data per profit unit measured with profitability, productivity and customer satisfaction. The data covered the years 1995-2006. As a result, the prerequisites for the successful adoption of an administrative innovation were defined, such as the top management involvement, support of the change agents and effective tools for implementation and measurement. The factors with the greatest effect on the depth of the implementation were the timing of the adoption and formalization. The results also indicated that the TQM model does have an effect on the company performance measured with profitability, productivity and customer satisfaction. Consequently this thesis contributes to the present literature (i) by taking into its scope an administrative innovation and focusing on the whole innovation implementation process, from the adoption, through diffusion until its consequences, (ii) because the studied factors with an effect on the innovation adoption and diffusion are multifaceted and grouped into individual, organizational and environmental factors, and a strong emphasis is put on the role of the individual change agents and (iii) by measuring the depth and consistency of the administrative innovation. This deep analysis was possible due to the availability of longitudinal data with triangulation possibilities.

On state and parameter estimation in chaotic systems

State-of-the-art predictions of atmospheric states rely on large-scale numerical models of chaotic systems. This dissertation studies numerical methods for state and parameter estimation in such systems. The motivation comes from weather and climate models and a methodological perspective is adopted. The dissertation comprises three sections: state estimation, parameter estimation and chemical data assimilation with real atmospheric satellite data. In the state estimation part of this dissertation, a new filtering technique based on a combination of ensemble and variational Kalman filtering approaches, is presented, experimented and discussed. This new filter is developed for large-scale Kalman filtering applications. In the parameter estimation part, three different techniques for parameter estimation in chaotic systems are considered. The methods are studied using the parameterized Lorenz 95 system, which is a benchmark model for data assimilation. In addition, a dilemma related to the uniqueness of weather and climate model closure parameters is discussed. In the data-oriented part of this dissertation, data from the Global Ozone Monitoring by Occultation of Stars (GOMOS) satellite instrument are considered and an alternative algorithm to retrieve atmospheric parameters from the measurements is presented. The validation study presents first global comparisons between two unique satellite-borne datasets of vertical profiles of nitrogen trioxide (NO3), retrieved using GOMOS and Stratospheric Aerosol and Gas Experiment III (SAGE III) satellite instruments. The GOMOS NO3 observations are also considered in a chemical state estimation study in order to retrieve stratospheric temperature profiles. The main result of this dissertation is the consideration of likelihood calculations via Kalman filtering outputs. The concept has previously been used together with stochastic differential equations and in time series analysis. In this work, the concept is applied to chaotic dynamical systems and used together with Markov chain Monte Carlo (MCMC) methods for statistical analysis. In particular, this methodology is advocated for use in numerical weather prediction (NWP) and climate model applications. In addition, the concept is shown to be useful in estimating the filter-specific parameters related, e.g., to model error covariance matrix parameters.

Mass transfer and particleinteractions in granular systems and suspension flows

The purpose of this study was to investigate some important features of granular flows and suspension flows by computational simulation methods. Granular materials have been considered as an independent state ofmatter because of their complex behaviors. They sometimes behave like a solid, sometimes like a fluid, and sometimes can contain both phases in equilibrium. The computer simulation of dense shear granular flows of monodisperse, spherical particles shows that the collisional model of contacts yields the coexistence of solid and fluid phases while the frictional model represents a uniform flow of fluid phase. However, a comparison between the stress signals from the simulations and experiments revealed that the collisional model would result a proper match with the experimental evidences. Although the effect of gravity is found to beimportant in sedimentation of solid part, the stick-slip behavior associated with the collisional model looks more similar to that of experiments. The mathematical formulations based on the kinetic theory have been derived for the moderatesolid volume fractions with the assumption of the homogeneity of flow. In orderto make some simulations which can provide such an ideal flow, the simulation of unbounded granular shear flows was performed. Therefore, the homogeneous flow properties could be achieved in the moderate solid volume fractions. A new algorithm, namely the nonequilibrium approach was introduced to show the features of self-diffusion in the granular flows. Using this algorithm a one way flow can beextracted from the entire flow, which not only provides a straightforward calculation of self-diffusion coefficient but also can qualitatively determine the deviation of self-diffusion from the linear law at some regions nearby the wall inbounded flows. Anyhow, the average lateral self-diffusion coefficient, which was calculated by the aforementioned method, showed a desirable agreement with thepredictions of kinetic theory formulation. In the continuation of computer simulation of shear granular flows, some numerical and theoretical investigations were carried out on mass transfer and particle interactions in particulate flows. In this context, the boundary element method and its combination with the spectral method using the special capabilities of wavelets have been introduced as theefficient numerical methods to solve the governing equations of mass transfer in particulate flows. A theoretical formulation of fluid dispersivity in suspension flows revealed that the fluid dispersivity depends upon the fluid properties and particle parameters as well as the fluid-particle and particle-particle interactions.

Fouling management and retention in nanofiltration of integrated paper mill effluents

Nanofiltration performance was studied with effluents from the pulp and paper industry and with model substances. The effect of filtration conditions and membrane properties on nanofiltration flux, retention, and fouling was investigated. Generally, the aim was to determine the parameters that influence nanofiltration efficiency and study how to carry out nanofiltration without fouling by controlling these parameters. The retentions of the nanofiltration membranes studied were considerably higher than those of tight ultrafiltration membranes, and the permeate fluxes obtained were approximately the same as those of tight ultrafiltration membranes. Generally, about 80% retentions of total carbon and conductivity were obtained during the nanofiltration experiments. Depending on the membrane and the filtration conditions, the retentions of monovalent ions (chloride) were between 80 and 95% in the nanofiltrations. An increase in pH improved retentions considerably and also the flux to some degree. An increase in pressure improved retention, whereas an increase in temperature decreased retention if the membrane retained the solute by the solution diffusion mechanism. In this study, more open membranes fouled more than tighter membranes due to higher concentration polarization and plugging of the membrane material. More irreversible fouling was measured for hydrophobic membranes. Electrostatic repulsion between the membrane and the components in the solution reduced fouling but did not completely prevent it with the hydrophobic membranes. Nanofiltration could be carried out without fouling, at least with the laboratory scale apparatus used here when the flux was below the critical flux. Model substances had a strong form of the critical flux, but the effluents had only a weak form of the critical flux. With the effluents, some fouling always occurred immediately when the filtration was started. However, if the flux was below the critical flux, further fouling was not observed. The flow velocity and pH were probably the most important parameters, along with the membrane properties, that influenced the critical flux. Precleaning of the membranes had only a small effect on the critical flux and retentions, but it improved the permeability of the membranes significantly.

Oxidation rates of carbon and nitrogen in char residues from solid fuels

Computational fluid dynamics (CFD) modeling is an important tool in designing new combustion systems. By using CFD modeling, entire combustion systems can be modeled and the emissions and the performance can be predicted. CFD modeling can also be used to develop new and better combustion systems from an economical and environmental point of view. In CFD modeling of solid fuel combustion, the combustible fuel is generally treated as single fuel particles. One of the limitations with the CFD modeling concerns the sub-models describing the combustion of single fuel particles. Available models in the scientific literature are in many cases not suitable as submodels for CFD modeling since they depend on a large number of input parameters and are computationally heavy. In this thesis CFD-applicable models are developed for the combustion of single fuel particles. The single particle models can be used to improve the combustion performance in various combustion devices or develop completely new technologies. The investigated fields are oxidation of carbon (C) and nitrogen (N) in char residues from solid fuels. Modeled char-C oxidation rates are compared to experimental oxidation rates for a large number of pulverized solid fuel chars under relevant combustion conditions. The experiments have been performed in an isothermal plug flow reactor operating at 1123-1673 K and 3-15 vol.% O2. In the single particle model, the char oxidation is based on apparent kinetics and depends on three fuel specific parameters: apparent pre-exponential factor, apparent activation energy, and apparent reaction order. The single particle model can be incorporated as a sub-model into a CFD code. The results show that the modeled char oxidation rates are in good agreement with experimental char oxidation rates up to around 70% of burnout. Moreover, the results show that the activation energy and the reaction order can be assumed to be constant for a large number of bituminous coal chars under conditions limited by the combined effects of chemical kinetics and pore diffusion. Based on this, a new model based on only one fuel specific parameter is developed (Paper III). The results also show that reaction orders of bituminous coal chars and anthracite chars differ under similar conditions (Paper I and Paper II); reaction orders of bituminous coal chars were found to be one, while reaction orders of anthracite chars were determined to be zero. This difference in reaction orders has not previously been observed in the literature and should be considered in future char oxidation models. One of the most frequently used comprehensive char oxidation models could not explain the difference in the reaction orders. In the thesis (Paper II), a modification to the model is suggested in order to explain the difference in reaction orders between anthracite chars and bituminous coal chars. Two single particle models are also developed for the NO formation and reduction during the oxidation of single biomass char particles. In the models the char-N is assumed to be oxidized to NO and the NO is partly reduced inside the particle. The first model (Paper IV) is based on the concentration gradients of NO inside and outside the particle and the second model is simplified to such an extent that it is based on apparent kinetics and can be incorporated as a sub-model into a CFD code (Paper V). Modeled NO release rates from both models were in good agreement with experimental measurements from a single particle reactor of quartz glass operating at 1173-1323 K and 3-19 vol.% O2. In the future, the models can be used to reduce NO emissions in new combustion systems.