abstract: Supply chains are increasingly complex as companies branch out into newer products and markets. In many cases, multiple products with moderate differences in performance and price compete for the same unit of demand. Simultaneous occurrences of multiple scenarios (competitive, disruptive, regulatory, economic, etc.), coupled with business decisions (pricing, product introduction, etc.) can drastically change demand structures within a short period of time. Furthermore, product obsolescence and cannibalization are real concerns due to short product life cycles. Analytical tools that can handle this complexity are important to quantify the impact of business scenarios/decisions on supply chain performance. Traditional analysis methods struggle in this environment of large, complex datasets with hundreds of features becoming the norm in supply chains. We present an empirical analysis framework termed Scenario Trees that provides a novel representation for impulse and delayed scenario events and a direction for modeling multivariate constrained responses. Amongst potential learners, supervised learners and feature extraction strategies based on tree-based ensembles are employed to extract the most impactful scenarios and predict their outcome on metrics at different product hierarchies. These models are able to provide accurate predictions in modeling environments characterized by incomplete datasets due to product substitution, missing values, outliers, redundant features, mixed variables and nonlinear interaction effects. Graphical model summaries are generated to aid model understanding. Models in complex environments benefit from feature selection methods that extract non-redundant feature subsets from the data. Additional model simplification can be achieved by extracting specific levels/values that contribute to variable importance. We propose and evaluate new analytical methods to address this problem of feature value selection and study their comparative performance using simulated datasets. We show that supply chain surveillance can be structured as a feature value selection problem. For situations such as new product introduction, a bottom-up approach to scenario analysis is designed using an agent-based simulation and data mining framework. This simulation engine envelopes utility theory, discrete choice models and diffusion theory and acts as a test bed for enacting different business scenarios. We demonstrate the use of machine learning algorithms to analyze scenarios and generate graphical summaries to aid decision making. / Dissertation/Thesis / Ph.D. Industrial Engineering 2012

Globally, Atlantic salmon aquaculture is faced with a critical challenge: How best to deliver long-term sustainable growth, whilst optimising the opportunity for the expansion of the industry presented by an increasing global seafood demand? The thesis presents a novel framework of complementary decision support approaches to enable decision-makers to better understand the factors influencing aquaculture development, and examine alternative production (growout) technologies that more effectively address the challenges associated with intensification and expansion. The framework was developed through a combination of fieldwork (international data-gathering), key stakeholder discussions, and the application of targeted qualitative and quantitative analytical approaches; using the Tasmanian industry as a Case Study. The initial research focused on shorter-term (tactical) decision support. A situational analysis defined the business environment, and appraised viable expansion options (offshore, closed-containment and extractive bio-remediation). An economic analysis of selected options then provided a comparison of financial performance and risk. The outputs of this initial component next informed strategic decision-making approaches; employing scenario analysis to explore plausible strategies for the adoption of land-based recirculating aquaculture systems; and qualitative modelling to understand the causal dynamics driving and regulating the industry, and their impact on technology selection. Whilst it was clear that business economic viability is paramount, the results suggested that societal acceptance (the Social License to operate) is playing an increasingly important role in influencing business decisions. There is no single ‘right' technological solution; social acceptance, in particular considerations regarding human wellbeing, trust, and animal welfare concerns, will shape the business environment and therefore technology selection. The research emphasised the importance of employing a balance of tactical and strategic decision-making techniques, and of engaging with a broad range of industry stakeholders. It also highlighted the complexity and dynamic nature of the industry and that key variances (economic, regional, strategic, technological, and temporal) must be included in decision-making.

This study aims to optimize the logistics processes of container terminals. Potentially powerful pipe-flow models of container terminal logistics processes have been neglected to date and modelling of terminals is rare. Because research which adopts a pipe flow and dynamic operational perspective is rare, a case application in Alexandria, Egypt collated empirical container and information flows using interviews and company records to describe its logistics processes and model container and information flows. The methodology used includes qualitative and quantitative methods and a descriptive methodology proceeds sequentially. Primary and secondary data were presented as a pipe flow model to show interrelations between the company’s resources and to identify bottlenecks. Simulation modelling used Simul8 software. Operational level modelling of both import and export flows simulated the actual inbound and outbound flows of containers from entry to exit. The import logistics process includes activities such as unloading vessels by quay cranes, moving containers by tractors to yard cranes to go for storage where customs procedures take place before exiting the terminal by customer’s truck. The export logistics process includes the activities associated with customers’ trucks, lifters, storage yards, tractors and quay cranes. The model takes into account the uncertainties in each activity. This study focuses on operational aspects rather than cost issues, and considers container flows rather than vessel flows. Although the simulated model was not generalized, implementation elsewhere is possible. Following successful validation of a base simulation model which reproduces the case company’s historical scenario, scenario testing empowered the case company to pro-actively design and test the impact of operational changes on the entire logistics process. The study evaluates a typical container terminal logistics system including both import and export containers in the presence of multiple uncertainties in terminal operations (e.g. quay crane operations, tractor operations, yard crane operations). Sensitivity testing and scenario analysis can empower terminal managers to make decisions to improve performance, and to guide terminal planners, managers, and operators in testing future investment scenarios before implementation.

In the UK, the debate surrounding energy production lies at the forefront of the political agenda, with growing emphasis on achieving an increasingly sustainable energy mix into the future. The nuclear option is especially debatable - issues such as waste management and decommissioning receive much attention. In addition, the many stakeholders interested in nuclear power display very divergent views on its sustainability. Since the turn of the century, nuclear power has received much attention globally, with many nations’ governments taking consideration of the potential benefits of new nuclear adoption. Conversely, the Fukushima nuclear disaster has led to new nuclear resistance in other nations, such as Germany, where plans have been made to stop nuclear power generation completely. This research aims to help inform the debate on nuclear power and the future UK electricity mix. A multi-criteria decision support framework (developed by the SPRIng Project) has been used for these purposes, taking into account technical, economic, environmental and social criteria.The methodology used in this work has involved: stakeholder consultation; use of future electricity scenarios; sustainability assessment of current and future electricity options (Pressurised Water Reactor, European Pressurised Reactor, European Fast Rector, coal, gas, solar and wind power, and coal carbon capture and storage [CCS] power); assessment of future electricity scenarios based on both sustainability impacts and stakeholder (expert and public) preferences for the sustainability indicators and electricity technologies. The sustainability assessment of future nuclear power options and coal CCS power have been carried out here for the first time in a UK-specific context.Based on the public and expert opinions on the importance of different sustainability indicators, results of the scenario analysis suggest that the scenario with a high penetration of low-carbon technologies (nuclear [60%] and offshore wind power [40%]) is the most sustainable. For the sample considered in this study, this finding is not sensitive to different stakeholder and public opinions on the importance of the sustainability indicators. However, when the stakeholder preferences for individual technologies are considered, scenarios with high penetration of renewables (26-40% solar and 20-48% wind) become the preferred options. This is due to the favourable stakeholder opinion on solar and wind power. In that case, the scenario with high penetration of nuclear is never the preferred option due to the low to moderate stakeholder preference for nuclear power.Therefore, the results from this research suggest that the ‘sustainability’ of different electricity options and scenarios is highly dependent on stakeholder preferences and priorities. Thus, for successful future deployment of these options and implementation of energy policy measures, transparency of information on the impacts of electricity options is key in ensuring that stakeholder opinions are founded in the actual rather than the perceived impacts of these options.

The thesis is composed of three main parts. The first part is theoretical and I deal here with economic relationships between macroeconomic magnitudes. Second part dedicated to the econometric theory of prognosis follows, in which I deal with different types of prognoses and prediction methods used at present. In the third, practical, part my intended aim is to create the best possible models of relations between fundamental macroeconomic magnitudes, using real Czech economy data, and to make simulation predictions of these magnitudes based on acquired models while utilising scenario analysis. First, I deal with choice of MSE and VAR models. Then follows the estimate of particular models and validation of prognostic capabilities of particular models for static and dynamic simulation. I conclude with elaboration of macroeconomic magnitudes prognosis while using scenario analysis.

The master thesis is dealing with the contemporary issues of operational risk management in financial institutions. Author sets a theoretical basis and legal background of the topic and describes the contemporary practices of managing the operational risk. Author focuses on the scenario analysis as a specific method which is described and evaluated. Scenario analysis is applied on the rogue trading risk. In the thesis there is created a model institution on which author applies the operational risk theory using best practices and expert opinions. The model situation provides the analysis of the processes of the financial institution and choose the suitable measures in order to defend against the risk. The author also analyses the past cases of rogue trading which helps to understand the prevention and the historical significance of the operational risk.

This thesis is composed of free main chapters. The first two chapters of is a theoretical part. The first chapter is devoted to the theory of economic policy and analysis of economic indicators. The second chapter concerns the econometric theory and describes vector autoregression models theory and econometric forecasting. In the third, practical part, aims to find out with the help of real data of the Czech economy impacts of alternative VAT rates on selected indicators of the czech economy, these indicators are gross domestic product, unemployment rate and consumer price index. As a tool to determine the impact of using models and vector autoregression method scenarios.

This diploma thesis presents results of analysis of monetary policy transmission mechanism in the Czech Republic employing the vector autoregressive (VAR) models. The responsible authority for monetary policy is Czech National Bank that has been using the inflation targeting regime to conduct its monetary policy since 1998. The inflation rate changes, i.e. the changes in repo rate represent a monetary tool for steering actual inflation rate towards the projected or "target" inflation rate. The linear correlation between 2 weeks repo rate and 1 month PRIBOR rate is confirmed. The transmission mechanism is examined within the VAR framework and the relationships between the 1 month PRIBOR rate, gross domestic product and inflation rate are studied. The VAR model including 1 lag is considered as the best performing model. The relationships among variables are analysed by related approaches -- Granger causality, impulse response functions and cointegration. The ability of model to create forecasts is assessed and the ex ante forecasts are produced for one-year horizon. The effects of alternative monetary policies are the subject of scenario analysis.

Les ménages sont responsables d’une part significative des consommations d’énergie et des émissions de CO2, en particulier si l’on tient compte des consommations d’énergie et des émissions indirectes liées aux processus de production des biens et services consommés. Plusieurs travaux scientifiques et recommandations d’organisations gouvernementales et d’associations non-gouvernementales soulignent que des modifications des modes de consommations seront sans doute nécessaires pour atteindre les objectifs climatiques fixés aujourd’hui. Notre thèse propose une méthode d’analyse prospective de changements de mode de vie, qui permet d’en estimer les impacts macro-économiques ainsi que ceux sur les consommations d’énergie et les émissions de CO2, tout en tenant compte de l'hétérogénéité des ménages en matière de comportements et de consommations d'énergie. Notre méthode explore les modes de consommation des ménages de manière fine, en prenant en considération le lien entre emplois du temps et consommations. En effet, les choix de consommation sont soumis non seulement à des contraintes de budget, mais également à des contraintes qui dérivent du temps à disposition et qui ne sont jamais prises en compte dans la prospective macro- économique. Nous construisons une base de données détaillant emplois du temps, dépenses et consommations d’énergie des ménages français et nous l’articulons à un modèle de prospective économique énergie-émissions par un processus de repondération itératif. Nous illustrons la portée de cet outil à travers l’analyse de trois scénarios, centrés respectivement sur la diffusion de nouvelles formes de mobilité (covoiturage et autopartage), la généralisation des achats en ligne et le retour vers le faire soi-même en matière d'alimentation. Pour les trois scénarios nous observons des réductions des consommations d’énergie et des émissions de CO2. Par exemple, les émissions totales diminuent de 2,3% en 2050 dans le scénario sur la mobilité. / Household energy consumption represents a significant share of final energy use, especially when both direct and embodied energy are taken into account. Several academic studies, as well as the recommendations of the United Nations and of non-governmental organisations, suggest that a shift in consumption patterns will be necessary to achieve sustainable development. The aim of our research is to analyse long-term scenarios of changes in lifestyle. We propose a methodology that allows to analyse the macro-economic impacts of these changes, as well as the impacts on energy use and CO2 emissions, while taking into account the heterogeneity of behaviours and energy consumptions among households. Consumption choices do not derive solely from monetary considerations but they are influenced by several factors. One binding constraint, never taken into account in macro-economic energy modelling, is the available time. For this reason, our analysis considers time use data in addition to expenditure and energy use data. We build a data base that combines time use, expenditure and energy consumption data for French households, which provides detailed information about household consumption patterns. Then, for scenario analyses, we link the data base with an Energy-EconomyEmissions model, using an iteration process based on a reweighting technique. We illustrate the methodology by exploring three areas of change in consumption patterns: cooking habits, ecommerce and shared transport (carpooling and car sharing). We obtain CO2 emissions reductions in all scenarios. As an example, emissions decrease by 2.3% by 2050 in the scenario focusing on transport.

With a large amount of tyres being discarded every year, the question of how to manage the end-of-life tyres (ELTs) has become a serious issue. Thus this study identifies different driving forces for this management and the most possible scenarios for the future management of ELTs. The study also compares the business as usual model with a waste hierarchy model to explore the possibilities for optimizing management of ELTs through cascading. This study collects opinions about the driving forces of ELT management from 29 experts working in the area. Important driving forces identified were: price of substitute products, recycled materials’ market, environmental legislation, and technology. This study also surveys 23 experts in the tyre area about the most possible scenarios for ELTs in 2020. One of the more believed in futures was: “Due to increasingly limited fossil fuels and a rise of sustainability awareness, applications for ELTs are growing both in material and energy recycling.” This suggests that a shift toward an equal recycling situation of ELTs among material and energy might be likely to happen by 2020. Based on the most possible scenario for ELTs in 2020, a comparison between waste hierarchy model and business as usual model has been performed. The result shows that the (cascading) waste hierarchy model would likely create more environmental benefits than business as usual model. This is done though the saving and cycling of more materials from energy recovery into material recycling.