Cette thèse porte sur l'étude de l'enthalpie de dissolution du dioxyde de carbone dans des solutions aqueuses d'amine. Pour développer des modèles théoriques décrivant les systèmes (CO2-amine-eau) pour les conditions appliquées aux procédés industriels, il est nécessaire d'avoir des données expérimentales reliant la solubilité et l'enthalpie. Dans cette étude, nous avons utilisé une unité de mélange construite au laboratoire que nous avons adapté à un calorimètre SETARAM C-80 pour mesurer l'enthalpie de solution du CO2 dans cinq solutions aqueuses d'amine, (la 2-Amino-2-méthyl-1-propanol (AMP), la monoéthanolamine (MEA), la diéthanolamine (DEA), la triéthanolamine (TEA) et la méthyldiéthanolamine (MDEA) (15 et 30 mass%) à des températures comprises entre 322.5 K et 372.9 K et des pressions allant de 0.5 à 5 MPa. Les données de la littérature ont été utilisées pour ajuster deux modèles thermodynamiques d'équilibre de phases (un simple et un rigoureux). Le premier modèle résume l'absorption du CO2 par une seule réaction, tandis que le second prend en compte toutes les réactions à l'équilibre. Le modèle simple reproduit nos enthalpies expérimentales à plus ou moins 10%, tandis que le modèle rigoureux reproduit nos données avec un écart compris entre 5 et 20% selon l'amine considérée. Le calcul de l'enthalpie dans le modèle rigoureux est fortement dépendant des données de la littérature utilisées pour la réaction de protonation de l'amine. Ceci souligne la nécessité d'acquérir de nouvelles données expérimentales sur ces constantes d'équilibre pour améliorer le modèle.

Στην παρούσα μεταπτυχιακή εργασία περιγράφεται η διαδικασία συνδυασμού προβλημάτων Επιχειρησιακής Έρευνας με την μεθοδολογία εύρεσης συγκριτικής αποδοτικότητας (DEA). Αρχικά, παρουσιάζεται μια γενική περιγραφή της μεθόδου DEA και μια συνοπτική επισκόπηση της σχετικής βιβλιογραφίας. Παρουσιάζεται ο τρόπος συνδυασμού της μεθόδου DEA και δύο κλασσικών μοντέλων χωροθέτησης εγκαταστάσεων, του μοντέλου με περιορισμό και του αντίστοιχου μοντέλου χωρίς περιορισμό στην χωρητικότητα. Για την επίτευξη αυτού του στόχου γίνονται οι απαραίτητοι χειρισμοί στην μέθοδο DEA ούτως ώστε να μπορεί να υπολογίζεται η αποδοτικότητα για όλες τις μονάδες λήψης απόφασης ταυτόχρονα – μέθοδος ταυτόχρονης DEA (Simultaneous DEA), εφόσον το κλασσικό μοντέλο βρίσκει την αποδοτικότητα μιας μονάδας λύνοντας μια φορά το γραμμικό πρόβλημα με τους συντελεστές βαρύτητας αυτής της μονάδας. Η λύση του πολυκριτήριου προβλήματος αναδεικνύει την αλληλεπίδραση μεταξύ κόστους και αποδοτικότητας, για τη λήψη απόφασης ανάλογα με τις ανάγκες που μπορεί ενυπάρχουν σε ένα αντίστοιχο πραγματικό πρόβλημα. Στην συνέχεια αναπτύσσεται για πρώτη φορά στη διεθνή βιβλιογραφία μια μεθοδολογία για το συνδυασμό δύο άλλων βασικών προβλημάτων, της κάλυψης και της σύμπτυξης συνόλου, αντίστοιχα, με την μεθοδολογία DEA. Στόχος είναι να μορφοποιηθεί ένα μοντέλο γραμμικού προγραμματισμού έτσι ώστε εκτός από το μέτρο απόφασης του κόστους για την κάλυψη ή σύμπτυξη ενός συνόλου-στόχου, από διαθέσιμα υποσύνολα να ληφθεί υπόψη και η αποδοτικότητα του εκάστοτε υποσυνόλου, η οποία εν τέλει θα επηρεάσει και την συνολική αποδοτικότητα του συνόλου-στόχου. Γίνεται ο συνδυασμός των μεθοδολογιών και αναπτύσσονται μεθοδολογίες πολυκριτήριας ανάλυσης που μπορούν να χρησιμοποιηθούν για την λήψη αποφάσεων που αφορούν την αποδοτική και οικονομική κάλυψη ή σύμπτυξη ενός συνόλου. Για την πιστοποίηση και τη διαπίστωση της λειτουργικότητας των προτεινόμενων μεθοδολογιών αναπτύσσονται παραδείγματα προβλημάτων, τα οποία και επιλύονται επιτυχώς. / In the present thesis, the combination of Operation Research Problems with the Data Envelopment Analysis (DEA) is performed in order to make optimal and efficient decisions. Firstly, a general description of DEA and a breath literature review is presented. Then, we show and test location modeling formulations that utilize data envelopment analysis (DEA) efficiency measures to find optimal and efficient facility location/allocation patterns. In addition, to the authors’ best knowledge, the combinations of DEA with the Set Covering Problem as well as Set Packing Problem are formulated as multiobjective problems, for first time in the literature. The main aim of the proposed models is to make cost-effective and efficient decisions regarding the Set Covering and Packing Problem, respectively. Numerical examples are developed in order to validate and test the novel models. The numerical results of multiobjective analysis demonstrate that the proposed methods are able to successfully find optimal and efficient solutions for real set covering, packing and partitioning problems.

No mundo todo busca-se a melhoria da qualidade das universidades através de avaliações periódicas. Também no Brasil tem sido realizado um grande esforço para a avaliação de universidades, especialmente as públicas. A departamentalização, vigente desde 1968, teve como conseqüência a separação entre o sistema administrativo e o acadêmico, com diferentes instâncias decisórias, diminuindo a integração e a flexibilidade da universidade. Essa situação vai de encontro ao que a sociedade atual está demandando: uma universidade dinâmica, que acompanhe as tendências tecnológicas, com alto nível de qualidade. Nesse contexto é que surge o interesse pela avaliação de departamentos de uma universidade, para verificar se estes, funcionando de uma maneira distinta, apresentam desempenhos similares. Mais especificamente, procura-se, neste trabalho: (1) definir, qualitativamente, os fatores relevantes para uma avaliação quantitativa multicriterial de departamentos acadêmicos de uma universidade; (2) efetuar comparação objetiva, via modelagem matemática, do desempenho dos departamentos; (3) diferenciar departamentos quanto à sua eficiência relativa; e (4) identificar os fatores que fazem departamentos serem mais (ou menos) eficientes. Focaliza-se a Universidade Federal do Rio Grande do Sul (UFRGS), de grande relevância no cenário nacional, com cursos de elevada qualidade, embora nem todos com o mesmo nível de excelência. O estudo emprega uma modelagem matemática, utilizando a técnica denominada Análise Envoltória de Dados (DEA, do inglês “Data Envelopment Analysis”). A aplicação da técnica aos departamentos da UFRGS oferece um mapeamento por eficiência relativa que, em vez de meramente criar uma escala ordinal (ranking), mostra o quanto um departamento é menos eficiente do que outro, em uma escala relacional, indicando, ainda, quais são os fatores que determinam eficiências distintas para os departamentos. Os resultados do estudo permitem: (1) destacar alguns pontos de ineficiência, que podem ser melhorados; (2) identificar características de departamentos mais eficientes, que possam ser repassadas aos menos eficientes, para que estes elevem seu desempenho; e (3) contribuir para a homogeneização da eficiência de todos os departamentos acadêmicos da UFRGS, com o intuito de aperfeiçoar a universidade como um todo.

No mundo todo busca-se a melhoria da qualidade das universidades através de avaliações periódicas. Também no Brasil tem sido realizado um grande esforço para a avaliação de universidades, especialmente as públicas. A departamentalização, vigente desde 1968, teve como conseqüência a separação entre o sistema administrativo e o acadêmico, com diferentes instâncias decisórias, diminuindo a integração e a flexibilidade da universidade. Essa situação vai de encontro ao que a sociedade atual está demandando: uma universidade dinâmica, que acompanhe as tendências tecnológicas, com alto nível de qualidade. Nesse contexto é que surge o interesse pela avaliação de departamentos de uma universidade, para verificar se estes, funcionando de uma maneira distinta, apresentam desempenhos similares. Mais especificamente, procura-se, neste trabalho: (1) definir, qualitativamente, os fatores relevantes para uma avaliação quantitativa multicriterial de departamentos acadêmicos de uma universidade; (2) efetuar comparação objetiva, via modelagem matemática, do desempenho dos departamentos; (3) diferenciar departamentos quanto à sua eficiência relativa; e (4) identificar os fatores que fazem departamentos serem mais (ou menos) eficientes. Focaliza-se a Universidade Federal do Rio Grande do Sul (UFRGS), de grande relevância no cenário nacional, com cursos de elevada qualidade, embora nem todos com o mesmo nível de excelência. O estudo emprega uma modelagem matemática, utilizando a técnica denominada Análise Envoltória de Dados (DEA, do inglês “Data Envelopment Analysis”). A aplicação da técnica aos departamentos da UFRGS oferece um mapeamento por eficiência relativa que, em vez de meramente criar uma escala ordinal (ranking), mostra o quanto um departamento é menos eficiente do que outro, em uma escala relacional, indicando, ainda, quais são os fatores que determinam eficiências distintas para os departamentos. Os resultados do estudo permitem: (1) destacar alguns pontos de ineficiência, que podem ser melhorados; (2) identificar características de departamentos mais eficientes, que possam ser repassadas aos menos eficientes, para que estes elevem seu desempenho; e (3) contribuir para a homogeneização da eficiência de todos os departamentos acadêmicos da UFRGS, com o intuito de aperfeiçoar a universidade como um todo.

No mundo todo busca-se a melhoria da qualidade das universidades através de avaliações periódicas. Também no Brasil tem sido realizado um grande esforço para a avaliação de universidades, especialmente as públicas. A departamentalização, vigente desde 1968, teve como conseqüência a separação entre o sistema administrativo e o acadêmico, com diferentes instâncias decisórias, diminuindo a integração e a flexibilidade da universidade. Essa situação vai de encontro ao que a sociedade atual está demandando: uma universidade dinâmica, que acompanhe as tendências tecnológicas, com alto nível de qualidade. Nesse contexto é que surge o interesse pela avaliação de departamentos de uma universidade, para verificar se estes, funcionando de uma maneira distinta, apresentam desempenhos similares. Mais especificamente, procura-se, neste trabalho: (1) definir, qualitativamente, os fatores relevantes para uma avaliação quantitativa multicriterial de departamentos acadêmicos de uma universidade; (2) efetuar comparação objetiva, via modelagem matemática, do desempenho dos departamentos; (3) diferenciar departamentos quanto à sua eficiência relativa; e (4) identificar os fatores que fazem departamentos serem mais (ou menos) eficientes. Focaliza-se a Universidade Federal do Rio Grande do Sul (UFRGS), de grande relevância no cenário nacional, com cursos de elevada qualidade, embora nem todos com o mesmo nível de excelência. O estudo emprega uma modelagem matemática, utilizando a técnica denominada Análise Envoltória de Dados (DEA, do inglês “Data Envelopment Analysis”). A aplicação da técnica aos departamentos da UFRGS oferece um mapeamento por eficiência relativa que, em vez de meramente criar uma escala ordinal (ranking), mostra o quanto um departamento é menos eficiente do que outro, em uma escala relacional, indicando, ainda, quais são os fatores que determinam eficiências distintas para os departamentos. Os resultados do estudo permitem: (1) destacar alguns pontos de ineficiência, que podem ser melhorados; (2) identificar características de departamentos mais eficientes, que possam ser repassadas aos menos eficientes, para que estes elevem seu desempenho; e (3) contribuir para a homogeneização da eficiência de todos os departamentos acadêmicos da UFRGS, com o intuito de aperfeiçoar a universidade como um todo.

Made available in DSpace on 2017-07-10T18:33:35Z (GMT). No. of bitstreams: 1 Alex Donega.pdf: 1640689 bytes, checksum: c3b70e0c374b81d46b9b19626c96910e (MD5) Previous issue date: 2016-09-05 / : Agriculture is considered an economic activity which contributes significantly to the production of part of the wealth earned in Parana state. In this way, the research on the efficiency of agriculture activity from the point of view of production costs becomes important, in order to know which costs can be reduced, to determine the possible constraints of inefficiency and to indicate the ways to increase efficiency, through the presentation of efficient and inefficient DMUs. Operations with machines, inputs, post-harvest, financial expenses and factor incomes are the input variables, and the final cost price of the sack is the output variable. The DEA model used is the output oriented BBC. The results show that 55.56% of the soybean crop, 38.89% of the wheat crop and 37.5% of the corn crop presented problems of efficiency in production costs during the studied period. It is possible to infer at least four possible explanations for the high level of inefficiency, namely: 1) Direct influence of the exchange rate on the price of herbicides, fertilizers and seeds (input); 2) Deficit in the static capacity of warehouses in Parana state; 3) Ratio between the variable "Financial Expenses" and the floating interest rate (SELIC rate); 4) Decrease in the sack price of the three commodities which coincides with a decrease in the efficiency of the crops. / A agrícultura é considerada uma atividade econômica que contribui significativamente para a produção de parte da riqueza auferida no Estado do Paraná. Desta forma, a investigação sobre a eficiência desta atividade sob a óptica dos custos de produção torna-se importante, a fim de se conhecer quais custos podem ser reduzidos, determinar os possíveis condicionantes da ineficiência e indicar os caminhos para o aumento da eficiência, por meio da apresentação de DMU s eficientes e ineficientes. As operações com máquinas, insumos, pós-colheita, despesas financeiras e rendas de fatores são as variáveis inputs e o preço final de custo da saca é a variável output. O modelo DEA empregado é o BBC com orientação por output. Os resultados remetem a interpretação de que 55,56% da safra da soja, 38,89% da safra do trigo e 37,5% da safra do milho dos anos pesquisados demonstraram problemas de eficiência nos custos de produção. É possível inferir ao menos quatro possíveis explicações para o alto nível de ineficiência, à saber: 1) Influência direta da taxa de câmbio no preço dos herbicidas, fertilizantes e sementes (input insumos); 2) Déficit na capacidade estática dos armazéns no Estado do Paraná; 3) Relação existente entre a variável "Despesa Financeira" e a taxa SELIC; 4) Queda do preço da saca das três commodities que coincide com uma queda na eficiência das safras.

The rapidly evolving global market scenario raised multiple challenges for an organization such as: change in customer needs and lifestyle, increased competition, compulsion to enter into new markets, pursue to innovate and so on, which raises an additional challenge for organization to sustain and succeed. In order to meet these multiple challenges, continuous New Product Development (NPD) turns out to be one of the essential tasks for any organization to improve market share, profitability and to succeed. In this scenario, a new product portfolio with best mix of new projects that ensures strategic alignment, balance of portfolio and improves organizations’ potential gain is compulsion. However, From the literature, it is observed that, ‘As nearly half of initial NPD ideas occur informally or without a specific goal, even a best performing organization requires a major improvement in the decision making process of Project Evaluation and Selection (PES)’. This emphasizes the significance of decision on Project Evaluation and Selection (PES) of NPD. Additionally, huge investments and resources need to be employed based on decision that is taken at PES phase of NPD. Thus PES turns out to be a crucial and essential phase of New Product Process (NPP). All these stated aspects of this challenging and crucial strategic decision of PES provoke for the requirement of an efficient management system and decision making model. In the literature the management system and decision making processes for formulation of portfolio is termed as “New Product Portfolio Management (NPPM)”. Though various researchers have been focusing on this particular issue of improving NPPM Performance, from the analysis of literature, to the best of our knowledge, it is observed that no one has identified or considered an exhaustive list of possible evaluative dimensions while taking the decision on PES of NPPM (PES-NPPM). This thesis makes an attempt to address this research gap, and the scope of this study is pertained to three sectors of manufacturing industry, namely, Automotive, Electronics and Machine Tools. Accordingly, the main objective of this thesis is “ In order to achieve this particular objective the following sub-objectives, methodologies, and analysis are carried out. For this purpose, first and foremost analysis of literature on PES is carried out. Accordingly, five evaluative dimensions are identified for PES-NPPM and they are: (i) Strategic Fit; (ii) Portfolio-Innovation Balance; (iii) Risk-Uncertainty Estimation; (iv) Cost-Revenue Estimation and (v) Optimized Resource Allocation. Furthermore, it is observed that, there is no study considering all the five evaluative dimensions simultaneously for PES-NPPM either to analyze their impact on performance of NPPM or to develop a decision making model. Thus, we are addressing a new problem configuration in the area of PES-NPPM. Additionally, though the requirements of multi-criteria models for PES-NPPM is discussed both in academic and practioners points of view, the real demonstration of the applicability of multi-criteria models are given a scant treatment in the literature. . By the end of the achieving this objective, we identified five distinct evaluative dimensions which are used in different combinations for PES-NPPM. Further, for measuring each of these five evaluative dimensions, we identified 23, 11, 15, 10, and 18 measurement variables respectively. Based on the evaluative dimensions considered in this study, a framework work is proposed for PES-NPPM. Due to the limitation of empirical evidences on considering the identified evaluative dimensions and respective measurement variables towards the proposed initial framework for PES-NPPM, another exploratory study: a case study method is carried out. In addition to the process of triangulation, the case study approach is carried out to understand (a) significance and nature of the identified measurement variables of all the five evaluative dimensions for PES-NPPM, and (b) real-life practices in decision making process of PES-NPPM and to identify the requirements of decision making tools. Accordingly, 12 case studies (4 each) from three manufacturing sectors, considered in this study, are conducted. Further, 12 case study reports are prepared and inferences are drawn. The inferences drawn are verified by conducting an individual brain-storming session with 3 academicians and 4 practitioners. The detailed analysis of the 12 case study reports endorsed the necessity of considering all the five identified evaluative dimensions in the proposed framework for PES-NPPM. In addition, the case study analysis revealed some of the variables originally considered for measuring the evaluative dimensions are not really the measurement variables, whereas those variables are expected to impact the decision making environment of PES-NPPM (or) NPPM Performance. Further those non-measurement variables are classified into (a) Characteristic Variables of PES-NPPM, and (b) Moderating variables for NPPM. Based on this, case study analysis identified 8 characteristic variables and 8 moderating variables. This specific observation resulted to analyze further the existing literature in order to identify if there exist any additional variables which impact decision making environment of PES-NPPM (or) NPPM Performance. Thus, from the analysis of literature and case study analysis 17 characteristic variable and 13 moderating variables are identified. Additionally, For this purpose, Partial Least Square – Path Modeling (PLS-PM) (or) regression analysis is conducted depending upon type of variables with 104 observations (representing 34, 39, and 31 observations of the three sectors respectively) to analyze the relationships between characteristic and moderating variables on decision-making environment of PES-NPPM and NPPM Performance respectively. From case study analysis, it is observed that the decision making tool required should provide: (a) ability to incorporate judgmental scores along with financial and other quantitative metrics, (b) ability to attain a balance of portfolio and consider interactions among project, and (c) ability to provide alternatives and rank the alternatives. In addition to the observation drawn from the case study analysis on the need of MCDM based tool(s), analysis of the literature is carried out to verify the same. As this problem scenario considers both quantitative and qualitative data for the development of a decision making tool, an appropriate technique/methodology needs to be employed. Based on analysis of literature and the case study reports, this study proposes an integrated Data Envelopment Analysis and Balanced Scorecard (DEA-BSC) model for individual PES. Further, the proposed DEA-BSC model is extended for evaluation of new product portfolio. In the process of formulation of new product portfolio, first, every new product project is evaluated with the proposed integrated DEA-BSC model. Second, an algorithm is designed to generate alternate portfolios with the selected set of efficient new product projects. Then, DEA-BSC model is employed to evaluate the generated portfolios. At this step, an accumulation functions are proposed which considers interactions among projects. These accumulation functions determine the overall input and output of the portfolio along with interactions involved. Accordingly, the proposed integrated DEA-BSC model for portfolio evaluation is expected to result in a balanced portfolio with profitable new product projects. In addition, the workability of the proposed integrated DEA-BSC model is demonstrated by developing a suitable numerical example. Finally, a sensitivity analysis is carried out on proposed DEA-BSC model to analyze the robustness of the results. In summary, this thesis examined a problem of decision making of NPPM. Further, this problem was retained with main focus on PES phase. Accordingly, the major contributions of this thesis are as follows: Identified an exhaustive lists of evaluative dimensions: (i) Strategic Fit; (ii) Portfolio-Innovation Balance; (iii) Risk-Uncertainty Estimation; (iv) Cost-Revenue Estimation and (v) Optimized Resource Allocation. Also identified the significance of these five dimensions in case of PES-NPPM. In addition, all the five evaluative dimensions are considered simultaneously for development of a multi criteria decision making tool for PES-NPPM. Identified the required measurement variables for each of the evaluative dimensions, considered in this study, that are essential for PES, and analyzed their influence on performance of NPPM. Identified and analyzed characteristic and moderating variables that influence decision making environment of PES-NPPM and performance of NPPM respectively. Identified the requirements of a decision making tool for PES-NPPM and developed an integrated DEA-BSC model for PES. To the best of our knowledge, the proposed integrated DEA-BSC model is considered to be the first hybrid model applied to PES-NPPM. Furthermore for implementing the proposed DEA-BSC model, an algorithm is proposed in this study and this is expected to assist decision maker for selecting the right set of projects for new product portfolio with higher development potential, profitability and minimize the associated risk. Identified possible project interactions caused due to external or internal factors and accordingly proposed an accumulation function to capture these interactions. Proposed an algorithm for formulation of new product portfolio and accordingly proposed a detail step-by-step procedure for implementation of the proposed integrated DEA-BSC model. Though this study analyzes the impact of characteristic variables on decision-making environment of PES-NPPM, we limit to link this impact to DEA-BSC Model. In this study, an attempt is made to capture the moderating effect on NPPM Performance, but this study limits to link this moderating effect in proposed DEA-BSC model. Finally, the validation of the workability of proposed DEA-BSC model is limited to the numerical example considered in the study and not to the real-life problems scenarios.

La prise en compte des problèmes environnementaux dans la responsabilité sociale des entreprises a généré en économie de nombreuses propositions. Parmi elles, le cadre d’analyse basé sur l’évaluation de la performance en utilisant notamment les techniques d’enveloppement des données (DEA) s’est très vite répandu dans la littérature théorique comme empirique. Ce travail de thèse s’inscrit dans cette logique en mettant l’accent sur la modélisation des technologies polluantes. Par ailleurs, la question des changements climatiques et de la forte contribution de l’agriculture et en particulier de l’élevage dans les émissions de gaz à effet de serre (GES) impose à ce secteur de relever aujourd’hui en plus du défi économique celui de l’amélioration de sa performance environnementale. L’objectif général de cette recherche doctorale est donc de fournir un nouveau cadre d’analyse théorique et empirique dans la modélisation des technologies polluantes afin d’évaluer l’éco-efficience des systèmes productifs, en particulier le cas des émissions de GES en élevage extensif de ruminants. Dans un premier temps, nous montrons les limites théoriques et méthodologiques des modèles existants. Néanmoins, nous insistons sur le fait que les approches basées sur l’estimation de plusieurs sous-technologies indépendantes pour prendre en compte les différents processus présents dans les systèmes productifs sont très prometteuses. Dès lors dans un deuxième temps, nous proposons une nouvelle extension de la méthode « by-production » qui repose sur l’introduction d’interconnections entre les différentes sous-technologies impliquées afin de construire un système plus unifié. Dans un troisième temps, une comparaison empirique utilisant des données d’exploitations de viande ovine de notre extension avec les approches existantes a révélé certaines incohérences de ces dernières. Enfin pour aller plus loin, nous élargissons dans un quatrième temps notre approche afin de prendre en compte les aspects dynamiques et notamment la présence de coûts d’ajustement. Les résultats de l’analyse empirique entreprise avec des données d’exploitations bovines allaitantes (viande) ont révélé la nécessité de prendre en compte ces aspects, mais ont aussi révélé la forte hétérogénéité existante dans les stratégies d’investissements des éleveurs. / The growing importance of environmental matters in social responsibility of firms has generated many frameworks of analysis in the economic literature. Among those frameworks, performance evaluation and benchmarking using the non-parametric Data Envelopment Analysis (DEA) have increased at a very fast rate. This PhD research focuses on models that include undesirable outputs such as pollution in the overall production system, to appraise eco-efficiency of decision making units (DMUs). Besides, the recent awareness on the large contribution of agriculture and particularly livestock farming to global warming, has highlighted for this sector the challenge of reaching both economic and environmental performances. In this line, the overall objective of this dissertation is to provide a theoretical and empirical background in modelling pollution-generating technologies and to suggest theoretical improvements that are consistent with the particular case of greenhouse gas emissions in extensive livestock systems. Firstly, we showed that all existing approaches that deal with undesirable outputs in the non-parametric analysis (i.e. DEA) have some strong drawbacks. However, the models grounded on the estimation of multiple independent sub-technologies offer interesting opportunities. Secondly, I developed a new framework that extends the by-production approach through the introduction of some explicit dependence constraints that link the sub-technologies in order to build a unified system. Thirdly, an empirical comparison, using a sample of French sheep meat farms, of this by-production modelling extension with the existing approaches, revealed some inconsistencies of these latter. Finally, we expanded this new by-production formulation to account for dynamic aspects related to the presence of adjustment costs. The application to the case of French suckler cow farms underlined the necessity of accounting for dynamic aspects and also showed high heterogeneity in investment strategies of these farmers.

In this thesis, I apply the sophisticated tools made available by the econometrics of panel data and treatment models to a range of different issues. In the first Chapter, an ECM model is used to test on the existence of financing constraints in firms’ investment and R&D, taken a proxy for the efficiency of market institutions and governance rules in different countries. In the second chapter we test an agency model linking pay-performance contracts of CEOS to the financial situation of a firm by using a UK panel data. In the third chapter I use a sophisticated treatment model to evaluate the effectiveness of Italian public subsidies to R&D. Finally, in the fourth chapter I try to evaluate the efficiency of Italian regional systems of public healthcare by controlling for socio-economic factors and quality of healthcare in a composite model using panel data estimation and efficient frontier techniques.<p>The first Chapter analyzes the investment behavior of a sample of R&D intensive firms which are quoted on the stock market from USA, UK and Japan for the period 1990-1998. By using an error correction model we test the elasticity of investment and R&D to cash flow in these countries to see by which measure different market institutions and corporate governance rules affects the cost of external financing. Contrary to previous studies, we find significant differences in the sensitivity to cash flow of the two types of investment, with R&D expenditure being much less sensitive than ordinary investment. This is not surprising given the more long-term nature of R&D expenditures. For what concerns the comparison between the different systems/countries, the USA stock markets confirms as the most efficient market providing outside financing at a much lower cost compared to other markets, especially for young, smaller firms.<p>The second Chapter is a joint work with Biagio Speciale. It uses the data on a panel of quoted UK firms over the period 1995–2002 to study the effects of financial leverage on managerial compensation. The change in the investors’ expectations that caused the recent collapse of the stock market tech bubble is a perfect example of natural experiment that has been used as a source of plausibly exogenous variation in the firm’s debt. The estimates show that pay-for-performance sensitivity is increasing in financial leverage, with the exception of the 10% most levered firms, giving rise at the end to a non-linear (inverted U-shape) relationship between the two variables. The chapter includes also a theoretical model accounting for this relationship where an higher leverage increases both the expected returns and the expected variance of investment returns: the first effect (determining increased pay-performance sensitivity) prevails for low leverage values and the second effect (determining decreased pay-performance sensitivity) prevails for high leverage values.<p>The third Chapter undertakes an empirical estimation of the additionality of public funding on both the propensity to initiate R&D activity and the intensity of R&D spending of Italian enterprises for the period 1998-2000, using data from the Third Community Innovation Survey and from firms' financial accounts. The chosen methodology (Endogenous Switching Type II-Tobit) takes into account the possibility that decisions about both starting an R&D activity (sample selection effect) and applying for/obtaining public funding (essential heterogeneity) are influenced by private knowledge of enterprises' idiosyncratic propensities in R&D spending. The present analysis shows that both these effects are indeed important and that they contribute to explain most of the additionality found with less sophisticated models.<p>The fourth Chapter investigates the underlying causes of variability of public health expenditure per capita (SSPC henceforth) between Italian regions. A fixed-effect panel data estimate on the SSPC (for the period 1997-2006) is used in the first part of the paper to account for regional differences in terms of physical, demographic, socio-economic characteristics and in terms of other variables that affect demand and supply of health services. In the second part, we take the ‘adjusted’ SSPC and proceed to estimate an "efficient production function" of the quality of health services through Data Envelopment Analysis. This procedure allows us to separate the share of expenditure used for the improvement of the quality from the one that can be traced only to an inefficient use of financial resources. A comparison of regional SSPC after factoring out the socio-economic factors and the quality of healthcare shows that big differences still remain and are even exacerbated, signalling big pockets of inefficiency and correspondingly a huge potential for cost savings. Finally, a preliminary analysis shows a positive correlation between the efficiency of regional public spending in healthcare and the level of social capital. / Doctorat en Sciences économiques et de gestion / info:eu-repo/semantics/nonPublished

Today, with various technology and business standards, organizations face rapid changes in both internal and external environments. To be able to rapidly respond to such changing environments, integration of software systems has entered among the top priorities of many organizations. However, despite extended use of software integration, methods for estimating the business value of implementing such integration are still missing. Besides presenting a conceptual model to define the benefits of systems integration and related causal relationships, this study proposes a method for measuring such benefits in monetary terms. In particular, we  demonstrate how a mathematical programming technique called Data Envelopment Analysis (DEA) can be used to evaluate the business value of software integration. Our method is illustrated using data from 12 organizations. The results indicate significant productivity gains by integrating software systems, which represent the value of software integration in organizations.