Depuis le début des années 1990, l'utilisation des agents mobiles dans le cadre de l'administration système et réseau est étudiée. Des plates-formes à agents mobiles répondant à cet usage ont été développées et se sont penchées en général sur les problèmes de l'interrogation des équipements actifs du réseau, de la supervision des noeuds des plates-formes, de la sécurité des noeuds visités, etc. Cependant, une contrainte opérationnelle provenant du fait que la présence d'éléments dans un réseau tend à évoluer dynamiquement n'avait pas été prise en compte. Une telle évolution requiert que les itinéraires de migration associés aux agents mobiles d'administration ne soient pas définis comme cela est fait en général, de manière statique par l'administrateur, mais qu'au contraire de tels itinéraires puissent être construits à la volée de sorte à refléter la topologie courante du réseau à administrer. Nous avons aussi constaté que les facilités de programmation obtenues par l'association d'une fonction d'administration à chacun des éléments d'un itinéraire, et ce, en fonction du type de l'élément, pourraient être bénéfique, en ce sens que cela pourrait grandement simplifier la définition d'agents mettant en oeuvre de nouvelles fonctions d'administration système et réseau.<br />Ce travail de thèse consiste à proposer la définition d'un mécanisme de fabrication puis d'utilisation d'itinéraires dynamiques pour l'administration système et réseau. Nous validons cette définition en fournissant une plate-forme complète ainsi qu'un cadre de programmation d'agents mobiles d'administration système et réseau. Pour ce faire, nous utilisons et étendons la bibliothèque ProActive pour le calcul parallèle, réparti et mobile. En effet, ProActive offre un cadre qui simplifie grandement la programmation et le déploiement d'agents mobiles en<br />environnement hétérogène et réparti. Le mécanisme d'itinéraires proposé suppose la connaissance de la topologie du réseau et des éléments qui y sont connectés. Nous implantons donc un mécanisme de découverte automatique de la topologie d'un réseau au niveau 2 OSI. Nous définissons ensuite un cadre générique de programmation d'agents mobiles bâtie autour de notre mécanisme d'itinéraires pour effectuer des tâches d'administration système et réseau. Pour valider notre approche et son intégration dans le monde de l'administration système et réseau, nous fournissons des exemples concrets de l'utilisation des agents mobiles d'administration et ce pour une large gamme de configurations réseau.

L'idée générale de la thèse est de proposer une modélisation d'un système d'information adaptable à tout type de projet mettant en présence des acteurs de territoires séparés, de cultures plurielles et de contingences structurelles et conjoncturelles différentes. La finalité sera de mettre en place un véritable « code de communication » autour d'un « projet de coopération » appropriable par l'ensemble des acteurs en intégrant leurs spécificités et celles de leurs environnements. Plus largement, cette thèse est un cadre de recherche permanent à capitalisation qui doit amener à enrichir « l'art projet » notamment en termes de valorisation des flux d'informations et de connaissances. Les deux dernières décennies ont été caractérisées par l‘émergence d'un environnement dont la complexité s'accroît de manière vertigineuse au vu du magma des flux d'informations à traiter. L'impact sur les entreprises, et plus généralement les organisations, a fondamentalement changé leur management et en particulier dans tous les processus « projet ». Au-delà des concepts de globalisation, de mondialisation, de mutation, nombre d'organisations vont chercher à créer de la valeur dans la mise en place de « projet de coopération ». Ces projets présentent des typologies bien particulières où l'information et la connaissance sont aussi bien des matières premières que des produits finis à leur réalisation. Dans cette optique, la notion de « l'environnement projet » devient de plus en plus prégnant d'autant que sa complexité se voient augmenter par une conjugaison de facteurs et d'acteurs qui ne sont pas uniquement issus de « l'environnement de proximité » du projet mais aussi d'un « environnement de connexion territoriale » qui redéfinit ainsi les « enjeux géospaciaux » du projet et qui pourrait se résumer à l'adage suivant : « penser global et agir local ». A travers cette lecture de l'environnement du projet, quel code de communication commun peut-on partager pour permettre de la mise en place d'une « architecture projet » en recherche d'efficience et qui doit amener un effet surgénérateur au projet, c'est-à-dire produire plus de richesses que de ressources consommées. Apporter une réponse à cette problématique, c'est avant tout faire émerger un modèle conceptuel de « proactivité managériale de l'information » autour de la notion de projet qui peut se décliner en différents éléments permettant de positionner précisément le projet dans son environnement en tenant compte « glocalement » de ses facteurs et de ses acteurs. Cette recherche est cadrée sur la genèse des composantes d'un « management proactif informationnel » au travers de retours d'expériences et résultats de construction d'un « modèle conceptuel informationnel projet » issu de mes travaux de recherches commencés en 1993.

<p>The media has responded toward unethical behavior in rainforests during the ‘80’s, triggering</p><p>the still upcoming trend of investigation of ethical considerations, as described by Macfarlane</p><p>(1995). Many researchers are constantly improving the ethical theories and showing</p><p>continued change in the perception of ethics (Svensson & Wood, 2007)</p><p>The purpose of this research is to identify the effect of ethics on the proactive or reactive</p><p>behavior of companies, with a downward tendency provided in the value chain.</p><p>The theoretical framework consists out of a set of theories supporting the goal of identifying</p><p>ethics within a company, applying a new business ethics model and measuring the pro-active</p><p>and re-active behavior of the stakeholder in the value chain. Finally, all the theories used, are</p><p>combined in an integrated theoretical model used to justify the collection of empirical data</p><p>and to give structure to the analysis.</p><p>The empirical data has been gathered through non structured and semi-structured interviews</p><p>with customers, employees within the case company, and a supplier to the case company.</p><p>These findings are complemented with secondary data gathered through websites, annual</p><p>reports, codes of conduct, media articles and others sources.</p><p>The findings of the study showed that the (in the theoretical framework) assumption of a</p><p>downward effect in perception and change in ethical considerations is not as obvious as</p><p>previously thought. Within this case study, the ethical influences come from the case</p><p>company, situated in the middle of the value chain, influencing both the consumer and the</p><p>supplier, in ethical considerations.</p>

<p>The authors of this paper will try to identify what different pressure factors have an influence over the Swedish mechanical pulp and paper industry. Moreover, the authors of this paper will try to identify whether different levels of environmental commitment can be identified within the organization.</p><p>The results show that all of the pressure factors that were identified in the case study, except for regulations and social pressure, are influenced by economic benefits and, therefore, a reactive stance is almost always preferred by the firm when there is a negative correlation between economic performance and environmental performance. Results also indicate a gap in the internal communication of the organization. This is not related to the topic of this paper but might be useful information for the company used in the case study.</p>

<p><strong>Background:</strong> The reasons why this study is conducted is because of the latest recession in the global economy. The current recession has made a lot of companies more aware of its cost, and in order to fit the new harsher economic climate the companies has to be more cost efficient. In order to do so many companies choose to reduce their amount of employees. When this happens in a recession, most of the layoffs are an effect of reactive causes, such as lower incoming orders, and fewer customers. This leads to a problem for the companies that don’t know how these kinds of layoffs will affect the value of the company’s stock. This leads to a question whether there is a significant pattern between layoffs that are a result of reactive reasons and the number of employees that are being laid off, in percentage of the total number of employees of the companies in question?</p><p><strong>Purpose:</strong> The reason for this study is to find out if there is any correlation between how many employees that is laid off, in percentage of the company’s total employees, and if different percentages have different impact in the valuations of the company’s stock.</p><p><strong>Approach:</strong> This study is made in the form of a modified event study and narrows down to the recession during 2008-01-01 to 2009-10-01, and only involves companies that are on trade on the stock market in Stockholm Sweden. The objects that are involved in this study are from the stock markets large and mid-cap sections. The criteria for companies in this study are that they should have made layoffs for reactive reasons in the time period that we are interested in. The variations in the company’s stock are compared to Stockholm’s stock markets OMX-index. The information needed to conduct this study is gathered from press realises and from financial reports made by the companies.</p><p><strong>Result and conclusions:</strong> The study shows that there is a correlation between how large percentage of the company’s total employees that are laid off and the effect of variations in the stock. The correlation is -0,306 but it is not statistical verified. This result was not in line with our hypotheses that we had concluded from earlier research and theories that are in the study. Earlier research on the American stock market shows a stronger connection between large percentage layoffs and how this makes the stock value to decrease.</p>

Background: The reasons why this study is conducted is because of the latest recession in the global economy. The current recession has made a lot of companies more aware of its cost, and in order to fit the new harsher economic climate the companies has to be more cost efficient. In order to do so many companies choose to reduce their amount of employees. When this happens in a recession, most of the layoffs are an effect of reactive causes, such as lower incoming orders, and fewer customers. This leads to a problem for the companies that don’t know how these kinds of layoffs will affect the value of the company’s stock. This leads to a question whether there is a significant pattern between layoffs that are a result of reactive reasons and the number of employees that are being laid off, in percentage of the total number of employees of the companies in question? Purpose: The reason for this study is to find out if there is any correlation between how many employees that is laid off, in percentage of the company’s total employees, and if different percentages have different impact in the valuations of the company’s stock. Approach: This study is made in the form of a modified event study and narrows down to the recession during 2008-01-01 to 2009-10-01, and only involves companies that are on trade on the stock market in Stockholm Sweden. The objects that are involved in this study are from the stock markets large and mid-cap sections. The criteria for companies in this study are that they should have made layoffs for reactive reasons in the time period that we are interested in. The variations in the company’s stock are compared to Stockholm’s stock markets OMX-index. The information needed to conduct this study is gathered from press realises and from financial reports made by the companies. Result and conclusions: The study shows that there is a correlation between how large percentage of the company’s total employees that are laid off and the effect of variations in the stock. The correlation is -0,306 but it is not statistical verified. This result was not in line with our hypotheses that we had concluded from earlier research and theories that are in the study. Earlier research on the American stock market shows a stronger connection between large percentage layoffs and how this makes the stock value to decrease.

The media has responded toward unethical behavior in rainforests during the ‘80’s, triggering the still upcoming trend of investigation of ethical considerations, as described by Macfarlane (1995). Many researchers are constantly improving the ethical theories and showing continued change in the perception of ethics (Svensson & Wood, 2007) The purpose of this research is to identify the effect of ethics on the proactive or reactive behavior of companies, with a downward tendency provided in the value chain. The theoretical framework consists out of a set of theories supporting the goal of identifying ethics within a company, applying a new business ethics model and measuring the pro-active and re-active behavior of the stakeholder in the value chain. Finally, all the theories used, are combined in an integrated theoretical model used to justify the collection of empirical data and to give structure to the analysis. The empirical data has been gathered through non structured and semi-structured interviews with customers, employees within the case company, and a supplier to the case company. These findings are complemented with secondary data gathered through websites, annual reports, codes of conduct, media articles and others sources. The findings of the study showed that the (in the theoretical framework) assumption of a downward effect in perception and change in ethical considerations is not as obvious as previously thought. Within this case study, the ethical influences come from the case company, situated in the middle of the value chain, influencing both the consumer and the supplier, in ethical considerations.

Det är inte längre en fråga om huruvida en organisation kommer att drabbas av en kris, utan det är snarare frågan om när. Det har skett en utveckling inom diskursen för kriskommunikation. Fokus ligger nu vid hur organisationer förebygger och förbereder för dessa extraordinära händelser snarare än att reagera när krisen väl har inträffat. Denna studie syftar till att vidga diskursen inom krishantering i en svensk kontext genom att undersöka svenska myndigheter och deras förmåga att vara proaktiva när det gäller krishantering, det vill säga arbeta redan före en kris. Baserad på Mitroffs ramverk för Crisis Leadership har myndigheterna analyserats genom en kvantitativ och kvalitativ undersökning. Av de 70 myndigheter som kontaktades skickades 42 enkäter tillbaka, vilket resulterade i en svarsfrekvens på 60 %. Resultaten från undersökningen visar att de svenska förvaltningsmyndigheterna inte uppfyller idealet för utveckla en proaktiv krishantering. De fyra faktorerna, kristyper, krismekanismer, krissystem och krisintressenter, är indikatorer för en proaktiv hållning. Dessa var inte representerade i de data som samlades in, utom den tredje faktorn som hade en proaktiv organisationskultur samt beredskapsaktiviteter, till exempel att ha en krisplan och en krisledningsgrupp. Slutsatsen är att svenska förvaltningsmyndigheter behöver utveckla en mer proaktiv inställning till sin krishantering rörande goda relationer med sina intressenter, med en krisportfölj som omfattar olika krisfamiljer, ta fram fler signaldetektorer och strategier för att förhindra krisen, till exempel företagens sociala ansvarsprogram (CSR-program). / There is no longer a question of whether an organization will suffer from a crisis; it is rather the question of when. It has been a shift in the discourse of the crisis communication area. Focus now lies in how to prevent and prepare for these extraordinary events instead of reacting when the crisis already has occured. This study aims to expand the crisis management area in a Swedish context by examining Swedish authorities and their capabilities to be proactive in the case of crisis management. Based of Mi- troffs framework for Crisis Leadership, the authorities capabilities have been analyzed through a quantitative and qualitative survey. From the 70 authorities that were contacted, 42 surveys were returned, resulting in an acceptable 60% response rate. The results from the survey show that the Swedish authorities do not fulfil the ideal for develop- ing a proactive crisis management. The four factors, crisis types, crisis mechanisms, crisis system and crisis stakeholder, are indicators for a proactive stance. These were not represented in the data that was gathered, except the third factor which were having a proactive organizational culture along with preparedness activities such as having a crisis plan and a crisis leadership team. The conclusion is that Swedish authorities needs to develop a more proactive approach to their crisis management regarding good relationships with their stakeholders, having a crisis portfolio that covers different crisis families, implement more signal detectors and strategies to prevent cri- sis, for example corporate social responsibilities program.

In traditional Wireless Sensor network's (WSN's), the sink is the only unconditionally trusted authority. If the sink is not connected to the nodes for a period of time then the network is considered as unattended. In Unattended Wireless Sensor Network (UWSN), a trusted mobile sink visits each node periodically to collect data. This network differs from the traditional multi hop wireless sensor networks where the nodes close to the sink deplete their power earlier than the other nodes. An UWSN can prolong the life time of the network by saving the battery of the nodes and also it can be deployed in environments where it is not practical for the sink to be online all the time. Saving data in the memory of the nodes for a long time causes security problems due to the lack of tamper-resistant hardware. Data collected by the nodes has to be secured until the next visit of the sink. Securing the data from an adversary in UWSN is a challenging task. We present two non-cryptographic algorithms (DS-PADV and DS-RADV) to ensure data survivability in mobile UWSN. The DS-PADV protects against proactive adversary which compromises nodes before identifying its target. DS-RADV makes the network secure against reactive adversary which compromises nodes after identifying the target. We also propose a data authentication scheme against a mobile adversary trying to modify the data. The proposed data authentication scheme uses inexpensive cryptographic primitives and few message exchanges. The proposed solutions are analyzed both mathematically and using simulations proving that the proposed solutions are better than the previous ones in terms of security and communication overhead.

Working memory (WM) capacity – the capacity to maintain and manipulate information in mind – plays an essential role in high-level cognitive functions. An important determinant of WM capacity is the ability to resolve interference of previously encoded but no longer relevant information (proactive interference: PI). Four different mechanisms of PI resolution involving binding and inhibition have been proposed in the literature, although debate continues regarding their role. Braver et al. (2007) introduced an important distinction in the PI resolution literature, proposing two general types of PI control mechanisms that occur at different time points: proactive control (involves preparation in advance of the interference) and reactive control (occurs after interference occurs). This thesis proposed that among these four functions involving binding and inhibition, item inhibition and binding could be involved in proactive control, while familiarity inhibition and episodic inhibition could be involved in reactive control. The question is which mechanism in each pair is indeed involved in proactive control and reactive control respectively, and how these proactive control and reactive control mechanisms work together to resolve PI. In addition, do these mechanisms play a role in the relationship between PI resolution and WM? In an individual differences study, individuals’ ability to resolve PI was assessed in memory tasks, with two versions of each that encouraged the use of either proactive or reactive control. In addition, measures were obtained of individuals’ ability of binding and inhibition in tasks that had minimal memory demands. Regression analyses showed contributions of binding and inhibition to PI resolution and WM. Moreover, these functions are responsible for the correlation between PI resolution and WM. In a neuroimaging study, the neural basis of proactive control was examined by comparing two memory tasks that differed in their demand on binding and inhibition. In addition, the brain regions engaged in reactive control was examined by contrasting trials involving interference or not. The thesis showed that item inhibition carried out by the left inferior frontal cortex (IFC) is involved in proactive control while episodic inhibition carried out by the left IFC and the posterior parietal cortex is involved in reactive control.