Thesis (Ph.D.) - James Cook University, 2008. / Typescript (photocopy) Bibliography: leaves 198-218.

Thesis (M.Sc. Biological Science)--University of Waikato, 2008. / Title from PDF cover (viewed September 18, 2008) Includes bibliographical references (p. 84-96)

Thesis (Ph. D.)--University of Virginia, 1998. / Includes bibliographical references (p. 105-112). Also available online through Digital Dissertations.

Thesis (Ph. D.)--University of Wisconsin, Madison, 2001. / eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references.

Thesis (Ph.D.)--Civil and Environmental Engineering, Georgia Institute of Technology, 2008. / Committee Chair: Garrow, Laurie Anne; Committee Member: Castillo, Marco; Committee Member: Ferguson, Mark; Committee Member: McCarthy, Patrick; Committee Member: Meyer, Michael.

Thesis (M. Sc.)--University of Canterbury, 2008. / Typescript (photocopy). "June 2008." Includes bibliographical references (leaves 74-84). Also available via the World Wide Web.

Thesis (Ph. D.)--University of Washington, 1996. / Vita. Includes bibliographical references (leaves [77]-87).

Thesis (Ph. D.)--Drexel University, 2006. / Includes abstract and vita. Includes bibliographical references.

Les nouvelles technologies de l’information deviennent un moyen incontournable pour réaliser des transactions instantanées dont tirent profit certaines chaînes logistiques. De ce fait, de nouveaux moyens liés aux opérations d’approvisionnement se développent. Leur émergence est directement liée à l’environnement volatile où évoluent désormais de plus en plus de chaînes logistiques. Les opérations d’approvisionnement du type «e-Procurement» sont des exemples de ces nouvelles pratiques où les chaînes logistiques sont qualifiées d’agiles. L’objectif de cette thèse est d’aborder des problématiques d’approvisionnement où un décideur est confronté au problème de choix de fournisseurs ainsi que la quantité de produits commandée durant le temps. Ces systèmes évoluent dans un environnement changeant caractérisé par des variations asynchrones et répétitives des prix d’achat et de commande ainsi que des capacités des fournisseurs et où l’évolution de ces données est inconnue. L’exemple des systèmes d’achat sur Internet ainsi que les systèmes d’enchères inversées en ligne s’inscrivent parfaitement dans la problématique traitée ici. Dans ce cadre, les approches classiques d’optimisation peuvent s’avérer inadaptées pour ce problème. Les travaux récents sur l’optimisation dynamique peuvent répondre à ce type de questionnement mais n’ont pour l’instant pas été étudiées dans le contexte des chaînes logistiques. Nous proposons, dans cette thèse, une approche basée sur des algorithmes génétiques dynamiques que nous avons illustrée avec trois cas d’application dans le cadre de l’approvisionnement en ligne. / In a context of increased competition between enterprises, supply chains are struggling to respond to an increasingly volatile and complex environment. With technological advances, current practices to build efficient supply chains have changed. Indeed, the enthusiasms of companies with the use of internet have lead researchers to find adequate methods to cope with the dynamic nature of logistics networks. The purpose of this thesis is to address a dynamic procurement issue under asynchronous and repetitive variations over time. The supply chain considered is composed of two levels (buyer-suppliers) operating in highly agile environment. The questions facing the buyer is how many units of product should be purchased and from which supplier in response to variation in term of price and capacity. Because of this highly changing environment characterized by frequents changes in a short time, most of the classical optimization approaches seems inadequate to address these problems. Recently, dynamic optimization has been used successfully to deal with such problems. However, we have no knowledge of its application in a supply chain context. We propose a dynamic genetic approach which is applied to an e-procurement context in aim to optimize the procurement process during time.

Three experiments arranged a concurrent chained schedule that probabilistically arranged reinforcement or extinction. In Experiments 1 and 2, the probability of obtaining food in the terminal link period, following a given left or right lever choice, was the complement of the probability that the initial link choice would produce a transition to the terminal link. Also, the probability of reinforcement in the terminal link was either signaled or unsignaled, depending upon condition. In Experiment 1, a steady-state environment kept the relative probabilities of reinforcement constant within-session and Experiment 2 varied the relative probabilities of reinforcement within-session. Experiment 3 arranged equal rates of terminal link transition to either a signaled-reinforcement or an unsignaled-reinforcement terminal link. The location of the signaled option and the relative probabilities of reinforcement changed within-session. The signaled option produced either a reinforcement-correlated terminal link stimulus (i.e., conditional reinforcement) or an extinction-correlated terminal link stimulus. The unsignaled alternative produced the same terminal link stimulus regardless of the outcome. Overall, Experiments 1 and 2 demonstrated that rats frequently favor the option providing higher rates of terminal link transition at the expense of the probability of terminal link unconditional reinforcement. However, in Experiment 2, this tendency was reduced when the probabilities of reinforcement were signaled, suggesting weak control by conditional reinforcement. Experiment 3 did not show preference for the reinforcement-correlated signaled option in rats. Rather, it appears overall preference was controlled by an avoidance of the extinction-correlated option.