Demand-driven type analysis for dynamically-typed functional languages

Thèse diffusée initialement dans le cadre d'un projet pilote des Presses de l'Université de Montréal/Centre d'édition numérique UdeM (1997-2008) avec l'autorisation de l'auteur.

Effets des changements climatiques sur la santé et la sécurité des travailleurs au Québec

Les impacts des changements climatiques sur la population sont nombreux et ont été relativement bien documentés, ce qui n’est pas le cas de ces impacts sur la santé et la sécurité des travailleurs. L’objectif de cette thèse est de documenter les effets négatifs des changements climatiques sur la santé et la sécurité des travailleurs dans une région d’un pays industrialisé à climat tempéré, comme le Québec. Pour y arriver, deux approches ont été utilisées : a) les dangers et les effets sanitaires ont été identifiés par une revue de la littérature validée par des experts nationaux et internationaux, et des priorités de recherche ont été établies à l’aide d’une méthode de consultation itérative, b) des modèles statistiques, utiles à l’estimation des impacts sanitaires des changements climatiques, ont été développés pour apprécier les associations entre la survenue de lésions professionnelles et l’exposition des travailleurs aux chaleurs estivales et à l’ozone troposphérique, deux problématiques préoccupantes pour le Québec. Le bilan des connaissances a mis en évidence cinq catégories de dangers pouvant affecter directement ou indirectement la santé et la sécurité des travailleurs au Québec (vagues de chaleur, polluants de l’air, rayonnements ultraviolets, événements météorologiques extrêmes, maladies vectorielles transmissibles et zoonoses) et cinq conditions pouvant entraîner des modifications dans l’environnement de travail et pouvant ultimement affecter négativement la santé et la sécurité des travailleurs (changements dans les méthodes agricoles et d’élevage, altérations dans l’industrie de la pêche, perturbations de l’écosystème forestier, dégradation de l’environnement bâti et émergence de nouvelles industries vertes). Quant aux modélisations, elles suggèrent que les indemnisations quotidiennes pour des maladies liées à la chaleur et pour des accidents de travail augmentent avec les températures estivales, et que ces associations varient selon l’âge des travailleurs, le secteur industriel et la catégorie professionnelle (manuelle vs autre). Des associations positives statistiquement non significatives entre les indemnisations pour des atteintes respiratoires aiguës et les concentrations d’ozone troposphérique ont aussi été observées. Dans l’ensemble, cette thèse a permis de dégager douze pistes de recherche prioritaires pour le Québec se rapportant à l’acquisition de connaissances, à la surveillance épidémiologique et au développement de méthodes d’adaptation. Selon les résultats de cette recherche, les intervenants en santé au travail et les décideurs devraient déployer des efforts pour protéger la santé et la sécurité des travailleurs et mettre en place des actions préventives en vue des changements climatiques.

Ambiguity on the insurer's side: the demand for insurance

Empirical evidence suggests that ambiguity is prevalent in insurance pricing and underwriting, and that often insurers tend to exhibit more ambiguity than the insured individuals (e.g., [23]). Motivated by these findings, we consider a problem of demand for insurance indemnity schedules, where the insurer has ambiguous beliefs about the realizations of the insurable loss, whereas the insured is an expected-utility maximizer. We show that if the ambiguous beliefs of the insurer satisfy a property of compatibility with the non-ambiguous beliefs of the insured, then there exist optimal monotonic indemnity schedules. By virtue of monotonicity, no ex-post moral hazard issues arise at our solutions (e.g., [25]). In addition, in the case where the insurer is either ambiguity-seeking or ambiguity-averse, we show that the problem of determining the optimal indemnity schedule reduces to that of solving an auxiliary problem that is simpler than the original one in that it does not involve ambiguity. Finally, under additional assumptions, we give an explicit characterization of the optimal indemnity schedule for the insured, and we show how our results naturally extend the classical result of Arrow [5] on the optimality of the deductible indemnity schedule.

Ambiguity on the insurer's side: the demand for insurance

Empirical evidence suggests that ambiguity is prevalent in insurance pricing and underwriting, and that often insurers tend to exhibit more ambiguity than the insured individuals (e.g., [23]). Motivated by these findings, we consider a problem of demand for insurance indemnity schedules, where the insurer has ambiguous beliefs about the realizations of the insurable loss, whereas the insured is an expected-utility maximizer. We show that if the ambiguous beliefs of the insurer satisfy a property of compatibility with the non-ambiguous beliefs of the insured, then there exist optimal monotonic indemnity schedules. By virtue of monotonicity, no ex-post moral hazard issues arise at our solutions (e.g., [25]). In addition, in the case where the insurer is either ambiguity-seeking or ambiguity-averse, we show that the problem of determining the optimal indemnity schedule reduces to that of solving an auxiliary problem that is simpler than the original one in that it does not involve ambiguity. Finally, under additional assumptions, we give an explicit characterization of the optimal indemnity schedule for the insured, and we show how our results naturally extend the classical result of Arrow [5] on the optimality of the deductible indemnity schedule.

Photothermal deflection measurement on heat transport in GaAs epitaxial layers

In this paper, we report the in-plane and cross-plane measurements of the thermal diffusivity of double epitaxial layers of n-type GaAs doped with various concentrations of Si and a p-type Be-doped GaAs layer grown on a GaAs substrate by the molecular beam epitaxial method, using the laser-induced nondestructive photothermal deflection technique. The thermal diffusivity value is evaluated from the slope of the graph of the phase of the photothermal deflection signal as a function of pump-probe offset. Analysis of the data shows that the cross-plane thermal diffusivity is less than that of the in-plane thermal diffusivity. It is also seen that the doping concentration has a great influence on the thermal diffusivity value. Measurement of p-type Be-doped samples shows that the nature of the dopant also influences the effective thermal diffusivity value. The results are interpreted in terms of a phonon-assisted heat transfer mechanism and the various scattering process involved in the propagation of phonons.

Photothermal deflection measurement on heat transport in GaAs epitaxial layers

In this paper, we report the in-plane and cross-plane measurements of the thermal diffusivity of double epitaxial layers of n-type GaAs doped with various concentrations of Si and a p-type Be-doped GaAs layer grown on a GaAs substrate by the molecular beam epitaxial method, using the laser-induced nondestructive photothermal deflection technique. The thermal diffusivity value is evaluated from the slope of the graph of the phase of the photothermal deflection signal as a function of pump-probe offset. Analysis of the data shows that the cross-plane thermal diffusivity is less than that of the in-plane thermal diffusivity. It is also seen that the doping concentration has a great influence on the thermal diffusivity value. Measurement of p-type Be-doped samples shows that the nature of the dopant also influences the effective thermal diffusivity value. The results are interpreted in terms of a phonon-assisted heat transfer mechanism and the various scattering process involved in the propagation of phonons

Photothermal deflection measurement on heat transport in GaAs epitaxial layers

In this paper, we report the in-plane and cross-plane measurements of the thermal diffusivity of double epitaxial layers of n-type GaAs doped with various concentrations of Si and a p-type Be-doped GaAs layer grown on a GaAs substrate by the molecular beam epitaxial method, using the laser-induced nondestructive photothermal deflection technique. The thermal diffusivity value is evaluated from the slope of the graph of the phase of the photothermal deflection signal as a function of pump-probe offset. Analysis of the data shows that the cross-plane thermal diffusivity is less than that of the in-plane thermal diffusivity. It is also seen that the doping concentration has a great influence on the thermal diffusivity value. Measurement of p-type Be-doped samples shows that the nature of the dopant also influences the effective thermal diffusivity value. The results are interpreted in terms of a phonon-assisted heat transfer mechanism and the various scattering process involved in the propagation of phonons

Photothermal deflection studies on heat transport in intrinsic and extrinsic InP

Laser induced transverse photothermal deflection technique has been employed to determine the thermal parameters of InP doped with Sn, S and Fe as well as intrinsic InP. The thermal diffusivity values of these various samples are evaluated from the slope of the curve plotted between the phase of photothermal deflection signal and pump-probe offset. Analysis of the data shows that heat transport and hence the thermal diffusivity value, is greatly affected by the introduction of dopant. It is also seen that the direction of heat flow with respect to the plane of cleavage of semiconductor wafers influences the thermal diffusivity value. The results are explained in terms of dominating phonon assisted heat transfer mechanism in semiconductors.

Studies on Heat Diffusion in Selected Conducting Polymers using Probe Beam Deflection and Photoacoustic Techniques

In the present study, radio frequency plasma polymerization technique is used to prepare thin films of polyaniline, polypyrrole, poly N-methyl pyrrole and polythiophene. The thermal characterization of these films is carried out using transverse probe beam deflection method. Electrical conductivity and band gaps are also determined. The effect of iodine doping on electrical conductivity and the rate of heat diffusion is explored.Bulk samples of poyaniline and polypyrrole in powder form are synthesized by chemical route. Open photoacoustic cell configuration is employed for the thermal characterization of these samples. The effect of acid doping on heat diffusion in these bulk samples of polyaniline is also investigated. The variation of electrical conductivity of doped polyaniline and polypyrrole with temperature is also studied for drawing conclusion on the nature of conduction in these samples. In order to improve the processability of polyaniline and polypyrrole, these polymers are incorporated into a host matrix of poly vinyl chloride. Measurements of thermal diffusivity and electrical conductivity of these samples are carried out to investigate the variation of these quantities as a function of the content of polyvinyl chloride.