Optimisation et D
Forum 'Annonces' - Sujet créé le 2009-05-25
Chers collègues,
Veuillez trouver le programme complet du workshop inaugural de la chaire Microsoft-CNRS Optimisation et Développement Durable, à l'école Polytechnique. Nous espérons vous voir nombreux dès 10h pour le lancement officiel de la chaire, http://www.polytechnique.fr/osd-chair/.
A noter :
· Au départ de Paris, gare RER Denfert-Rochereau, un bus gratuit ralliera directement l'École Polytechnique. Départ 9h10.
· L'entrée est libre, cependant il est préférable de s'inscrire en ligne, http://www.polytechnique.fr/inscriptions/inscription_externe_evenement.php?id=10.
Merci de diffuser aux personnes susceptible d'être intéressées.
Inaugural Workshop
Microsoft-CNRS Chair
Optimisation for Sustainable Development
at École Polytechnique
Wednesday, June 3, 2009
Organizers
* Philippe Baptiste, CNRS/LIX, École Polytechnique, Philippe.Baptiste@polytechnique.fr
* Youssef Hamadi, Microsoft Research Cambridge, Microsoft Research-INRIA Joint Centre, youssefh@microsoft.com
Program
· 14:00-14:30 Stochastic Optimization
o Pascal Van Hentenryck, Online Stochastic Optimization for Sustainable Development Opportunities and Challenges, Brown University
· 14:30-15:00 Bio-Economy
o Luc Doyen, Stochastic viability for the sustainable management of natural resources, CNRS CERSP
· 15:00-15:30 Economy
o Christian Gollier, Optimal environmental policy under ambiguity, University Toulouse-I, Laboratoire d'Economie des Ressources Naturelles
· 15:30-16:00 Coffee-break
· 16:00-16:30 Physics
o Mehdi Metaiche and John Palmeri, Optimization of Reverse Osmosis and Nanofiltration Desalination Systems, CNRS, University of Montpellier, University of Toulouse
· 16:30-17:00 Energy
o Claude Le Pape and Vincent Mazauric, Operational and Prospective Optimization Problems for Energy Efficiency, Schneider Electric
· 17:00-17:30 Scheduling
o Kirk Pruhs, A science of power management, University of Pittsburgh
· 17:30-18:00 Energy
o Laurent Schmitt, New technological challenges related to Energy Optimisation in the Energy Industry, Alstom
· 18:00-19:00 Cocktail
Abstracts and biographies
Online Stochastic Optimization for Sustainable Development Opportunities and Challenges
Progress in telecommunications is enabling novel applications of optimization technology in which decisions must be taken in real-time and under uncertainty. Online stochastic optimization approaches these applications by combining optimization technology and predictive modeling to provide solutions of higher quality. This talk first reviews these recent developments and illustrate them on applications in disaster management. It then discusses how these techniques may have a fundamental impact in sustainable development and, in particular, in renewable power and describe a specific effort in this direction.
Biography: Pascal Van Hentenryck is professor of computer science at Brown University. He is the author of five books published by the MIT Press and the author of the innovative optimization systems CHIP, Numerica, OPL, and Comet, all of which have been commercialized. Pascal is a fellow of the Association for the Advancement of Artificial Intelligence (AAAI) and has been awarded the 2002 INFORMS ICS Award for research excellence at the intersection of computer science and operations research, the 2006 ACP Award for research excellence in constraint programming, and an honorary degree from the University of Louvain in 2008. His research interests span optimization, programming languages, software verification, and computational biology.
Stochastic viability for the sustainable management of natural resources
Abstract: A basic issue for a sustainable management of natural resources is the reconciliation of ecological and economic requirements with an intergenerational equity perspective. Such a goal is made more complex by the presence of numerous uncertainties in the systems at stake including climatic, habitat, demographic or market disturbances. Stochastic or robust viability and more generally viable control under uncertainty is here proposed as a relevant modeling framework to deal with such issues (DeLara & Doyen, 2008). Such approach does not strive to determine optimal or steady-state paths and decisions for the joint dynamics of resources and exploitations, but rather aims at maintaining the trajectories of systems within satisfying normative bounds and constraints that mix ecological, economic and social requirements. Hence the approach offers a multi-criteria perspective which allows to easily accommodate monetary values and non monetary criteria and to combine commodities and services derived from biodiversity. The approach also provides ways to analyze and control (if possible) the risks and vulnerability of ecological-economic systems dynamics. Hence conceptual links to Population Viability Analysis, precautionary approaches and conservation biology are strong. Moreover it is closely related to the maximin or rawlsian approach which provides important insights regarding intergenerational equity. Connections with resilience concept or recovery problems are also strong. From the methodological viewpoint, it can be proved how a dynamic programming method applies for stochastic or robust viable control. Examples related to the management of renewable resources and biodiversity (fisheries, agriculture, hunting) illustrate the general ideas.
Biography: Luc Doyen, HDR in applied mathematics, has received an education in control theory, optimization and mathematical economics. He currently holds a permanent position at the CNRS (Centre National de la Recherche Scientifique), where he is especially involved in bio-economic modelling, viable management of biodiversity and mathematics of sustainability. A book « Sustainable management of natural resources models and methods » recently published by Springer and more than 30 publications in international peer-reviewed journals emphasize the mix between the applied and theoretical dimensions of his research activity and its inter-disciplinary nature balancing ecology, environmental economics, modeling and mathematics. The applied component of his research relies on contracts and grants (ANR biodiversity, ANR Systerra, ACI MEDD, IFB-GICC, program FEAST) pointing out the transfer of the theoretical results toward national or international institutions such as INRA, IFREMER, WorldFishCenter, CSIRO regarding the management of fisheries, agriculture and biodiversity. Luc Doyen has also coordinated a seminar entitled « Développement Viable » during four years at the ENS in Paris and belong to different networks (RTP M3D,...) for the animation of research. He is reviewer of various international journals and has participated to different committees (University Paris 6 , IRD, ACI, CEMAGREF) for the evaluation of research. Its numerous teachings (Masters, Engineering schools) and supervisions (Ph-D, masters, .) also highlight the interdisciplinary feature of his activities at the interface between mathematical and numerical modeling, economics and ecology.
Optimal environmental policy under ambiguity
Abstract: The aim of this lecture is to present explore new decision models involving two dimensions : ambiguous uncertainty and the distant future. It is related to the problem of defining sustainability in the absence of objective probabilities about either the long-term costs and benefits of the environmental policy, or about the quality of the environment itself. We recognize that economic agents are ambiguity-averse, i.e., they prefer to get 100 with a sure probability of ½ than to get 100 with an unknown probability whose mean is ½. We explore the consequences of this additional hypothesis on the optimal investment strategy, and on the choice of the long-term discount rate. We link our analysis to the notion of the precautionary principle, and to the debate about the optimal intensity of reduction of CO2 emissions.
Biography: Christian Gollier is currently Deputy Director of the Toulouse School of Economics (TSE), Research Director at the Institut d'Economie Industrielle (IDEI) and Director of the Laboratoire d'Economie des Ressources Naturelles (LERNA), a research center associated to INRA and CNRS on environmental economics. He has published more than eighty articles in top-tier economic journals and he is also Associate Editor, editor or co-editor of various top-field scientific journals. He has also published 7 books on risk including "The Economics of Risk and Time", MIT Press, winner of "the 2001 Paul A. Samuelson Award" and of the "2002 Prix Risques-les Echos". He also publishes a monthly column in the french newspapers Les Echos. His current topics of research extend from Decision Theory under Uncertainty to Environmental Economics through Finance, Investment, Consumption Theory, Insurance Economics and Cost-Benefit Analysis, with a special interest for long term (sustainable) effects.
Optimization of Reverse Osmosis and Nanofiltration Desalination Systems
Abstract: Faced with ever increasing demands for water worldwide, membrane technology is being called upon to play an increasingly important role in providing solutions to water supply problems. Reverse Osmosis (RO), and increasingly Nanofiltration (NF), membrane processes are being used to design sea and brackish water desalination plants to provide potable and irrigation water for coastal populations, even in places where such technology was previously considered to be too costly. Hybrid systems coupling RO and NF are also being adopted in the design of integrated systems that are more e
Veuillez trouver le programme complet du workshop inaugural de la chaire Microsoft-CNRS Optimisation et Développement Durable, à l'école Polytechnique. Nous espérons vous voir nombreux dès 10h pour le lancement officiel de la chaire, http://www.polytechnique.fr/osd-chair/.
A noter :
· Au départ de Paris, gare RER Denfert-Rochereau, un bus gratuit ralliera directement l'École Polytechnique. Départ 9h10.
· L'entrée est libre, cependant il est préférable de s'inscrire en ligne, http://www.polytechnique.fr/inscriptions/inscription_externe_evenement.php?id=10.
Merci de diffuser aux personnes susceptible d'être intéressées.
Inaugural Workshop
Microsoft-CNRS Chair
Optimisation for Sustainable Development
at École Polytechnique
Wednesday, June 3, 2009
Organizers
* Philippe Baptiste, CNRS/LIX, École Polytechnique, Philippe.Baptiste@polytechnique.fr
* Youssef Hamadi, Microsoft Research Cambridge, Microsoft Research-INRIA Joint Centre, youssefh@microsoft.com
Program
· 14:00-14:30 Stochastic Optimization
o Pascal Van Hentenryck, Online Stochastic Optimization for Sustainable Development Opportunities and Challenges, Brown University
· 14:30-15:00 Bio-Economy
o Luc Doyen, Stochastic viability for the sustainable management of natural resources, CNRS CERSP
· 15:00-15:30 Economy
o Christian Gollier, Optimal environmental policy under ambiguity, University Toulouse-I, Laboratoire d'Economie des Ressources Naturelles
· 15:30-16:00 Coffee-break
· 16:00-16:30 Physics
o Mehdi Metaiche and John Palmeri, Optimization of Reverse Osmosis and Nanofiltration Desalination Systems, CNRS, University of Montpellier, University of Toulouse
· 16:30-17:00 Energy
o Claude Le Pape and Vincent Mazauric, Operational and Prospective Optimization Problems for Energy Efficiency, Schneider Electric
· 17:00-17:30 Scheduling
o Kirk Pruhs, A science of power management, University of Pittsburgh
· 17:30-18:00 Energy
o Laurent Schmitt, New technological challenges related to Energy Optimisation in the Energy Industry, Alstom
· 18:00-19:00 Cocktail
Abstracts and biographies
Online Stochastic Optimization for Sustainable Development Opportunities and Challenges
Progress in telecommunications is enabling novel applications of optimization technology in which decisions must be taken in real-time and under uncertainty. Online stochastic optimization approaches these applications by combining optimization technology and predictive modeling to provide solutions of higher quality. This talk first reviews these recent developments and illustrate them on applications in disaster management. It then discusses how these techniques may have a fundamental impact in sustainable development and, in particular, in renewable power and describe a specific effort in this direction.
Biography: Pascal Van Hentenryck is professor of computer science at Brown University. He is the author of five books published by the MIT Press and the author of the innovative optimization systems CHIP, Numerica, OPL, and Comet, all of which have been commercialized. Pascal is a fellow of the Association for the Advancement of Artificial Intelligence (AAAI) and has been awarded the 2002 INFORMS ICS Award for research excellence at the intersection of computer science and operations research, the 2006 ACP Award for research excellence in constraint programming, and an honorary degree from the University of Louvain in 2008. His research interests span optimization, programming languages, software verification, and computational biology.
Stochastic viability for the sustainable management of natural resources
Abstract: A basic issue for a sustainable management of natural resources is the reconciliation of ecological and economic requirements with an intergenerational equity perspective. Such a goal is made more complex by the presence of numerous uncertainties in the systems at stake including climatic, habitat, demographic or market disturbances. Stochastic or robust viability and more generally viable control under uncertainty is here proposed as a relevant modeling framework to deal with such issues (DeLara & Doyen, 2008). Such approach does not strive to determine optimal or steady-state paths and decisions for the joint dynamics of resources and exploitations, but rather aims at maintaining the trajectories of systems within satisfying normative bounds and constraints that mix ecological, economic and social requirements. Hence the approach offers a multi-criteria perspective which allows to easily accommodate monetary values and non monetary criteria and to combine commodities and services derived from biodiversity. The approach also provides ways to analyze and control (if possible) the risks and vulnerability of ecological-economic systems dynamics. Hence conceptual links to Population Viability Analysis, precautionary approaches and conservation biology are strong. Moreover it is closely related to the maximin or rawlsian approach which provides important insights regarding intergenerational equity. Connections with resilience concept or recovery problems are also strong. From the methodological viewpoint, it can be proved how a dynamic programming method applies for stochastic or robust viable control. Examples related to the management of renewable resources and biodiversity (fisheries, agriculture, hunting) illustrate the general ideas.
Biography: Luc Doyen, HDR in applied mathematics, has received an education in control theory, optimization and mathematical economics. He currently holds a permanent position at the CNRS (Centre National de la Recherche Scientifique), where he is especially involved in bio-economic modelling, viable management of biodiversity and mathematics of sustainability. A book « Sustainable management of natural resources models and methods » recently published by Springer and more than 30 publications in international peer-reviewed journals emphasize the mix between the applied and theoretical dimensions of his research activity and its inter-disciplinary nature balancing ecology, environmental economics, modeling and mathematics. The applied component of his research relies on contracts and grants (ANR biodiversity, ANR Systerra, ACI MEDD, IFB-GICC, program FEAST) pointing out the transfer of the theoretical results toward national or international institutions such as INRA, IFREMER, WorldFishCenter, CSIRO regarding the management of fisheries, agriculture and biodiversity. Luc Doyen has also coordinated a seminar entitled « Développement Viable » during four years at the ENS in Paris and belong to different networks (RTP M3D,...) for the animation of research. He is reviewer of various international journals and has participated to different committees (University Paris 6 , IRD, ACI, CEMAGREF) for the evaluation of research. Its numerous teachings (Masters, Engineering schools) and supervisions (Ph-D, masters, .) also highlight the interdisciplinary feature of his activities at the interface between mathematical and numerical modeling, economics and ecology.
Optimal environmental policy under ambiguity
Abstract: The aim of this lecture is to present explore new decision models involving two dimensions : ambiguous uncertainty and the distant future. It is related to the problem of defining sustainability in the absence of objective probabilities about either the long-term costs and benefits of the environmental policy, or about the quality of the environment itself. We recognize that economic agents are ambiguity-averse, i.e., they prefer to get 100 with a sure probability of ½ than to get 100 with an unknown probability whose mean is ½. We explore the consequences of this additional hypothesis on the optimal investment strategy, and on the choice of the long-term discount rate. We link our analysis to the notion of the precautionary principle, and to the debate about the optimal intensity of reduction of CO2 emissions.
Biography: Christian Gollier is currently Deputy Director of the Toulouse School of Economics (TSE), Research Director at the Institut d'Economie Industrielle (IDEI) and Director of the Laboratoire d'Economie des Ressources Naturelles (LERNA), a research center associated to INRA and CNRS on environmental economics. He has published more than eighty articles in top-tier economic journals and he is also Associate Editor, editor or co-editor of various top-field scientific journals. He has also published 7 books on risk including "The Economics of Risk and Time", MIT Press, winner of "the 2001 Paul A. Samuelson Award" and of the "2002 Prix Risques-les Echos". He also publishes a monthly column in the french newspapers Les Echos. His current topics of research extend from Decision Theory under Uncertainty to Environmental Economics through Finance, Investment, Consumption Theory, Insurance Economics and Cost-Benefit Analysis, with a special interest for long term (sustainable) effects.
Optimization of Reverse Osmosis and Nanofiltration Desalination Systems
Abstract: Faced with ever increasing demands for water worldwide, membrane technology is being called upon to play an increasingly important role in providing solutions to water supply problems. Reverse Osmosis (RO), and increasingly Nanofiltration (NF), membrane processes are being used to design sea and brackish water desalination plants to provide potable and irrigation water for coastal populations, even in places where such technology was previously considered to be too costly. Hybrid systems coupling RO and NF are also being adopted in the design of integrated systems that are more e