CP-AI-OR 2012 - May, 2012


Title of the Conference/Workshop: 9th International Conference on Integration of Artificial Intelligence (AI) and Operations Research (OR) techniques in Constraint Programming

Location: Nantes, France

Dates: May-June, 2012

url: http://www.emn.fr/z-info/cpaior-2012/


Brief description of the conference:

The CPAIOR 2012 conference aims at gathering researchers from constraint programming (CP), artificial intelligence (AI) and operations research (OR) to present new techniques or applications in combinatorial optimization and to provide an opportunity for researchers in one area to learn about techniques in the other ones.

Organized by the Ecole des Mines de Nantes in partnership with the LINA and the INRIA, the 9th international CPAIOR conference  will be held in Nantes, in France, from May 28 to June 1, 2012.

The CPAIOR 2012 conference aims at gathering researchers from constraint programming (CP), artificial intelligence (AI) and operations research (OR) to present new techniques or applications in combinatorial optimization and to provide an opportunity for researchers in one area to learn about techniques in the other ones.

Insight on the topics of the conference :

  • Inference and relaxation methods.
  • Integration methods.
  • Modeling methods.
  • Innovative Applications of CP/AI/OR techniques.
  • Implementation of CP/AI/OR techniques and optimization systems.

 Campus21 publications


Invited Talk: Optimization with Variable Energy Prices.


Abstract

The global energy market is undergoing major changes, in many countries market-based mechanisms are used to determine a real-time price for electricity.  This is driven by volatile prices for fossil fuels, an increasing market share of renewable, but hard to predict, energy sources like solar PV (photo-voltaic) and wind energy, and a need to reduce CO2 emissions by penalizing inefficient and polluting energy generators. While whole-sale prices can vary widely during a day, most residential and industrial consumers are still using fixed-price energy tariffs, as they can not easily exploit the advantages of variable prices.
We consider the role that optimization can play in helping users to plan their energy consumption over time, which reduces their energy bill, helps utility companies to reduce peak demands, and thus improve the overall environmental impact. For this we introduce a family of energy cost aware resource constraints, which can model time and volume dependent energy consumption.We show how a generalization of existing LP relaxations for the cumulative constraint can be used to perform strong propagation for these constraints. In order to plan the energy consumption into the future we also have to be able to predict the real-time price hours or even days ahead. Experiments based on data from the Irish energy market indicate that high quality schedules can be found, even if the future price is not known very accurately, but also that improving the accuracy of the forecast based on standard quality measures is not enough to guarantee even better schedule costs. We conclude with presenting some ongoing work on energy management in residential homes and public spaces like university campuses, which extends our
optimization models with thermal models for heating and air conditioning and energy storage.