Applying genetic algorithms to reduce information asymmetry in the Brazilian power market

Under the current regulatory framework of the Brazilian power market, Distribution Companies (Discos) are supposed to declare to the Ministry of Mines and Energy (MME) their energy requirements for the five years ahead. Such information is used for the purposes of both planning and organizing of auctions. It is of Independent Power Producers´ great interest to reduce the existent information asymmetry, deliberately created by the auctioneer to encourage competition in the auctions. Knowing the incentives driving Discos in their energy requirements declarations, it is possible to build an optimization model that emulates Discos´ behavior when deciding those declarations. Besides, with the usage of public information of auction outcomes, it seems possible to get closer to Discos´ private information using some ldquoreverse engineeringrdquo. A genetic algorithm optimization model is used for this purpose.