Business Intelligence Using Information Gap Decision Theory and Data Mining Approach in Competitive Bidding

Since the nineties, many electric utilities and power network companies have undergone and are still experiencing dynamic change in the ways of doing business, from a vertically integrated industry to an open market system. The operational planning activity of a generation company (GenCo) is no longer a cost-minimizing process, but seeks to maximize its net profit subject to physical constraints and market factors. The objective of this research is to develop a strategic bidding decision-making unit that not only considers the technical aspects of unit operation such as capacity limits but also incorporates information about other market participants and the volatility of the market prices. These additional market factors are significant in such an oligopoly market because they influence the amount of electricity sold and purchased, hence affecting the net profit gained. This project proposes an economic model that attempts to data mine the available historical and current market data in a deterministic two-market-participant environment. Further stochastic analysis is performed using the information gap decision theory concept to quantify the uncertainty that arises. The data mining approach can also be justified for information acquisition to reduce uncertainty, hence improving the information gap model.