The first, written with Nuno Cardoso, aims to attenuate the problem in the computation of hedging sets for a collection of this same sets. A practical example of this problem is when you want to build certain teams of people with the same skills. In this situation, you can have lots of people but it is very difficult to match teams. Regarding the computation, this problem is too intensive and the thorough search algorithms are too expensive to be applied in actual problems.

This publication therefore proposes an efficient algorithm to solve the existing problem in the process of multiple computers. According to the results, the new algorithm is 34,000 times faster than the previous algorithm, but is also able to efficiently use the processing power of multiple computers to speed up the process.

The second publication, written with Lúcio Sanchez Passos and Rosaldo Rossetti, acts in the control of industrial plants and market management such as the electricity market. The article identifies which agent, or, which intelligent and autonomous entity acts negatively on the system in question. These agents always work for their own benefit but also work for the good functioning of the system in which they operate.

According to the authors, the serious problem of these agents is when one of them works in the opposite direction of the others. This article presents a way to quickly detect the agent that interferes negatively with the system. For this, the ESFL-MAS technique presented in this publication, sum periodic reports from each of the agents that are part of the system, as well as the provision of each other. This technique also identifies the agent that interferes negatively, establishing a causal relationship between the poor provision of this agent and the system malfunctioning.

Is important to stress that, the technique presented, the ESFL-MAS can be used with about 78% of certain probability in the identification of the agent that interfere negatively in the system.