Abstract
A maximin multiobjective particle swarm optimization algorithm variant is presented, in the context of parallel robotic manipulator design. The choice of a particular structural configuration and its dimensioning is a central issue to the performance of these manipulators. A solution to the dimensioning problem, normally involves the definition of performance criteria as part of an optimization process. In this paper the kinematic design of a 6-dof parallel robotic manipulator is analyzed. Two dynamic performance criteria are formulated and non-dominated optimal solutions are found through a multi-objective particle swarm optimization algorithm. Preliminary simulation results are presented.

Abstract
In this paper an overview of two humanoid simulation platforms is provided: Simspark, a 3D Robocup simulator and the robotics simulator SimTwo. Although these two simulators have different background, today they share the same humanoid robot model, namely the Albaderan NAO robot. According to the fact that developing reliable and robust biped locomotion and low level humanoid behaviors are still challenging tasks, simulation has an important role in improving humanoids movement development approaches. In this paper the two humanoid robotic simulators will be compared in face of simulating Humanoid low level behaviors. The comparison is based on identifying the same role and locomotion approaches. The results show that Simspark is closer to reality than SimTwo.

Abstract
Electrical utilities are confronted daily with unpredictable events in their grids, which may lead to severe security level repercussions in the system, far exceeding all the security principles used for operation and consequently jeopardizing the essential service to the consumers. Incidents are unpredictable disturbances and recent experiences prove that severe contingencies happen. Given these facts, the need to evaluate harsher contingencies arose and such analysis must be rigorous and exhaustive. The security principles used for planning and system operation determine that, given an incident in the Electrical Power System (EPS) which involves the breakdown of any grid element or the simultaneous failure of a double circuit overhead line or the failure of the largest generator in service the supply interruption shall never take place (excluding single-feeding points without alternative) or permanent overloads. To analyse the steady-state security of an EPS it is required too, accurately and efficiently identify the critical contingencies set, i.e., those that when occur may endanger the system's security. The steady-state simulations of the Portuguese Transmission System (PTS) were made in PSS/E software from Siemens PTI using snapshots that represent pictures of the real system. All these security studies produce a large amount of data and information. Recently, the Rough Sets Theory (RST) has been used successfully to handle efficiently problems where large amounts of data are produced. RST constitute a framework for inducing minimal decision rules. These rules in turn can be used to perform a classification task. The main goal of the rough set analysis is to search large databases for meaningful decision rules and finally acquire new knowledge. This approach is based on four main topics: Indiscernibility, Approximation, Reducts and Decision rules. A reduct is a minimal set of attributes from the whole attributes set that preserves the partitioning of the finite set of objects and therefore the original classes. It means that if the redundant attributes are eliminated the reducts are found. Decision rules extracted knowledge, can be used when classifying new objects not in the original information system. In this paper it is proposed an efficient study and contingency analysis in the PTS using the RST. The developed methodology produces a system operation classification, distinguishing in four possible states: normal, alert, emergency I and emergency II. These different operating states correspond to a four levels of security. The four states can be classified horizontally as secure, in normal state and insecure for the remaining ones. The computer programs SecurMining2.0 developed, were applied to the Portuguese test power network.

Abstract
A good relationship between companies and customers is a crucial factor of competitiveness. The improvement of service levels has become a key issue to develop and maintain a loyal relationship with customers. This paper proposes a method for promotions design for retailing companies, based on knowledge extraction from transactions records of customer loyalty cards, aiming to improve service levels and increase sales. At first, customers are segmented using k-means and then the segments' profile is characterized according to the rules extracted from a decision tree. This is followed by the identification of product associations within segments, which can base the identification of the products most suitable for customized promotions. The research reported is done in collaboration with an European retailing company.

Abstract
The recent evidence demonstrating the importance of perishables in terms of store choice and shopping experience makes these products a very interesting topic in many different research areas. Nevertheless, the production planning research has not been paying the necessary attention to the complexities of production systems of such items. The evidence that consumers of perishable goods search for visual and other cues of freshness, such as the printed expiry dates, triggered the development of a multi-objective lot-sizing and scheduling model taking this relevant aspect into account by considering it explicitly as an objective function. A hybrid genetic algorithm based on NSGA-II was developed to allow the decision maker a true choice between different trade-offs from the Pareto front. Computational experiments were based on a case study, reported in the literature, concerning a diary company producing yogurt.

Abstract
Several optimization methods can find good solutions for different instances of the Traveling Salesman Problem (TSP). Since there is no method that generates the best solution for all instances, the selection of the most promising method for a given TSP instance is a difficult task. This paper describes a meta-learning-based approach to select optimization methods for the TSP. Multilayer perceptron (MLP) networks are trained with TSP examples. These examples are described by a set of TSP characteristics and the cost of solutions obtained by a set of optimization methods. The trained MLP network model is then used to predict a ranking of these methods for a new TSP instance. Correlation measures are used to compare the predicted ranking with the ranking previously known. The obtained results suggest that the proposed approach is promising. © 2011 IEEE.