Antonio José Fernández Leiva

Hybrid Cooperation Models for the Tool Switching Problem

The Tool Switching Problem (ToSP) is a hard combinatorial optimization problem of relevance in the field of flexible manufacturing systems (FMS), that has been tackled in the literature using both complete and heuristic methods, including local-search metaheuristics, population-based methods and hybrids thereof (e.g., memetic algorithms). This work approaches the ToSP using several hybrid cooperative models where spatially-structured agents are endowed with specific localsearch/ population-based strategies. Issues such as the intervening techniques and the communication topology are analyzed via an extensive empirical evaluation. It is shown that the cooperative models provide better results than their constituent parts. Furthermore, they not only provide solutions of similar quality to those returned by the memetic approach but raise interest prospects with respect to its scalability.

Practices of advanced programming: Tradition versus innovation

This paper compares two different approaches for teaching complex programming concepts in lab sessions. The first approach is based on the widely held assumption that the programming knowledge and skills which the student obtains are predominantly acquired from classroom activities undertaken at the instigation of the teacher. This automatically implies that the student must attend a minimum number of lab sessions. The second approach is to motivate student learning via a full-term activity based on the implementation of computer games from scratch.

Design of emergent and adaptive virtual players in a war RTS game

Basically, in (one-player) war Real Time Strategy (wRTS) games a human player controls, in real time, an army consisting of a number of soldiers and her aim is to destroy the opponents assets where the opponent is a virtual (i.e., non-human player controlled) player that usually consists of a pre-programmed decision-making script. These scripts have usually associated some well-known problems (e.g., predictability, non-rationality, repetitive behaviors, and sensation of artificial stupidity among others). This paper describes a method for the automatic generation of virtual players that adapt to the player skills; this is done by building initially a model of the player behavior in real time during the game, and further evolving the virtual player via this model in-between two games. The paper also shows preliminary results obtained on a oneplayer wRTS game constructed specifically for experimentation.

A memetic cooperative optimization schema and its application to the tool switching problem

This paper describes a generic (meta-)cooperative optimization schema in which several agents endowed with an optimization technique (whose nature is not initially restricted) cooperate to solve an optimization problem. These agents can use a wide set of optimization techniques, including local search, population-based methods, and hybrids thereof, hence featuring multilevel hybridization. This optimization approach is here deployed on the Tool Switching Problem (ToSP), a hard combinatorial optimization problem in the area of flexible manufacturing. We have conducted an ample experimental analysis involving a comparison of a wide number of algorithms or a large number of instances. This analysis indicates that some meta-cooperative instances perform significantly better than the rest of the algorithms, including a memetic algorithm that was the previous incumbent for this problem.

Memetic and Hybrid Evolutionary Algorithms

This chapter presents an overview of hybridization mechanisms in evolutionary algorithms. Such mechanisms are aimed to introducing problem knowledge in the optimization technique by means of the synergistic combination of general–purpose methods and problemspecific add-ons. This combination is presented in this work from two wide perspectives: memetic algorithms and cooperative optimization models. Memetic algorithms are based on the smart orchestration of global (population-based) and local (trajectorybased) techniques, using an algorithmic scheme in which the latter are often subordinated to the former. As to cooperative models, they are based on the collaboration of different optimization techniques that exchange information in order to boost their respective performances. Both approaches, memetic algorithms and cooperative models, provide a framework to achieve synergistic algorithmic combinations for the resolution of large-scale combinatorial problems.

Memetic Algorithms and Complete Techniques

As mentioned in previous chapters in this volume, metaheuristics (and specifically MAs) have a part of their raison d’etre in practically solving problems whose resolution would be otherwise infeasible by means of other non-heuristic approaches. Such alternative non-heuristic approaches are complete methods that –unlike heuristics– do guarantee that the deviation from optimality of the solution they will provide is somehow bounded (and as a particular case, that the optimal solution will be found). These methods are eventually limited by the curse of dimensionality, yet they may still constitute a very interesting resource either from the application point of view, or from the lessons that can be learnt from them. Indeed, in some sense these approaches could be considered complementary to metaheuristics rather that mere “rivals”. Even more so in the case of MAs, whose philosophy has been since its inception much more flexible and integrative rather than dogmatic or exclusive.

On the Use of Human-Guided Evolutionary Algorithms for Tackling 2D Packing Problems

We consider a 2D packing problem in which a collection of rectangular objects have to be arranged within a larger rectangular area of fixed width, such that its height is minimized. This problem is tackled using evolutionary algorithms that combine permutational decoders and GRASP-based principles. It is shown that this approach can be improved by allowing the user interact with the algorithm, tuning the greediness of the genotype-to-phenotype decoding. Experiments are presented on three different problem instances with sizes ranging from 19 up to 49 objects.