Search

The paper introduces a bio-inspired approach to solve the route allocation problem (RAP) in the transportation networks. The approach extends a well-known meta-heuristics algorithm with the real life constraints that are dealt with in the scheduling process (i.e. the uniform distribution of routes diversity for vehicles, the average distance travelled in a month, the driver’s rest between subsequent trips etc.). The paper is focusing on the engineering aspects of employing bio-inspired algorithms (which proved to have near-optimal results for toy-like problems) to a real-life application domain. The approach proved to be capable of preserving the software components (agents) to the complexity and dynamics of the situation when the RAP requires incremental extensions of constraints to reflect the traffic conditions in the transportation network. [ABSTRACT FROM AUTHOR]

ACS algorithms have been used in solving NP-hard and optimization problems in recent years. ACS ant colonies are homogeneous, but natural colonies are not. In this paper, a new ACS algorithm is proposed. It uses heterogeneous ant colonies which are evolved using a new type of genetic algorithm. Experimental results obtained from solving TSP problem, show the superiority of proposed algorithm over classical ACS. [ABSTRACT FROM AUTHOR]

Commonly used standard deterministic procedures are not efficient enough to solve complex optimization problems. The main drawback of the classical optimization techniques is that large number of objective functions may show appearance of local minima. In such case, these methods fail, and it is difficult to localize the global minimum. One of frequently used solution is application of probabilistic methods i.e. Monte Carlo. These procedures, based on random search, are less sensitive to the local minima. The optimization methods, based on the Artificial Neural Network appear a competitive tools in relation to standard sequential or probabilistic optimization algorithms. The objective of the work is Artificial Neural Networks optimization strategy development and application to the minimization problem of anisotropic effect in deformed material. The anisotropic effect observed in deformed material is an undesired phenomenon, which can cause the elimination of thin products where high quality is required. Therefore, it is very important to look for the deformation technology, which minimizes the anisotropy effect. The work deals with the thin cup bulge forming process. The experiment of axisymertrical bulging process includes unbounded expansion of thin walls of cup being formed by superelastic die (Fig. 1). The objective function is based on the final shape of formed element, while the optimization variable is the punch velocity. The anisotropic model of material was implemented to the FEM code and simulations of the deformation process ware performed. The comparison between experimental results and FEM simulations was shown. © 2004 American Institute of Physics [ABSTRACT FROM AUTHOR]