Informatica

Ligand Based Virtual Screening methods are used to screen molecule databases to select the most promising compounds for a query. This is performed by decision-makers based on the information of the descriptors, which are usually processed individually. This methodology leads to a lack of information and hard post-processing dependent on the expert’s knowledge that can end up in the discarding of promising compounds. Consequently, in this work, we propose a new multi-objective methodology called MultiPharm-DT where several descriptors are considered simultaneously and whose results are offered to the decision-maker without effort on their part and without relying on their expertise.

Crossover operators play a very important role by creation of genetic algorithms (GAs) which are applied in various areas of computer science, including combinatorial optimization. In this paper, fifteen genetic crossover procedures are designed and implemented using a modern C# programming language. The computational experiments have been conducted with these operators by solving the famous combinatorial optimization problem – the quadratic assignment problem (QAP). The results of the conducted experiments on the characteristic benchmark instances from the QAP instances library QAPLIB illustrate the relative performance of the examined crossover operations.

The aim is to develop simple for industrial use neuro-fuzzy (NF) predictive controllers (NFPCs) that improve the system performance and stability compensating the nonlinear plant inertia and time delay. A NF plant predictor is trained from real time plant control data and validated to supply a main model-free fuzzy logic controller with predicted plant information. A proper prediction horizon is determined via simulation investigations. The NFPC closed loop system stability is validated based on a parallel distributed compensation (PDC) approximation of the NFPC. The PDC can easily be embedded in industrial controllers. The proposed approach is applied for the real time air temperature control in a laboratory dryer. The improvements are reduced overshoot and settling time.

A new heuristic algorithm for solution of bi-objective discrete competitive facility location problems is developed and experimentally investigated by solving different instances of a facility location problem for firm expansion. The proposed algorithm is based on ranking of candidate locations for the new facilities, where rank values are dynamically adjusted with respect to behaviour of the algorithm. Results of the experimental investigation show that the proposed algorithm is suitable for the latter facility location problems and provides good results in sense of accuracy of the approximation of the true Pareto front.

In the paper we propose a genetic algorithm based on insertion heuristics for the vehicle routing problem with constraints. A random insertion heuristic is used to construct initial solutions and to reconstruct the existing ones. The location where a randomly chosen node will be inserted is selected by calculating an objective function. The process of random insertion preserves stochastic characteristics of the genetic algorithm and preserves feasibility of generated individuals. The defined crossover and mutation operators incorporate random insertion heuristics, analyse individuals and select which parts should be reinserted. Additionally, the second population is used in the mutation process. The second population increases the probability that the solution, obtained in the mutation process, will survive in the first population and increase the probability to find the global optimum. The result comparison shows that the solutions, found by the proposed algorithm, are similar to the optimal solutions obtained by other genetic algorithms. However, in most cases the proposed algorithm finds the solution in a shorter time and it makes this algorithm competitive with others.