The development of solution search method using improved jumping frog algorithm

Abstract

The object of research is the decision making support systems. The subject of the research is the decision making process in management tasks using an advanced jumping frog algorithm (JFA), an advanced genetic algorithm and evolving artificial neural networks. A method of finding solutions with the use of improved JFA is proposed. The research is based on the JFA to find a solution regarding the object state. Evolving artificial neural networks are used to train frog agents (FA). The method has the following sequence of actions: – an input of initial data; – processing of initial data taking into account the degree of uncertainty; – calculation of the value of the criterion of optimality of each permutation from the initial FA population – global search of FA; – an improvement of the FA position in the search space; – a regulation of the speed of vehicle movement. – an improvement of the working conditions of JFA; – the FA rearrangement; – an unification of all memplexes into one group; – the verification of the fulfillment of the conditions of JFA operation; – the search for the best FA; – training of the FA knowledge bases. The originality of the proposed method consists in the arrangement of the FA taking into account the uncertainty of the initial data, the improved procedures of global and local edge taking into account the degree of data noise about the analysis object state, the adjustment of the degree of data noise during the FA movement, the adjustment of the speed of the FA movement. Also, the peculiarity of the proposed method is the use of an improved procedure for FA training. The use of the method makes it possible to increase the efficiency of data processing at the level of 14–18 % due to the use of additional improved procedures. The proposed method should be used to solve the problems of evaluating complex and dynamic processes in the interests of solving national security problems