Comparison and use of optimization algorithms in information systems

DOI: 10.31673/2412-9070.2025.042240

Abstract

The article explores the possibilities of enhancing the efficiency of enterprise information systems (IS) through the implementation of optimization algorithms. It presents a general problem statement highlighting the importance of utilizing modern and highly adaptable enterprise information systems, as well as the necessity of their improvement and optimization. The article briefly reviews recent studies and publications in this area. The aim of the article is formulated as a theoretical analysis of five modern optimization algorithms.
The following five methods are analyzed: Genetic Algorithm, Particle Swarm Optimization, Differential Evolution, Simulated Annealing, and the Bee Algorithm. Their main advantages are studied, while specific disadvantages in practical application are also highlighted. The areas of application within information systems are identified, including resource management, logistics, planning, and others. In modern enterprise information systems – where processing large volumes of data and making real-time decisions is critically important – the choice of an effective optimization algorithm plays a significant role. Results of the comparative analysis show that none of the algorithms is universal for all types of tasks; however, each has the potential to be part of the optimization process in enterprise information systems. The conducted research allows us to conclude that the choice of an optimal algorithm depends on the specific task, requirements for decision-making speed and accuracy, availability of resources, and other factors. The comparative characteristics of the optimization algorithms are given and summarized in Table 1.
In many cases, the use of hybrid approaches – combining the advantages of multiple algorithms – can provide the best results. This article also outlines future research plans concerning the use of optimization algorithms in improving enterprise information systems, including the development and review of hybrid models that integrate the strengths of several algorithms to achieve a better balance between accuracy, speed, and stability.

Keywords: enterprise information systems; information systems improvement; genetic algorithm; particle swarm optimization; differential evolution; simulated annealing; bee algorithm; optimization algorithms.