Development of an Evolutionary Optimization Method for Financial Indicators of Pharmacies

Abstract

The work is devoted to the problem of optimizing the financial performance of pharmacies. To solve this problem, a genetic method of multicriteria optimization was developed with a mutation operator modification to study the degree of influence of factors on the financial performance of pharmacies and the choice of optimization model. The fundamental difference between the developed genetic algorithm and its existing counterparts is the ability to control the mathematical distribution of the values of the solution, which prevents premature convergence of the genetic algorithm and uses all proposed genes in fractions according to the chosen distribution model. A comparative analysis of the work of classical GA and modified versions shows that the best results are achieved in the cognitive-style determination. Three modifications of the mutation operator were developed. The application of the developed methods will lead to a more effective use of the pharmacy area, to reduce unmet demand and, ultimately, to reduce the retail cost of drugs by reducing the costs of storing and servicing the suboptimal loading of the pharmacy.