An evolutionary method for solving the traveling salesman problem in the field of pharmacy business by optimizing the work of the drug delivery device is proposed in this paper. Modifications of three methods of initialization of the initial population of the genetic algorithm are developed. The software implementation is proposed to solve the problem of a sales-man in the pharmacy business by optimizing the process of drug delivery, using modified evolutionary methods. Unlike existing methods, the modified version of the evolution method allows to choose the original method of population initialization when solving the problem of traveling salesman, which, in turn, allows to generate more adapted chromosomes (chromosomes with better values of fitness functionality) at the stage of initialization and thereby improve the results algorithm. It is also possible to graphically monitor the process of solving the seller's problem, get the result in text and graphic forms. The principles of objectoriented programming, namely the use of classes, the principle of data encapsulation and inheritance, were used when writing the program. UML diagrams of classes, sequences, activities, states and cooperation were used to visualize the structure and functional relationships of the modules of the developed software.