Automation of Scheduling Training Sessions in Educational Institutions using Genetic Algorithms

In connection with the improvement of the university management system based on information and communication technologies, the problem of automating the scheduling of training sessions is becoming more and more challenging issue. The solution to this problem depends on the specifics of a particular educational institution and requires the availability of "flexible" methods and algorithms. In addition, the development of information and communication technologies has led to the development of such a direction as distance learning which is especially valuable during the coronavirus pandemic. Taking into account the development and implementation of distance learning in conjunction with traditional training at the university, the task was set to combine the training schedule for two different forms of training. This article describes the development and implementation of the compilation and optimization of the schedule for semester classes in universities based on a genetic algorithm. For this algorithm, various stages of work are defined and implemented: the initial and output data for scheduling are determined, the constraints imposing on schedules are determined, the fitness function is determined based on the weight coefficients of the constraints, the rules of the genetic algorithm at various stages of its work are determined - the generation of the initial population, crossover operator and mutation operators, conditions for selecting individuals and condition for completing the algorithm work and selecting the final schedule.