An immune evolution mechanism for the study of stress factors in supervised and controlled systems

Reinterpretation and adaptation of knowledge from technical sciences into the field of sports science is at the forefront of advancing performance. Evolution-based systems open the possibility of evaluating different stress factors that appear in the systematization of training programs. Our aim was to demonstrate that it is possible to use evolution mechanisms to model the real life influence of stress factors during performance training of athletes, a concept that can be generalized to other supervised or controlled systems. We have recorded data from one former basketball player to be used in the development of a simulation model based on immune genetic algorithms. We developed an immunization scheme that is parameterized for simulating different training outcomes based on type of athlete. Results confirm one of these cases is possibly the most similar to real life situations. Thus, we obtained an evolution model that aligns with the generative experiment as proof of concept for the evaluation of performance under stress. In particular, for sports science, we may have found a way to analyze training programs before their execution and to spot weaknesses in them.