Player Management in Soccer Using Particle Swarm Optimization

Manual searching and optimization are less efficient as what is conventionally followed by Coaches or Managers in Soccer to select a squad of 18 members for an upcoming game in soccer, which is less efficient in their approach to problems which are voluminous and technically inclined and involves high complexity and large impediments. Since Soccer involves huge set of data which needs to be gathered as and when an event occurs at real time it becomes more challenging as there can be critical scenarios raised during practice sessions or matches and it is difficult to correlate all the events manually to take decisions. Hence evolutionary intelligence based computational algorithm such as Particle Swarm Optimization algorithm is applied to measure various parameters of trainings and matches on players hence optimizing the players leading to an efficient player selection. The actual data of a private club was taken to implement the algorithm using Particle Swarm Optimization (PSO) for predicting the coefficients of various models to implement on the proposed five models. To minimize the total cost incurred to the club on players we deploy a fitness function to achieve it.