Artificial worlds modeling of human resources management through evolutionary learning

Abstract

Human resource management plays a vital role in our current society. We describe an evolutionary learning model, composed of a hierarchy of submodels, that simulates human resource management. The submodel at the highest level is the environment submodel, modeled as a 2-D array of grid locations. The next level in the model hierarchy is the organization submodel, consisting of leaders and jobs. The last level in the model hierarchy is the subordinate submodel that each individual is represented by a two-level genotype-to-phenotype structure. The whole system can be pictured as an artificial world that allows different modes of interactions, including interactions among subordinates and environments, among subordinates and leaders, and among subordinates and jobs. The goal is to improve the fitness of people to their environments, leaders, and jobs through evolutionary learning. Evolution can occur at the level of managing migration among different environments, at the level of improving relationship among leaders and subordinates, and at the level of enhancing the capabilities of subordinates for specific jobs. Our simulation results showed that increasing migration mobility can improve the fitness of people, but in the meantime it leads to the clustering of people (i.e., some areas are extremely crowded whereas some have no resident). The results also showed that subordinates respond differently to different types of leaders and jobs.