Fuzzy cognitive maps learning using Artificial Bee Colony optimization

Most of the dynamic systems are hard to express in mathematical models due to their complex, nonlinear and uncertain characteristics. Thus, advanced methodologies are needed, using human experience, present expert knowledge and historical data. Hence fuzzy cognitive maps are quite convenient, simple, powerful and practical tools for simulation and analysis of these kinds of dynamic systems. Yet, human experts are subjective and cannot handle relatively complex fuzzy cognitive maps (FCMs); hence, new approaches are required to develop for an automatic building of fuzzy cognitive maps. In this study, Artificial Bee Colony (ABC) global optimization algorithm is proposed for the first time in literature for an automated generation of fuzzy cognitive maps from historical data. An ERP management model is used as the illustrative example to obtain the data for training and validation. The obtained results show the success of the ABC learning for FCMs.