The development of metaheuristics and Boolean Satisfiability representation plays an important part in a neural network (NN) and Artificial Intelligence (AI) communities. In this paper, a new hybrid discrete version of the artificial dragonfly algorithm (DADA) applying a minimum objective function in agent-based modelling (ABM) obeying a specified procedure to optimize the states of neurons, for optimal Boolean Exact Satisfiability representation on NETLOGO as a dynamic platform. We combined the artificial dragonfly algorithm for its random searching ability that encourages diverse solutions and formation of static swarm’s mechanism to stimulus computational problems to converge to the best global optimal search space. The global performance of the proposed DADA was compared with genetic algorithm (GA) that are available in the literature based on the global minimum ratio (gM), Local Minimum Ratio (yM), Computational time (CPU) and Hamming distance (HD). The final results showed good agreement between the proposed DADA and discrete version of GA to efficiently optimize the Exact-kSAT problem. It found that DADA-ABM has high potentiality for optimizing or modelling a network that is very hard or often impossible to capture by exact or traditional optimization modelling techniques such as Boolean satisfiability problem is better than existing methods in the literature.