RFID network planning (RNP) is the problem of allocating readers in a work area so that the constructed reader network can cover all tags, while satisfying some constraints such as minimum interference, minimum number of used readers, etc. As a multi-objective optimization problem, RNP is proven to be NP-hard, so natural-based methods are often used to find the best solution in a given period. Approaches based on GA, PSO, Cuckoo Search, and even artificial neural networks have also been proposed to solve this RNP problem. However, a problem for the natural-based methods class is choosing the best hyperparameters. This paper proposes an integration of the genetic algorithm and the Hopfield network, in which the RNP problem is optimized by the Hopfield network based on the hyperparameters determined by the genetic algorithm. Experimental results show that the integrated model achieves impressive results with a given working area in which the tags are evenly distributed. However, in return, the execution time increases significantly due to searching and determining the optimal hyperparameter set.