In this paper, we investigate network location problem for congestible facilities, considering balking, the reluctance of customers to join the queue upon arrival. Balking often happens when the estimated waiting times by customers are more than their patience threshold. Based on the number of current customers, each customer decides to join or leave the system upon the arrival. Since balking is a usual issue in the real world situations, taking it into consideration increases the reality of mathematical models. In the proposed model, the selection of candidate sites and the assignment of customers to facilities are performed in such a manner that the total lost demands is minimized. We use a monotonically decreasing function to calculate arrival rates for facilities. In order to evaluate the proposed mathematical model, some numerical examples are presented and solved using GAMS.