Wireless mesh networks (WMNs) can provide wireless services across a wide area. However, mesh router placement in WMNs is a challenging task and is classified as NP-hard problem. In this research work, we propose and implement a hybrid simulation system to optimize mesh router placement in WMNs based on Particle Swarm Optimization (PSO), Hill Climbing (HC) and Distributed Genetic Algorithm (DGA). We implemented in our system Simplex Crossover (SPX) method which we compare the performance with Unimodal Normal Distribution Crossover (UNDX) method considering three router replacement methods: Constriction Method (CM), Random Inertia Weight Method (RIWM) and Fast Convergence Rational Decreasing Vmax Method (FC-RDVM). We carried out simulations for Two islands distribution of mesh clients. The simulation results show that all mesh routers are connected and all clients are covered for both crossover methods. However, the UNDX method has better load balancing than SPX method. Comparing three router replacement methods, the RIWM has better load balancing than CM and FC-RDVM for both UNDX and SPX crossover methods.