Hydrogen consumption in oil refineries increases sharply because of more and more heavy and sour crude oil processing, which also makes hydrogen sulfide a considerable contaminant in off-gases of hydrotreaters. This work presents a simulation-based optimization model for synthesis of hydrogen networks with H2S removal. Aspen HYSYS is employed for rigorous process and thermodynamic modeling of the H2S removal unit. The proposed model is solved using the genetic algorithm combined with the linprog and fmincon solvers in the Matlab platform. The optimal hydrogen sources and the recirculated absorbent fed into the H2S removal unit as well as the optimal design of the hydrogen network can be determined simultaneously. A case study is performed to illustrate the application and effectiveness of the proposed model. The result shows that the introduction of H2S removal can decrease the fresh hydrogen consumption by 43% and the total annualized cost by 17%. (C) 2019 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.