Multiscale process modeling is very appealing methodology for process optimization since it highlights certain issues that remain unexplored with conventional methodologies and debottlenecks certain potentialities that remain unexploited in chemical plants. In this work, a kinetic model with 2400 reactions and 140 species is implemented in a proper reactor network to characterize the thermal furnace and the waste heat boiler of sulfur recovery units; the network with detailed kinetics is the kernel of a Claus process simulation that include all the unit operations and the catalytic train. By doing so, reliable estimation of acid gas conversion, elemental sulfur recovery, and steam generation is achieved with the possibility to carry out an integrated process-energy optimization at the total plant scale.