Melting layers (MLs) contain complex microphysical processes that largely influence the precipitation systems. They usually manifest as bright bands in radar observations. Most radar forward operators embedded in numerical weather models showed a limited capacity to simulate MLs because bulk microphysics schemes (BMPs) neglect the melting state of particles to reduce the computational complexity. To present polarimetric simulations closer to observations, a new ML simulation algorithm was proposed for radar forward operators, which considered the evolutions of melting particle numbers and the coexistence of the liquid drops. The new simulator was tested using three double-moment schemes [i.e., Morrison, National Severe Storm Laboratory (NSSL), and WRF 6 class (WDM6)] for the simulations of Typhoon Hato (2017). Compared with polarimetric radar observations, it was found that the peak and thickness of the simulated bright bands were overestimated in the Morrison and NSSL scheme and underestimated in the WDM6 scheme using the simulating algorithm raised by past research. The new simulator significantly improved the simulation of MLs in the three schemes by simulating a more realistic evolution of particle size distribution of melting particles. The microphysical factors influencing polarimetric simulations are thoroughly examined, providing valuable insights for better understanding the polarimetric signatures of mixed-phase clouds. A forward operator with improved capability to represent melting particles will also help us develop better optimization-based radar retrieval methods for precipitation with mixed-phase processes.