High pressure die casting (HPDC) is a widely used process to manufacture complex shapes of aluminium castings with short cycle times of processing for the automotive industry. Most of the die casting companies use specialized commercial software (MAGMASOFT®) to simulate molten metal filling and solidification. This approach works satisfactorily and is industrially accepted for defect prediction capability. In die casting, die cooling is an effective technique to reduce internal porosity. But the actual casting behaviour on account of die cooling is not completely captured in the approach used in the simulation. One of the main reasons for the lack of precision in simulation is the uncertainty in assigning the boundary conditions (heat transfer coefficients) for the die cooling channel. The commercial software (like MAGMASOFT®), although good at simulating molten metal filling and solidification, it lacks in robust approach for prescribing heat transfer coefficients as boundary conditions. Generally, some recommended and averaged heat transfer coefficient (HTC) values are prescribed as the boundary conditions. In this study, a coupled simulation strategy is developed to determine more accurate HTC values of the die cooling channel. The computational domain is divided into two sub-domains, i.e., casting zone and die cooling zone. MAGMASOFT® is used in the casting zone for solidification analysis and ANSYS Fluent® is used in the die cooling zone for fluid flow and heat transfer simulation. A series of iterations are performed by exchanging information at the interface between the two sub-domains till the convergence is achieved. Experiments are also conducted with a thermocouple inserted in the die and the actual die temperature readings are measured. The converged MAGMASOFT® simulation results are compared with the measurements and a good comparison is observed. Also, the influence of flow rate on HTC is studied by conducting trials with two different flow rates and the castings produced are compared with the help of CT scan analysis, microstructure evaluation and thermal imaging. The results obtained are very promising and emphasize the importance of prescribing calculated (rather than assumed) values of heat transfer conditions as boundary conditions in the simulation. With the coupled simulation approach developed in this work, parametric studies are conducted to maximize the heat removal rate from die, for different flow velocity, nozzle diameter, cooling hole diameter and nozzle-to-surface spacing.