Heat transfer is very important in steel industries because heating is an economical way to alter a steel component's properties to suit the end user's requirement. Heating interrupted by quenching is a well-known procedure to treat steel components. Much work had been done on changing the mechanical properties of the steel component and little on the fundamental of the process, which is the heat transfer between the quenching medium and steel component. Theoretically, we should be able to predict hardness at any points of steel provided that the relationship between its cooling rate and hardness and the temperature field and its evolution with time are known. Nevertheless, based on metallurgy literature, steel such as hot rolled bar; the allowable hardness is 35HRC. In this paper, we will discuss the approach of finite element software ANSYS Workbench. The ANSYS simulation will be done in 1-D, 2-D and three dimensional. Beside that, the results will be verified by using ANSYS software. From the result, the hardness prediction can be used as a rough guideline to assist the quality engineer prior to production by comparing to a typical standard hardness value set. This will help to reduce the risk of trial and error during production. In this manuscript we discussed the hardness prediction of the quenched bar is difference according to its quenching media which for water have the highest value of 50.32 HRC, 25.46 HRC for stationary oil and the lowest hardness is 14.82 HRC for salt bath. Hardness increase as cooling rate increases. Once the cooling temperature decreased, the steel bar will need to take longer time to quench. This was caused the hardness of steel bar decreased. The effect of quenchants media gives variable of hardness of steel bar by according to its cooling rate. As a conclusion, this paper has fulfilled all the objectives where the affected parameters have been identified to improve the workability of steel bar during cooling by using ANSYS software [ 1-D, 2-D and 3-D] owing to lower cost, safer and reliable procedure than experimental work.