The cooling time of an injection molded part directly affects the production cost and efficiency. In order to accurately estimate the cooling time, it is necessary to define the ejection condition. One common criterion for part ejection is when the mean or maximal temperature is equal or less than an estimated ejection temperature, T-e [1]. But in industrial practice, the ejection condition is determined by the part＇s deformation or warpage due to ejection forces and subsequent cooling [2]. This research uses a combination of experimental, analytical, and statistical methods to discuss different ejection criteria and their rationality thereby developing a more effective criterion based on part stiffness. This stiffness criterion facilitates more accurate estimation of cooling rime at early stages of product development. This knowledge will help the designer to improve the design and reduce the cost by increasing the production efficiency while simultaneously ensuring injection molded part quality.