This paper focuses on air bearing surface (ABS) design optimizations in order to reduce the lift-off force during the unloading process while satisfying the desired static flying performances. Since it takes a huge amount of computational time to solve time-dependent dynamic load/unload (L/UL) equations, an approximate lift-off force is created as a function of the air bearing suction force and flying attitude parameters by the kriging method. The design framework is employed in wrapping effectively and connecting the kriging model and the static analyzer to the optimizer. An optimization problem is formulated to minimize the amplitude of the lift-off force during the unloading process while keeping the flying height, pitch, and roll angles within suitable ranges over the entire recording band as well as reducing the possibility of slider-disk contact in steady state. Then, two different sizes of slider models are optimally designed for L/UL applications with 1-in disk drive. The L/UL simulation results show that the optimized ABS designs have reduced the lift-off force in the loading process by approximately 62% and 11% for pico and femto design respectively, while satisfying desired static flying performance. In addition, results demonstrated that the optimum slider incorporated with the suspension were not only properly unloaded onto the ramp but also smoothly loaded onto the rotating disk. Therefore, it is believed that the proposed design approach works efficiently in ABS designs for L/UL applications.