Incompressible Smoothed Particle Hydrodynamics (ISPH) is an effective method for simulating complex free-surface flows. Significant efforts have been devoted to improving the accuracy of spatial discretization in particle methods, including high-order schemes such as SPH(2) [1]. In addition to these developments, time-marching approaches known as split-type algorithms have been proposed to reduce algorithm-dependent errors inherent to particle methods, thereby improving temporal accuracy [2]. Among various time-integration schemes, the classical Runge–Kutta method [3] is a well-established approach for achieving high-order accuracy. However, when applied to ISPH, it requires solving the computationally expensive Pressure Poisson Equation at every intermediate stage, leading to a trade-off between numerical accuracy and computational efficiency. To overcome this limitation, this study proposes an efficient time-integration framework for ISPH based on the Fast Projection method [4], combined with the split-type algorithm to preserve high spatial accuracy. In the proposed approach, pressure at intermediate stages are evaluated using an approximated pressure field extrapolated from previous time steps. As a result, the PPE is solved only once at the final stage to strictly enforce incompressibility. This strategy enables high-order time integration while keeping computational costs comparable to conventional single-stage schemes. The proposed method is validated through the Taylor–Green vortex flow problem, under both Eulerian and Arbitrary Lagrangian–Eulerian (ALE) descriptions. Numerical results show that the proposed framework significantly reduces numerical errors compared to conventional first-order explicit Euler integration, maintaining computational efficiency. Overall, the method provides a robust and efficient framework for high-accuracy particle-based flow simulations.