Efficient random walk algorithm for computing conductivity in continuum percolation systems

Random walks can be used to obtain the diffusion constant and thus the conductivity for continuum percolation problems. This paper presents an efficient algorithm that allows walkers to move very large distances in one step. The algorithm uses a first‐passage time distribution for d‐dimensional spherical surfaces. Results are given for overlapping nonconducting disks in two dimensions. Depending on the density of disks, it is found that the present algorithm is about 5 to 50 times faster than an equivalent algorithm using fixed step lengths.