It is essential to consider the fluidity of a debris flow front when calculating its impact. Here we flume-tested mono-granular and bi-granular debris flows and compared the results to those of numerical simulations. We used sand particles with diameters of 0.29 and 0.14 cm at two mixing ratios of 50 percent:50 percent and 30 percent:70 percent. Particle segregation was recorded with a high-speed video camera. We evaluated the fronts of debris flows at 0.5-second intervals. Then we numerically simulated one-dimensional debris flows under the same conditions and used the mean particle diameter when simulating mixed-diameter flows. For the mono-granular debris flows, the experimental and simulated results showed good agreement in terms of flow depth, front velocity, and flux. However, for the bi-granular debris flows, the simulated flow depth was less, and both the front velocity and flux were greater than those found experimentally. These differences may be attributable to the fact that the dominant shear stress was caused by the concentration of smaller sediment particles in the lower flow layers; such inverse gradations were detected in the debris flow bodies. Under these conditions, most shear stress is supported by smaller particles in the lower layers; the debris flow characteristics become similar to those of mono-granular flows, in contrast to the numerical simulation, which incorporated particle segregation with gradually decreasing mean diameter from the front to the flow body. Consequently, the calculated front velocities were underestimated; particle segregation at the front of the bi-granular debris flows did not affect fluidity either initially or over time.