During puncture tests of Japanese cereal snacks, the force increases and decreases alternately. We herein compare the force–deformation curves recorded by two different testing machines and show that the number of changes in the curves depends on the testing machine. Thus, it is impossible to compare results obtained using different instruments. By removing the higher‐frequency components of the force–deformation curves, small events are easily missed. The number of large events decreases when lower‐frequency components are eliminated. This suggests the importance of providing the information on the frequency range of the testing machines. Nevertheless, the number of large force changes is similar between the examined machines. To model the size–frequency distribution, we selected a parametric probabilistic model from among the Weibull, exponential and Pareto distributions using Akaike information criterion and found that the Weibull or exponential distributions have a fit better than the Pareto distribution.
PRACTICAL APPLICATIONS
The methods developed in this work can be used to evaluate the quality of crisp snack food. By analyzing the samples obtained from a lot, samples with poor texture because of abnormal moisture levels and/or ingredients can be discerned and can be used to accept or reject the lot.