KeSTE

KeSTE (Kernel Sampling Technique Evaluation) is a program to evaluate sampling techniques employed for the detection of impurities in large kernel lots, either virtually created or actually sampled (see Actual samples evaluation). In the first case, the program takes a unique approach to sampling, based on a two-step procedure. First, the program allows creating populations (i.e., virtual kernel lots) with user defined characteristics: total number of seeds (lot size), degree of contamination (percentage of impurities), number of different contaminants present in the lot (one vs. multiple sources of impurities), and level of spatial aggregation of the contaminant/s (degree of lot heterogeneity). Several instances of a population can be created, differing because of the distribution of impurities within the constraints imposed. Second, the created population can be sampled using either a random or a systematic sampling scheme. The user can specify a range of incremental samples and a range of increment sizes (number of kernels per increment) to be sampled. The evaluation of actual samples provides estimates of the sampling error (expressed as standard deviation) associated to different sampling numerosities (i.e., number of increments collected to produce the bulk sample). Also in this second case the program takes a unique approach based on a two-step procedure: first, contamination estimates obtained for each individual increment are collected and used to measure their variability as function of number of sampled increments. Second, the maximum possible sampling error is estimated for any sampling numerosity (i.e., number of collected increments) by fitting an exponential model. Response surfaces are built to provide estimates of possible sampling errors as function of the different sampling scheme considered, and to identify proper sampling techniques. The program was developed using MS Excel, which is needed to run the program. Intended use of the current release of the program is explained in the page What use for KeSTE?.