Soaking is performed in soybeans to promote water absorption for late processing. This processing needs to determine the minimum time to obtain the maximum absorption to turn economically acceptable the production process, but the traditional methods can not be used to establish this "optimum", then statistical sensory methods were used. These methods have implicit two concepts : the sensory judgement and the experimental error concept. Twenty cultivars of soybean grown in Brazil were evaluated by a panel and an expert in experimentation to visually establish "optimal points" in three kinds of graphics. The consistency of the panel between the judgements of joints of full points and joints of quadratic points of was 80.43% (p < 0.05), and the consistency of the expert in the same graphics was 83.22 %. The consistency between the panel and the expert for these graphics were 77.01 % and 83.22% (p < 0.05). There was no significant difference in the efficiency, expressed in relative quadratic errors, of the judgements between the panel and the expert. The expert judged that cultivar MSBR-21 (Buriti) saturated, the panel disagreed. The OCEPAR-4, Davis and IAC-12 cultivars showed the lowest minimum time to reach saturation (from 7.5 to 9.5 hours) and MSBR-21 (Buriti) and Bragg showed the greatest (>14 hours).
sensory evaluation; graphics; water absorption; soybean; optimization; soaking
