Tomohiro Tasaki* and Kohei Urano*
* Laboratory of Safety and Environmental Engineering, Yokohama National University

+ Correspondence should be addressed to Kohei Urano:

(79-5 Tokiwadai Hodogaya-ku Yokohama city 240-8501 Japan)

Abstract
Conventional equations to estimate sampling variability in leaching tests and content tests are insufficient because these equations are based on the assumption that wastes consist of a single composition. In this paper, the influence of the variability of sampled compositions on the variability of test data of multicomposition wastes, which consist of plastics, fabrics, metals, etc., has been statistically analyzed. Results show that the equations that estimate the coefficient of variation of sampling were derived as functions of sampling weight, particle diameter, average apparent density, which could be measured easily, and a newly defined sampling parameter, which represented a diversity of compositions and was inherent to waste. Using the derived equations, the coefficients of variation of test data for some typical weights, numbers, distributions of particle diameter and average apparent densities of samples were estimated and discussed.

Key words: sampling, multicomposition wastes, variability, leaching test, content test