A Case Study of an Experimental Design to Laboratory Process Improvement

Abstract

In a chemical analysis laboratory it is extremely important to identify and quantify the contribution of the stages of the analytical process in the final variability of the results, thus facilitating the directing of efforts in the search for continuous improvement. In this type of laboratory, the improvement corresponds to the reduction of variability, which guarantees greater reliability of the analytical results, and one of the ways to achieve this reduction is to carry out an experiment. In this work, a practical example of a chemical industry will be presented in which the experiment allowed to quantify the contribution of the steps: (I) preparation of the samples, (II) of the different technicians and (III) the measurement itself. An experiment with random factors and a hierarchical factor (nested factor) was used and the results were analyzed using an ANOVA. From the analysis of the results, it was identified that the major contributor to the final variability was the preparation of the samples. Efforts were directed at this stage, with a review of the procedure and training of technicians, which provided a general improvement in the reliability of the analytical results. With a second similar experiment it was possible to prove the effectiveness of the improvement actions.