Manufacturing Technology Today

M. Vijay Kumar ¹, W. Yerriswamy ¹

Affiliations
1 JSS Academy of Technical Education, Bangalore, IN

Abstract

An important part of many quality improvement and quality development programs is an evaluation of the effect of process variables upon performance. The product quality is depending on number of independent factors. Their effects on dependent factors are evaluated through empirical investigation. In practice, a small number of controllable variables contribute to a vital share of the effect of the product quality. These variables do not necessarily produce a constant effect on the product. The question would therefore arise as to how efficiently and economically the contribution of each of these factors can be assessed individually and also collectively to produce the total effect on the product performance. An approach that fulfills these requirements is available in the statistically designed experiments. The statistically designed experiment permits simultaneous consideration of all the possible factors that are suspected to have a bearing on the problem under consideration. Even a limited number of experiments would enable experimenter to uncover the vital factors on which further trails would lead the researcher to track down their most desirable combination which will yield the expected results. Scanning a large number of variables is one of the objectives that a statistically designed experiment would fulfill in many problem situations. The purpose of this study is to develop a mathematical model to predict the task time taken by the operator to assemble the needle and thread by varying the working environmental conditions like temperature, light and noise. The adequacies of the model are then evaluated using MINITAB statistical software and analysis of variance (ANOVA) technique. The model developed is checked for its adequacy. Results of confirmation experiments showed that the model can predict the optimum task time with reasonable accuracy.

Keywords

Design of Experiment, ANOVA, Experiments, Treatment, Replication.

Full Text