A feasibility study of a system for computer-aided selection of materials for industrial production

This study was a comparative analysis of two methods of selecting materials for industrial production. The specific objectives were to find out if (a) a computer-aided materials selection system would select materials which were not significantly different from materials chosen by practicing professional engineers; and (b) the time required by the computer system to select materials would be significantly less than the time required by the engineers. The criterion measures were (a) the closeness of fit of properties to design specifications for each material selected; and (b) the amount of time consumed by each subject in the selection of each material. The research design selected was a pre-experimental Static Group Comparison with repeated (time series) observations. Five plastics selection engineers were identified from throughout Eastern Iowa, and three selection problems were submitted to each of the five engineers and to the computer selection system. The resultant data were analyzed using a Friedman Two-Way Analysis of Variance and the Mann-Whitney U test. The data supported the hypothesis of no difference in the materials selected (p = .7872). However, the time required to select materials was significantly less (p = .0355) for the computer than for the five engineers. On the basis of the analysis, it was concluded that the computer selected materials which were not significantly different from those selected by the five engineers, but that the computer took significantly less time to make the selections. The implications of the results of this study are that, for the field of materials selection, computer technology seems to offer a real alternative to manual searches of selection data. Further research is recommended to develop larger and more complete data bases, and to develop better modeling techniques to allow the computer to do a better job of selecting the individual materials