NAME: Fan Data TYPE: Designed experiment SIZE: Data for four experiments (4 variables with 8 observations, 4 variables with 16 observations, 4 variables with 32 observations, 4 variables with 64 observations) DESCRIPTIVE ABSTRACT: This dataset contains the results of a experiment designed to determine the best of two levels for each of three factors used in manufacturing industrial fans: (1) the shape of the hole in the fan blade (spyder); (2) the type of assembly method; and (3) the shape of the hub or barrel. Destructive testing was used to break the spyder loose from the barrel for each of the fans. The breaking torque was measured in foot-pounds. SOURCES: The data represent the results of an actual industrial experiment. Company representatives have asked that the name of the company not be revealed. VARIABLE DESCRIPTIONS: Data columns 1, 5, 9, and 13 are values for the two types of hole tested (coded -1 and 1) Data columns 2, 6, 10, and 14 are values for the two types of assembly method used (coded -1 and 1) Data columns 3, 7, 11, and 15 are values for the two barrel shapes used (coded -1 and 1) Data columns 4, 8, 12, and 16 are the breaking torques in foot-pounds for the various factor combinations Data columns 1 - 4 give the results of the experiment with one observation for each factor combination Data columns 5 - 8 give the results of the experiment with two replications for each factor combination Data columns 9 - 12 give the results for the experiment with four replications for each factor combination Data columns 13 - 16 give the results for the experiment with eight replications for each factor combination SPECIAL NOTES: Columns are delimited with spaces. STORY BEHIND THE DATA: A mid-western company that manufactures large industrial fans recently acquired a smaller competitor. The companies used different manufacturing methods to attach the fan blades (spyders) to the fan hubs. In an attempt to determine the best manufacturing process, an experiment was designed to investigate: (1) two types of hole in the spyder, (2) two hub (or barrel) shapes, and (3) two assembly methods. Destructive testing was used to break the spyder from the hub and the torque required to break the part was measured in foot-pounds. The full factorial design resulted in eight factor combinations. Since the fans had to be destroyed in order to test the effectiveness of the manufacturing methods, the company's engineers wanted to use as few fans as possible in the experiment. The minimum number of fans necessary to estimate the main effects and the two factor interactions was eight. The authors encouraged the engineers to use a larger sample size than one observation per factor combination. This dataset can be used to show how smaller sample sizes may lead to incorrect conclusions about the process under investigation. If the engineers had conducted only the original experiment, they would have missed an opportunity to significantly strengthen the fans manufactured by the company. PEDAGOGICAL NOTES: Analysis of Variance (ANOVA) can be used to determine which factors are important in producing a stronger fan. Students analyzing the first four columns of data would conclude that only the barrel shape was an important factor. The average breaking torque of the best barrel shape was 176 foot-pounds. Conversely, an analysis of columns 5 through 8 (two replications of each factor combination) would suggest that all three factors are important in producing a strong fan. Students will observe that, by increasing the sample size to 32 and finally 64 observations, all main effects and all interactions become significant. In fact, the average breaking torque for the best combination of the factor levels is 196 foot-pounds. By identifying the best combination of factor levels and using those settings in the manufacturing process, the company can produce a stronger fan than otherwise possible. SUBMITTED BY: LeRoy A. Franklin Rose-Hulman Institute of Technology 550 Wabash Avenue Terre Haute, IN USA LeRoy.A.Franklin@Rose-Hulman.edu Belva J. Cooley University of Montana School of Business Administration Missoula, MT 59812 belva.cooley@business.umt.edu