Lab 3 - Engineering Anthropometry

.docx

School

Northeastern University *

*We aren’t endorsed by this school

Course

7315

Subject

Industrial Engineering

Date

Dec 6, 2023

Type

docx

Pages

Uploaded by MegaMonkey3868

IE 7315 Human Factors Engineering Lab Series – Prof. Y. Lin Lab 3 - Engineering Anthropometry Introduction: To design any new product, it is important for the engineer to understand the key characteristics of their target populations. A field that can be useful is anthropometry, which involves the scientific study of the measurements and proportions of the human body. This lab examines regression equations regarding human anthropometric data, based on the methodology established by Karl H.E. Kroemer. During the lab, you will measure the common body dimensions via practice. You will also formulate your own equations using a Multiple Linear Regression (MLR) on the class data. MLR is a statistical model that assumes that a dependent variable ( predicted value ) can be estimated by decomposing into a weighted sum of causal factors ( predictors ) plus an error term which takes the form: Objectives: ● Understand the theory and the application of ergonomics ● Use appropriate techniques and equipment to take body measurements ● Learn how to use correlation and perform data analysis utilizing MLR Apparatus: Flexible Measuring Tape, Tape Measure, Ruler, Weighing Scale Methods: 1. Measure your body dimensions listed in Table 1. You can also welcome to use your roommate’s, friend’s body dimensions in Table 1. 2. In the Data Analysis session, use the dataset posted by TA to work on Step 1, 2, and 3, and use the data in Table 1 for Step 4. 3. Finish the rest of the report. 1

Table 1 Form for data recording: Body Measurements Unit of Measurement No. in Figure Subject Value Stature Height cm 1 Eye height (standing) cm 2 Shoulder Circumference cm 3 Chest Circumference cm 4 Chest Depth cm 5 Sitting Height cm 6 Eye height (sitting) cm 7 Elbow Rest Height Sitting cm 8 Popliteal Height cm 9 Buttock-Knee Length cm 10 Weight kg --- The numbers in Table 1 Column 3 correspond with the descriptions and images on page 3. 2

Anthropometric Measurements Facilitator 1. Stature/Stature height: Stand up straight with heels and back against a wall. Measure the distance from the top of head to the floor. 2. Eye Height (standing): Stand up straight with heels and back against a wall. Measure from the bridge of the nose (in line with pupils) to the floor. 3. Shoulder Circumference : Stand up straight with arms resting beside body. Measure circumference around shoulders 4. Chest Circumference: Wrap a flexible measuring tape level around the torso, parallel to the ground and in line with the nipples. 5. Chest Depth : Sit up straight on a level surface with their buttocks and back against a wall. Place a ruler flat across their chest. Measure the distance between the ruler and the wall. 6. Sitting height : Sit up straight on a level surface and measure the distance from top of head to the level surface. 7. Eye Height (sitting) : Sit up straight on a level surface. Measure from the bridge of the nose (in line with the pupils) to the desk. 8. Elbow Rest Height Sitting : Sit up straight on a level surface. Bend arm at a 90 degree angle. Measure the distance from bottom of elbow to the level surface. 9. Popliteal Height : Do a “wall sit” against the wall. Measure from the floor to the underside of the leg at a point approximately 5 inches from the back of the knee. 10. Buttock-Knee Length: Sit up straight on a level surface. Measure from the back of the buttocks to the front of the knee. Other Measurements: Not shown in Figures Weight : Weight should be recorded in kg. 3

Data Analysis 1. Step 1 : The correlation coefficient, r, tells you how strong a linear relationship is between two variables. In step 1, you need to develop a correlation coefficient matrix using Excel or Minitab. Helpful links are provided. Excel: http://www.excel- easy.com/examples/correlation.html Minitab: https://online.stat.psu.edu/stat501/lesson/creating-correlation-matrix 2. Step 2 : you need to identify all body measurement variables that are good-predicted candidates . To identify them, you can: 1) highlight one correlation coefficient (the absolute value of the cell in the matrix ) which are above 0.7 as yellow , 2) then highlight the body measurement variable (the row’s and column’s name) as red. 3) iterate 1) and 2) until all correlation coefficient values larger than 0.7 are selected. Your matrix’s format should look similar to Fig. 1 Matrix Example. 4) the good-predicted candidates are highlighted in red. According to Kroemer’s theory [1], it is a useful practice in anthropometry and in human engineering to require that the predictor variable explains at least 50% of the variation of the predicted value, which means that the correlation coefficient must be at least 0.7. To identify the good candidates, this “ 0.7 convention ” is important for basing decisions on a correlation and, relatedly, for the development and use of regression equations, which express the average of one variable as a function of another variable. Stature Height Eye Height(sta nding) Shoulder Circumfere nce Chest Circumfere nce Chest Depth Sitting Height Eye Height(sitti ng) Elbow Rest Height Sitting Popliteal Height Stature Height 1 Eye Height(sta nding) 0.9694526 1 Shoulder Circumfere nce 0.2216678 0.2352518 1 Chest Circumfere nce 0.1171283 0.149894 0.5220091 1 Chest Depth 0.5084341 0.4557278 0.3409543 0.3103507 1 Sitting Height 0.6194507 0.5703459 0.0225733 0.2838817 0.4814886 1 Eye Height(sit ing) 0.6004135 0.5669637 -0.0765091 0.262292 0.3959144 0.9227959 1 Elbow Rest Height Sitting 0.3049392 0.276114 0.1356201 -0.1386274 0.1692876 0.1868201 0.2331008 1 Popliteal Height 0.7660161 0.7074816 0.2781535 0.0947447 0.1792383 0.4520044 0.404479 0.058339 1 Figure 1 Correlation Coefficient Matrix Example 3. Step 3 : Select three good-predicted candidates from step 2, treat one candidate as the predicted and the other two as predictors, and run a multiple linear regression . Iteratively run the MLP on the other two combinations to create the regression equations. An example of regression equations is given in Table 2. You can lower the correlation coefficient bar from 0.7 to 0.65 unless the data variation is too large to select three good-predicted candidates.

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help