Academic Integrity: tutoring, explanations, and feedback — we don’t complete graded work or submit on a student’s behalf.

The manager of a plate steel fabricator plant wants to develop a standard data s

ID: 3208764 • Letter: T

Question

The manager of a plate steel fabricator plant wants to develop a standard data system to

estimate the time required to arc weld the plates together to form the welded assembly.

The standard data system will be based on the use of a mathematical formula in which the

arc welding time depends on three variables in the job: (1) the length of the weld path, (2)

the plate thickness, and (3) the number of parts in the welded assembly. For a given

assembly, the plate thickness would be the same for all parts. Because the plant

specializes in low carbon plate steel, the welding voltage, current, and other machine

settings are constant for all jobs to be included in the coverage of this standard data

system. The data that have been collected from previous direct time studies are presented

in the table below.

Develop an equation to predict the welding time based on length, thickness, and

number of parts using linear regression in Excel. You may not need to include

every variable. Compute a regression equation r value (correlation coefficient) for

the final equation.

Do you think the equation fits the data well, based on the r value? (Give a yes/no

and a reason.)

Define the range of values for length, thickness, and number of parts for which

your equation is valid

Length L (m) Thickness t (mm) Number of parts np Welding Time T (min) 0.68 6 2 2.47 3.26 8 5 10.85 1.45 6 3 5.21 2.10 10 7 8.69 2.66 6 4 8.71 4.75 6 8 16.35 0.81 10 3 3.30 1.82 8 4 7.01 1.49 8 5 5.61 2.05 6 4 6.99 3.79 6 2 10.56 2.92 8 3 9.65 4.22 10 7 14.65 3.66 10 5 12.45 1.78 8 4 6.61

Explanation / Answer

Result:

Develop an equation to predict the welding time based on length, thickness, and

number of parts using linear regression in Excel. You may not need to include

every variable. Compute a regression equation r value (correlation coefficient) for

the final equation.

Regression Analysis

0.995

Adjusted R²

0.994

n

15

R

0.998

k

3

Std. Error

0.308

Dep. Var.

Welding Time T (min)

ANOVA table

Source

SS

df

MS

F

p-value

Regression

215.6464

3  

71.8821

755.58

5.14E-13

Residual

1.0465

11  

0.0951

Total

216.6929

14  

Regression output

confidence interval

variables

coefficients

std. error

   t (df=11)

p-value

95% lower

95% upper

Intercept

-0.8193

0.4217

-1.943

.0780

-1.7474

0.1088

Length L (m)

2.7777

0.0821

33.846

1.79E-12

2.5971

2.9584

Thickness t (mm)

0.0830

0.0547

1.519

.1571

-0.0373

0.2033

Number of parts np

0.4208

0.0606

6.948

2.43E-05

0.2875

0.5541

The regression equation

Welding Time =-0.8193 +2.7777 * Length L (m)+ 0.0830 *Thickness t (mm)+ 0.4208 *Number of parts np

R=0.998

Do you think the equation fits the data well, based on the r value? (Give a yes/no and a reason.)

Yes. R square =0.995.

99.5% of variance in welding time is explained by the model. The equation fits the data well.

Define the range of values for length, thickness, and number of parts for which your equation is valid

The equation is valid for,

length (0.68 to 4.75), thickness ( 6 to 10), and number of parts (2 to 8).

Regression Analysis

0.995

Adjusted R²

0.994

n

15

R

0.998

k

3

Std. Error

0.308

Dep. Var.

Welding Time T (min)

ANOVA table

Source

SS

df

MS

F

p-value

Regression

215.6464

3  

71.8821

755.58

5.14E-13

Residual

1.0465

11  

0.0951

Total

216.6929

14  

Regression output

confidence interval

variables

coefficients

std. error

   t (df=11)

p-value

95% lower

95% upper

Intercept

-0.8193

0.4217

-1.943

.0780

-1.7474

0.1088

Length L (m)

2.7777

0.0821

33.846

1.79E-12

2.5971

2.9584

Thickness t (mm)

0.0830

0.0547

1.519

.1571

-0.0373

0.2033

Number of parts np

0.4208

0.0606

6.948

2.43E-05

0.2875

0.5541