Ross Hawlins had done it all at Hawlins Manufacturing company founded by his grandfather 63 years ago. Among his many duties, Ross oversaw all the plant’s operations, a task that had grown in responsibility given the company’s rapid growth over the past three decades. When Ross’s grandfather founded the company, there were only two manufacturing sites. Expansion and acquisition of competitors over the years had caused that number to grow to more than 50 manufacturing plants in 18 states.
Hawlins had a simple process that produced only two products, but the demand for these products was strong, and Ross had spent millions of dollars upgrading his facilities over the past decade. Consequently, most of the company’s equipment was less than 10 years old on average. Hawlins’s two products were made for local markets because prohibitive shipping costs prevented shipping them long distances. Product demand was sufficiently strong to support two manufacturing shifts (day and night) at every plant, and every plant had the capability to manufacture both products sold by Hawlins. Recently, the management team at Hawlins noticed that there were differences output levels across the various plants. They were uncertain what, if anything, might explain these differences. Clearly, if some plants were more productive than others, there might be some meaningful insights that could be standardized across plants to boost overall productivity.
Ross asked Lisa Chandler, an industrial engineer at the company’s headquarters, to conduct a study of the plant’s productivity. Lisa randomly sampled 159 weeks of output from various plants together with the number of plant employees working that week, the plants’ average age in years, the product mix produced that week (either product A or B), and whether the output was from the day or night shift. The sampled data are in the file Hawlins Manufacturing. The Hawlins management team is expecting a written report a expecting a written report and a presentation by Lisa when it meets again next Tuesday.
Required Tasks: In your report
Identify the primary issue of the case. Identify the dependent and independent variables.
Briefly summarize the data (include an explanation of the results of descriptive statistics of the data, variables included, how the data was collected, and any pertinent information about the data available in the case study)
Dummy variables will have to be created for two of the dependent variables. Explain what dummy variables are, how they are created, and what role they play in regression estimation.
Describe the method that can lead to the following results table.
Justify the statistical model used to produce the following table of results for the case.
Describe which variables are statistically significant. What values provide this information? Comment on the overall fit of the model? Is there any need for further improvement? What values in the results table offer information about the suitability of the overall model?
Provide a short report that describes your analysis and explains in management terms the findings of your model. Be sure to explain which variables, if any, are significant explanatory variables. Provide a recommendation to management. The report must consider practical recommendation(s) in the light of the results of the statistical analysis.
SUMMARY OUTPUT
Regression Statistics
Multiple R 0.886131
R Square 0.785227
Adjusted R Square 0.779649
Standard Error 114.62
Observations 159
ANOVA
df SS MS F Significance F
Regression 4 7397040 1849260 140.7594 2.24E-50
Residual 154 2023212 13137.74
Total 158 9420251
Coefficients Standard Error t Stat P-value Lower
95% Upper 95%
Intercept 827.2881 150.0717 5.5126 1.46E-07 530.8233 1123.7529
Number of Employees 8.8730 0.8062 11.0062 3.77E-21 7.2804 10.4657
Avg Age of Plant (Years) -6.1680 4.3710 -1.4111 0.1602 -14.8028 2.4669
Shift (Day or Night ) 54.9363 35.3308 1.5549 0.1220 -14.8594 124.7319
Product Mix (A or B) 133.8896 35.0992 3.8146 0.0002 64.5514 203.2277
