You are on page 1of 3

SEITZ CORPORATION: PRODUCING QUALITY GEAR-DRIVEN AND LINEAR-MOTION PRODUCTS

The Seitz Corporation, a QS 9000 certified organization based in Torrington, Connecticut, is a leading
designer and manufacturer of thermoplastic motion control systems and components and an industry
leader in plastics and gear trains. Founded in 1949 by the late Karl F. Seitz, the company began as a small
tool-making business and grew slowly. In the late 1960s, the company expanded its services to include
custom injection molding. As their customer base grew to include leading printer manufacturers, Seitz
developed and patented a proprietary line of perforated form handling tractors. Utilizing its injection-
molding technology, the company engineered an all-plastic tractor called Data Motion, which replaced
costly metal versions. By the late 1970s, business was booming, and Data Motion had become the
worldwide industry leader.
In the 1980s, foreign competition entered the business equipment market, and many of Seitz’s
customers relocated or closed shop. The ripple effect hit Seitz as sales declined and profits eroded.
Employment at the company dropped from a high of 313 in 1985 to only 125 in 1987. Drastic changes
had to be made at Seitz.
To meet the challenge in 1987, Seitz made a crucial decision to change the way it did business.
The company implemented a formal five-year plan with measurable goals called “World- Class
Excellence Through Total Quality.” Senior managers devoted many hours to improving employee
training and involvement. New concepts were explored and integrated into the business plan. Teams
and programs were put into place to immediately correct deficiencies in Seitz’s systems that were
revealed in customer satisfaction surveys. All employees from machine operators to accountants were
taught that quality means understanding customers’ needs and fulfilling them correctly the first time.
Once the program started, thousands of dollars in cost savings and two new products generating
almost $1 million in sales resulted. Annual sales grew from $10.8 million in 1987 to $19 million in 1990.
Seitz’s customer base expanded from 312 in 1987 to 550 at the end of 1990.
In the decade of the 1990s, Seitz continued its steady growth. By 1999, Seitz was shipping
products to 28 countries, and customers included Xerox, Hewlett Packard, Canon, U.S. Tsubaki, and
many more worldwide. By 1998, sales topped the $30 million mark. In January 2000, the company
established the Seitz Motion Control Systems Co., Ltd., in Changzhou, China, about 150 miles northwest
of Shanghai, to provide product and tooling design engineering, sourcing and supply chain management
services, and contract manufacturing. The Seitz Corporation headquarters is located in Torrington,
Connecticut in an 80,000 square foot facility with over 150 associates, 50 molding machines ranging in
size from 35 tons to 770 tons, an in-house tooling department, and a corporate staff. While the primary
core competency of the Seitz Corporation is rotary and linear motion control, making them an industry
leader in plastics and gear trains, Seitz offers a full range of product design and tooling services.

Discussion

1. Seitz’s list of several hundred business-to-business customers continues to grow. Managers would like
to know whether the average dollar amount of sales per transaction per customer has changed from last
year to this year. Suppose company accountants sampled 20 customers randomly from last year’s
records and determined that the mean sales per customer was $2,300, with a standard deviation of
$500. They sampled 25 customers randomly from this year’s files and determined that the mean sales
per customer for this sample was $2,450, with a standard deviation of $540. Analyze these data and
summarize your findings for managers. Explain how this information can be used by decision makers.
Assume that sales per customer are normally distributed.
Answer:
Last Year’s Mean Transactions: n1 = 20; x1 = 2300; s1 = 500
This Year’s Mean Transactions: n2 = 25; x2 = 2450; s2 = 540

H0: μ1 - μ2 = 0
Ha: μ1 - μ2 ≠ 0

df = n1 + n2 - 2 = 20 + 25 - 2 = 43;  = .05; /2 = .025

t.025,43 = ± 2.021

t = (2300-2450)-0 / [(500)2 (19) + (540) 2 (24) * 1/20+1/25 = -0.96

Because t = -0.96 > t.025, 43 = -2.021, the decision is to fail to reject the null hypothesis. There is not
enough evidence here to say that there is any difference in the average dollar amount of sales between
this year and last.

2. One common approach to measuring a company’s quality is through the use of customer satisfaction
surveys. Suppose in a random sample, Seitz’s customers are asked whether the plastic tractor produced
by Seitz has outstanding quality (Yes or No).Assume Seitz produces these tractors at two different plant
locations and that the tractor customers can be divided according to where their tractors were
manufactured. Suppose a random sample of 45 customers who bought tractors made at plant 1 results
in 18 saying the tractors have excellent quality and a random sample of 51 customers who bought
tractors made at plant 2 results in 12 saying the tractors have excellent quality. Use a confidence
interval to express the estimated difference in population proportions of excellent ratings between the
two groups of customers. Does it seem to matter which plant produces the tractors in terms of the
quality rating received from customers? What would you report from these data?

Answer:
n1 = 45; x1 = 18
n2 = 51; x2 = 12

ˆp1 = 18/45 = .4000


ˆ p2 = 12/51 = .2353

(.40-.2353) ± 1.96 (.4)(.6)/45 + (.2353)(.7647)/51 = .1647 ± .1845


-.0198 < p1 - p2 < .3492

The point estimate of the difference in quality of tractors at the two plants is 16.47%. However, due to
the relatively small samples, the error of the interval is 18.45% which is greater than the point estimate.
Combining the error of the interval with the point estimate results in the confidence interval shown
above. Note that zero is in the interval indicating that there is a possibility that there is no difference in
the quality ratings of tractors produced at the two plants. If this were a hypothesis testing problem, then
the decision would be to fail to reject the null hypothesis based on the confidence interval’s inclusion of
zero.

3. Suppose the customer satisfaction survey included a question on the overall quality of Seitz measured
on a scale from 0 to 10 where higher numbers indicate greater quality. Company managers monitor the
figures from year to year to help determine whether Seitz is improving customers’ perceptions of its
quality. Suppose random samples of the responses from 2008 customers and 2009 customers are taken
and analyzed on this question, and the following Minitab analysis of the data results. Help managers
interpret this analysis so that comparisons can be made between 2008 and 2009. Discuss the samples,
the statistics, and the conclusions.

Answer:
2008 vs. 2009:
t = -1.83 with a p-value of .07. This is not significant at  = .05. There is no significant difference in the
mean ratings between 2008 and 2009. This is underscored by the confidence interval that includes zero.
However, if  = .10 were used, there would be a significant difference. Examining the means reveals that
the mean score for 2009 was higher. The sample sizes were 75 for 2008 and 93 for 2009.

4. Suppose Seitz produces pulleys that are specified to be 50 millimeters (mm) in diameter. A large batch
of pulleys is made in week 1 and another is made in week 5. Quality control people want to determine
whether there is a difference in the variance of the diameters of the two batches. Assume that a sample
of six pulleys from the week 1 batch results in the following diameter measurements (in mm): 51, 50, 48,
50, 49, 51.Assume that a sample of seven pulleys from the week 5 batch results in the following
diameter measurements (in mm): 50, 48, 48, 51, 52, 50, 52. Conduct a test to determine whether the
variance in diameters differs between these two populations. Why would the quality control people be
interested in such a test? What results of this test would you relate to them? What about the means of
these two batches? Analyze these data in terms of the means and report on the results. Assume that
pulley diameters are normally distributed in the population.

You might also like