H.

14 SUPPLEMENT H > MEASURING OUTPUT RATES <

> PROBLEMS <

1. During a time study in a machine shop, five observations b. Determine the sample size necessary if the estimate
of a milling operation performed by an operator whose of the select time for the work elements is to be within
rating factor is 95 percent yielded the following times (in 3 percent of the true mean 95 percent of the time.
minutes): 40, 48, 48, 46, and 42. The allowance for this c. Is the sample size chosen to determine the standard
type of operation is 15 percent. time adequate? If not, how many additional cycles
a. Determine the normal time for this operation. should be observed?
b. Determine the standard time for this operation. 3. A time study involving the preparation of hamburgers at
2. The manager of Stetson and Stetson Company is trying to Bill’s fast-food restaurant used the snap-back method to
develop a time standard for the powder filling and pack- obtain the data (in minutes) shown in Table H.2. Allow-
ing operation. This operation has five work elements, ances typically constitute 15 percent of normal time. The
each of which is performed once every cycle. The schedule calls for 300 hamburgers to be prepared during
allowance for each work element is 18 percent. The oper- the lunch rush. If each part-time employee works 190
ation was studied for 20 cycles, and the following sum- minutes per day, how many employees will be needed?
mary data were obtained: 4. A cook at Bill’s restaurant (see Problem 3) has devised a
new method of quickly flipping and pressing hamburgers
that he believes will save time in cooking the second side of
Work Select Time Standard Deviation hamburgers (work element 3 in Table H.2). The cook asked
Element (minutes) (minutes) a peer to conduct a time study for this work element, with
the results shown in Table H.3. This cook is renowned for
1 0.40 0.021 superior strength and speed in hamburger flipping and
2 0.20 0.011 pressing. The rating factor for this study is 1.2. Allowances
3 0.31 0.018 typically constitute 15 percent of the normal time.
4 0.15 0.005 a. What is the average of select times for revised work
element 3? For the revised normal time?
5 1.25 0.085
b. What is the revised normal time per cycle? The revised
standard time?
a. Determine the standard time for the filling and pack- c. The managers seem very interested in this revised
ing operation. Assume the rating factor = 1.0. method for work element 3. They say that if they could

TABLE H.2 Time Study Results of Hamburger Preparation

Observation

Work Element 1 2 3 4 5 t F RF NT

1. Prepare patty 0.45 0.41 0.50 0.48 0.36 1 0.9

2. Cook first side 0.85 0.81 0.77 0.89 0.83 1 1.2
3. Flip, press, and cook other side 0.60 0.55 0.59 0.58 0.63 1 1.2
4. Assemble 0.31 0.24 0.27 0.26 0.32 1 1.0
Normal time per cycle (NTC) =

TABLE H.3 Work Element 3 Time Study

Observation

Work Element 1 2 3 4 5 6 7 8 9 10

Revised work element 3 0.45 0.31 0.50 0.48 0.39 0.31 0.44 0.29 0.33 0.40
 > PROBLEMS < H.15

be sure the average of the select times for this study 7. A time study has been conducted on a cellular telephone
was within ±13 percent of the true average time for assembly operation. The data shown in Table H.6 (in
this new method, they could afford to buy health minutes) were obtained. A standard time within ±3 per-
insurance for the part-time cooks. How many obser- cent of the true mean with 95 percent confidence is
vations would be required to be 98 percent confident desired.
that the average of select times for this study was a. Calculate the standard time for the assembly
within ±13 percent of the true mean? operations.
d. The cook is not only quick with a spatula but also b. How many more observations will be required?
quick with a calculator and a bit suspicious of man-
agement’s motives. If the average of the select times 8. Consider the recorded observations of 10 cycles
found in part (a) were inflated by 13 percent, how of the cup packaging operation, shown in Fig-
many cooks would Bill’s restaurant need? ure H.4.
5. The information (in minutes) shown in Table H.4 pertains a. Determine the select times for each work ele-
to a package filling operation at the Black Sheep Wool ment, the normal time for the cycle, and the
Company. When three bags are full, the third work ele- standard time per package.
ment involves transporting the three bags down the lane. b. Suppose that we want a sample size that gives an
What is the normal cycle time for this operation? average time within ±5 percent of the true
6. A time analyst for the Super-Fast speedway pit crew average 95 percent of the time. Did we make
observed the mechanic in charge of changing both front enough observations? If not, how many more
tires during a pit-stop practice session. Her job is divided should we make?
into six work elements and a preparation time between c. Suppose that all we wanted was a precision of
drills. The element times (in seconds) for the first six ±10 percent. How many additional observations
cycles are shown in Table H.5. would we need?
a. Calculate the normal time for changing tires.
9. Management in a large hospital is planning to install
b. What sample size is appropriate for estimating the a computer to reduce the time spent by nurses doing
average time for work element 3 within ±1 percent of paperwork. First, management needed to know how
the true mean with 99 percent confidence? much time nurses spend doing paperwork in order to
c. What is the standard time for changing tires if the estimate the potential savings from the computer
allowance is 20 percent? installation. A work sampling study comprising

TABLE H.4 Time Study Results of the Package Filling Operation

Observation

Work Element 1 2 3 4 5 6 7 8 9 10 11 12 F RF

1. Fill bag 0.20 0.22 0.24 0.18 0.20 0.21 0.22 0.19 0.24 0.18 0.19 0.25 1.00 1.2
2. Sew closed 0.40 0.38 0.37 0.41 0.41 0.40 0.36 0.37 0.41 0.42 0.39 0.36 1.00 0.8
3. Transport 0.82 0.84 0.73 0.85 0.33 1.1

TABLE H.5 Pit Crew Time Study Data

Observation

Work Element 1 2 3 4 5 6 F RF

1. Wait for car lift 2.9 3.2 2.6 0.5 1.0

2. Remove lugs 3.3 3.8 3.6 3.8 4.0 4.3 1.0 0.9
3. Switch tires 6.4 7.1 6.8 7.3 6.2 6.4 1.0 1.2
4. Tighten lugs 4.1 3.8 3.5 4.9 3.5 4.2 1.0 0.8
5. Move to right side 3.8 — 4.3 — 3.2 — 0.5 1.2
6. Clear away for drop — 2.1 — 2.7 — 2.0 0.5 0.9
H.16 SUPPLEMENT H > MEASURING OUTPUT RATES <

TABLE H.6 Cellular Telephone Time Study Data

Observation

Work Element 1 2 3 4 5 6 7 8 F RF

1. Assemble unit 0.78 0.70 0.75 0.80 0.79 0.82 0.81 0.80 1.0 1.2
2. Insert batteries 0.20 0.21 0.16 0.19 0.23 0.25 0.24 0.26 1.0 1.0
3. Test 0.61 0.60 0.55 0.57 0.63 0.61 0.62 0.60 1.0 0.9
4. Package 0.41 0.36 0.45 0.37 0.39 0.40 0.43 0.44 1.0 1.1

Operation: Coffee cup packaging Date: 1/23 Observer: B. Larson

Observations
Work Element 1 2 3 4 5 6 7 8 9 10 t F RF

1. Get two t
cartons r 0.48 4.85 9.14 13.53 17.83 0.5 1.05

2. Put liner t
in carton r 0.59 2.56 4.94 6.82 9.25 11.23 13.61 15.50 17.93 19.83 1.0 0.95

3. Place cups t
in carton r 1.33 3.24 5.65 7.51 9.98 11.93 14.29 16.24 18.64 20.55 1.0 1.10

4. Seal carton, t
set aside r 2.43 4.39 6.72 8.60 11.10 13.04 15.38 17.32 19.74 21.68 1.0 0.90
Normal time for cycle:
Allowances (% of total time): 15 % Standard time: minutes per piece

FIGURE H.4

500 observations taken at random over a week yielded 10. The manager of a loading dock is concerned about the time
the following data: spent by his crew in nonproductive activities (e.g., waiting for
paperwork, idle time, etc.). Although he is not sure what the
true proportion of nonproductive time is, he believes that it
Activity Number of Observations is close to 20 percent. If he wants to use work sampling to
estimate this proportion with 95 percent confidence, a maxi-
Attending to patients 180 mum error of 3.5 percent, how many samples should he take?
Moving between stations 40 11. Mayor Jonathan (Johnny) Johnson of Graft City is running
Consulting with doctors 60 for reelection. At a big rally in the city park, volunteers will
assemble signs, “A vote for Johnny is a vote for Graft,” to
Taking a break or idling 50
be placed on front lawns and city property. To ensure that
Doing paperwork 170 the rally will go smoothly, the mayor directs the public
works department to conduct a preliminary time study to
estimate the rate of work and number of city workers that
a. Estimate the proportion of time that the nurses spend will be needed to make signs during the rally. The results
doing paperwork. (in seconds) are shown in Figure H.5.
b. Construct a 95 percent confidence interval for your a. Because of the chaos and uncontrolled environment
estimate. at the city park rally, allowances will be 25 percent of
c. If an hour of a nurse’s time costs the hospital $40, es- the normal time. Determine the normal time for the
timate the annual savings in cost if the installation of cycle and the standard time.
the computer cuts time spent on paperwork by 80 per- b. The mayor doesn’t like to leave things to chance.
cent. Assume 24 hours per day, 365 days per year of Suppose that he wants 99 percent confidence that
operation. each work element’s average time from the study is
 > PROBLEMS < H.17

Operation: Yard sign assembly Date: 9/27 Observer: Jerimiah (Jerry) Johnson
Observations
Work Element 1 2 3 4 5 6 7 8 9 10 t F RF

1. Get stake t
and sign r 8 39 70 1 07 1 42 1 81 207 254 282 31 2 1 .0 1 .05

2. Put glue t
on stake r 14 46 75 112 1 51 1 85 21 4 259 285 31 6 1 .0 0.8

3. Place sign, t
four staples r 25 60 90 1 26 1 68 1 96 236 270 298 338 1 .0 0.9

4. Check assem., t
set aside r 30 64 97 1 32 1 76 1 99 245 276 303 343 1 .0 1 .2
Normal time for cycle:
Allowances (% of total time): 25% Standard time: seconds per piece

FIGURE H.5

within ±5 percent of the true average. Did the public passports? What proportion of time do the clerks
works department make enough observations? If not, spend on each activity?
how many more should be made? b. If a machine to sell special stamps could be purchased
12. The information systems department of Universal Life outright for $3,500, would you recommend buying it?
Insurance Company wants to determine the proportion of Discuss.
time that the data entry operator is idle. The following 14. As manager of an encoding department in a bank, you are
information was gathered randomly using work sampling: concerned about the amount of time your encoder clerks
have to spend cleaning their machines because of mal-
Number Number Total functions. You obtained a proposal to modify the design
of Times of Times Number of of the machines to reduce the number of malfunctions.
Date Clerk Busy Clerk Idle Observations The modification will reduce the amount of time spent
cleaning the machines by 75 percent. You employ 20
8/22 11 2 13 encoder clerks at an average salary of $36,000 for working
2,000 hours per year. To help you decide whether the pro-
8/23 12 3 15
posal is worth considering, you had a pilot work sampling
8/24 11 3 14 study made, which provided the following results:
8/25 12 4 16
8/26 13 1 14
8/27 13 3 16 Activity Observations
8/28 6 6 12 Processing checks 52
Cleaning machine 15
If the department wants a 99 percent confidence level
Other duties 25
and a degree of precision of ±0.01, how many more obser-
vations are needed? Breaks 8
Total 100
13. The manager of the Valley Forge post office is interested
in the amount of time that window clerks spend on ancil-
lary services such as selling special issue stamp sets or
commemorative T-shirts and helping customers with a. Estimate the value of the annual labor savings from
passport applications. Three clerks, each earning $36,000 modifying the encoding machine design.
per year, staff the windows. When they are not needed at b. Construct a 95 percent confidence interval for
the window, they sort mail for the carriers. The results of a your estimate. Would you suggest a larger sample
preliminary work sampling study are shown in Table H.7, size? Why? (Hint: Base your confidence interval
where entries reflect number of occurrences. on the normal approximation to the binomial
a. For a degree of precision of ±0.05, is the sample size distribution where the standard error is
adequate for special stamp sets? For T-shirts? For σ p = pˆ(1 − pˆ )/ n .)
H.18 SUPPLEMENT H > MEASURING OUTPUT RATES <

TABLE H.7 Valley Forge Work Sampling Data

Day Selling Postage Priority Mail Special Stamp Sales T-Shirt Sales Passports Other Total

1 6 1 1 2 10
2 6 1 1 1 1 10
3 9 1 10
4 6 1 1 1 1 10
5 8 1 1 10
6 7 2 1 10
7 7 1 1 1 10
8 6 1 1 2 10
9 8 1 1 10
10 6 3 1 10

ADVANCED PROBLEM from the sample with the actual proportions and explain
Problem 15 requires prior reading of the Simulation supplement. how to obtain estimates corresponding to any desired confi-
dence level and within any specified maximum error.
15. You have been asked by your boss to make a presentation at a. Use a random number table to select 20 times during
the next management committee meeting on the methodol- the week you will “observe” the punch press. Use a
ogy of work sampling and its use in your machine shop. A random number first to select the day and then use
consultant has recommended this approach to manage- another random number to select the time of day.
ment for use in estimating the proportion of nonproductive Omit the time period 12 P.M. to 1 P.M. each day because
time for production equipment, such as a punch press. The that is lunchtime, during which the machine will be
committee members want an insider to brief them on this shut down. Put these 20 times in an observation
method. Sampling in real time is not possible during the schedule.
meeting, so you decide to develop a chart that is a “virtual b. Using your observation schedule, determine from
representation” of use of the punch press in the “real Figure H.6 what the machine is doing from Figure H.6
machine” shop. The chart is to cover all time intervals during at these times. Determine the sample proportion of
the five working days of the week. To develop this virtual the time spent in each category.
representation of the actual operation, you ask an assistant
c. What are the actual proportions of time spent in each
to observe it each day of the week and record running,
category from Figure H.6 ? How do they compare with
setup, idle, and breakdown times along a time line from 8:00
the sample estimates?
A.M. to 5:00 P.M. After making the observations, your assistant
summarized his findings in a chart, as shown in Figure H.6. d. Determine the sample size needed to ensure accuracy
You intend to use this chart to simulate work sampling, within 4 percent with 95 percent confidence.
thereby executing a work sampling plan in minutes that e. Based on your experiment, what can you tell manage-
would normally take a week. You can then compare the pro- ment about sample sizes and the accuracy of the
portions of time devoted to different activities estimated estimates?

FIGURE H.6
Time 8 9 10 11 12 1 2 3 4 5

Monday S R I R B S R S R
Tuesday R S R R I S R
Wednesday I R B R R B S R B
Thursday S R S R R S R I
Friday R S R I R S R
S = Setup R = Running I = Idle B = Breakdown
 > CASE < H.19

CASE The Facilities Maintenance Problem at Midwest University

Sean Allen is the manager in charge of facilities maintenance pleted because more than one type of craft was required. For
at Midwest University. Located on a 500-acre tract of land instance, repairing a set of wall-mounted bookshelves in a
outside St. Louis, Missouri, Midwest University is home to dormitory room required both a carpenter and a painter.
15,000 students. Allen is responsible for maintaining all the Personnel in each craft were scheduled independently.
physical facilities on campus, which comprise 60 buildings. As Allen thought about what to do, he kept coming back
They include dormitories, academic buildings, administra- to what he had learned in the continuous improvement work-
tion and office buildings, two athletic stadiums, and a basket- shops about “getting closer to the customer” and establishing
ball coliseum. To carry out this function Allen manages a large, cross-functional work teams that focus on processes, not out-
diverse workforce that has traditionally been segmented by puts. A new structure with enhanced job responsibilities
skilled craft into electricians, carpenters, plumbers, painters, might just be the answer. The big questions were “What kind
heating and air conditioning specialists, masons, dry wallers, of organizational structure would make sense?” and “How
and so on. Allen also is responsible for the custodial and could he minimize time spent traveling back and forth across
cleaning crews for each facility. campus and more effectively utilize his skilled craftspeople?”
A recurring nightmare for Allen has been the inability of Finally, there was the issue of implementation. After he
facilities maintenance to respond quickly to work-order had designed a new organizational structure and established
requests. A review of the data indicated that a response time new job responsibilities, how could he get the facilities main-
of 5 to 10 days was not unusual. This was unacceptable. tenance personnel to support the changes? One phrase kept
Allen applied what he had learned in a series of continu- going through Allen’s mind: “You get what you measure.” In
ous improvement workshops that focused on problem iden- addition, any reorganization would mean a realignment of
tification, data collection and analysis, and problem resolu- employee performance evaluation and recognition procedures.
tion. He soon discovered that 85 percent of the work-order
requests took less than an hour to handle. Furthermore, QUESTIONS
almost 40 percent of the requests were for routine mainte-
nance items such as clogged drains, burned-out light bulbs, 1. How would you restructure the facilities maintenance
and loose towel racks. His analysis led him to the conclusion organization at Midwest University?
that an ineffective organizational structure was a primary 2. What can Sean Allen do to alleviate the problem of exces-
cause of the long response times. sive travel time for work crews?
Facility maintenance personnel were grouped by craft 3. As Allen redesigns job responsibilities, how should he
and centrally located at the physical plant offices. As work evaluate his personnel’s performance? What should he
orders were received, Allen would try to prioritize the requests measure? How should he reward employees?
and allocate craft personnel to fix the problem. Scheduling
work to be done was complicated. Both the importance of the
job and the location had to be considered. Maintenance per-
sonnel often spent a large portion of their time traveling back
and forth across campus, going from one job to the next. Source: This case was prepared by Dr. Brooke Saladin, Wake Forest
Allen also discovered that jobs frequently could not be com- University, as a basis for classroom discussion.
H.20 SUPPLEMENT H > MEASURING OUTPUT RATES <

> SELECTED REFERENCES <

Ellis, Christian M., and Lea A. P. Tonkin. “Mature Teams Rewards “Motorola: Training for the Millennium.” Business Week (March 28,
and the High-Performance Workplace: Change and Oppor- 1994), pp. 158–162.
tunity.” Target, vol. 11, no. 6 (1995).
Niebel, Richard W. Motion and Time Study, 8th ed. Homewood,
Gephart, Martha A. “The Road to High Performance.” Training Ill.: Irwin Professional Publication, 1988.
and Development, vol. 49 (June 1995), pp. 29–44.
Schonberger, Richard J. Building a Chain of Customers: Linking
Hammer, Michael. “Reengineering Work: Don’t Automate, Business Functions to Create the World Class Company. New
Obliterate.” Harvard Business Review (July–August 1990), York: Free Press, 1990.
pp. 104–112.
Sherman, Stratford. “Levi’s: As Ye Sew, So Shall Ye Reap.” Fortune
Herzberg, F. “One More Time: How Do You Motivate Employ- (May 12, 1997), pp. 104–116.
ees?” Harvard Business Review (September–October 1987),
“What the Experts Forgot to Mention.” Inc. (September 1993),
pp. 109–120.
pp. 66–77.
Katzenbach, Jon R., and Douglas K. Smith. “The Discipline
“When the Going Gets Tough, Boeing Gets Touchy-Feely.” Busi-
of Teams.” Harvard Business Review (March–April 1993),
ness Week (January 17, 1994), pp. 65–67.
pp. 111–120.
Knights, David, Hugh Willmott, and David Collision, eds. Job
Redesign. Hants, England: Gower, 1985.