1.0 INTRODUCTION In Note 5, you
learned that job design determines the content of a job. This note is devoted to work measurement, which is the
process of creating labour standards
based on the judgment of skilled observers. Actually, job times are vital inputs for manpower planning,
estimating
labour costs, scheduling,
budgeting and designing incentive systems.
2.0 OBJECTIVES
By the end of this note, you should be able to:
(i) Define a standard time
(ii) Discuss and compare time study methods
(iii) Describe work sampling and perform calculations
(iv) Compare stopwatch time study and work sampling.
3.0 MAIN CONTENT
3.1 Relevance of Work Measurement
Work measurement refers to the process of estimating the amount of
worker time required to generate one
note of output. Its ultimate goal is usually to develop labour standards that will be used
for planning and controlling operations,
thereby achieving labour productivity.
Job times are important inputs for manpower planning, estimating labour
costs, scheduling, budgeting, and
designing incentive systems. In addition, from the workers' standpoint time standards provide an
indication of expected output. Time
standard reflects the amount of time it should take an average worker to do a given job, working under typical
conditions. The standards include
expected activity time plus allowances for probable delays.
Whenever a time standard is
developed for a job, it is essential to provide a complete description of the parameters of the job because the
actual time to do the job is sensitive to
given methods, tools and equipment, raw materials inputs and
workplace arrangement. For instance,
changes in product design or changes in job
performance brought about by a methods study should necessitate a new
time study to update the standard time.
3.2 Standard Time as
Management Tool.
Managers use Standard Time in the following ways:
(i) Establishing Prices and Costs: Managers can use labour and
machine time standards to develop costs
for current and new products, create
budgets, determine prices and arrive at make or - buy decisions.
(ii) Motivating Workers: Standards can be used to define a day's
work or to motivate workers to improve
their performance. For example, under an
inventive compensation plan, workers can earn a bonus for output
that exceeds the standard.
(iii) Comparing alternative process designs: Time standards can
also be used to compare different
routings for an item and to motivate new work
methods and new equipment.
(iv) Scheduling: Managers need time standards to assign task to
workers and machines in ways that effectively
utilize resources.
(v) Capacity Planning: Managers can use time standards to
determine current and projected capacity
requirements for given demand
requirements. Work-force staffing decisions also, may require time estimates.
(vi) Performance Appraisal: A worker's output can be compared to
the standard output over a period of
time in order to evaluate worker
performance and productivity. A manager's performance can similarly be measured by comparing actual costs to
standard costs of a process.
3.3. Methods of Work Measurement
Organisations develop time standard in a number of different ways.
The most common methods of work
measurements are:
(i) The time study method
(ii) The elemental standard data approach
(iii) The predetermined data approach and
(iv) The work sampling method.
The particular method chosen usually depends on the purpose of the
data. For example, if a high degree of
precision is needed in comparing actual work
method results to standard, a stopwatch study or pre-determined times might be required. On the other hand,
an analyst who wants to estimate the
percentage of time that an employee is idle while waiting for materials requires a work sampling method. We shall be
examining the time study method,
elemental standard data approach, and the work sampling method in the sections that follow.
3.3.1 Time Study Method
In this method, analysts use stopwatches to time the operation
being performed by workers. These
observed times are then converted into labour standards that are expressed in minutes per note of output
for the operation. A time study usually
consists of four steps:
Step 1: Selecting Work Elements
Each work element should have definite starting and stopping
points so as to facilitate taking
stopwatch readings. It has been suggested that work elements that take less than three seconds to complete
should be avoided since they are often
difficult to time. The work element
selected should correspond to a standard work method that has been running smoothly for a period of
time in a standard work environment.
Efforts should also be made to identify and separate incidental operations that are not normally involved in
the task from the repetitive work.
Step 2: Timing the Elements
Here, the analyst times a worker trained in the work, in order to
get an initial set of observations. The
analyst may use either the continuous method,
recording the stopwatch reading for each work element upon its
completion, or the snap-back method,
re-setting the stopwatch to zero upon completion of each worker element. In the case of the latter, the analyst uses
two watches, one for recording the
previous work element, and the other for timing the present work
elements.
In case that data include a
single, isolated time that differs greatly from other times recorded for the same element, it is
advisable for the analyst to investigate
the cause of the variation. Any irregular occurrence such as a dropped tool or a machine failure, should not
be included in calculating the average
time for the work element. The average observed time based only on representative times is called the select
time (t).
Step 3: Determining Sample Size
It is usual for analysts using the time study method to look for
an average time estimate that is very
close to the true long range average most of the time. The following formular, based on the normal
distribution is used to determine the
required sample size, n: d n = 2 ( )( ) t
Where
n = required sample size
P = precision of the estimate as a proportion of the true
value
t = select time for a work element
d = standard deviation of representative observed times for a work
element.
Z = number of normal standard deviations needed for the
desired confidence.
Where = Accuracy or maximum acceptable error
Typical values of Z for these formulars are:
For example, a Z value of
1.96 represents ±1.96 standard deviations from the mean, leaving a total of 5 percent in the
tails of the standardized normal curve.
The precision of the estimate, P is expressed as a proportion of the
true (but unknown) average time for the
work element. Let us make use of an
example given by Krajewski and Ritzman (1999) as an illustration of this step.
The example A coffee cup
packaging operation has four work elements. A preliminary study provided the following results: Select
Work Element Standard
deviation, Select Time, t Sample d
(minutes)(minutes)size
1. Get two cartons 0.0305
0.50 5
2. Put liner in carton 0.0171
0.11 10
3. Place cups in carton 0.0226
0.71 10
4. Seal carton and set 0.0241 1.10 10
Aside
Work element 1 was observed only 5 times because it occurs once
every two work cycles. The study covered
the packaging of 10 cartons. Determine the
appropriate sample size if the estimate for the select time for any work
element is to be within 4 percent of the
true mean 95 percent of the time.
Solution For this problem P= 0.04 and Z = 1.96 The sample size for each work element must be
calculated, and the largest must be used
for the final study so that all estimates will meet or exceed the desired precision.
All fractional calculations were rounded to the next largest
integer. To be sure that all select
times are within 4 percent of the true mean 95 percent of the time, we must have a total of 58 observations
because of work element 2. Consequently,
we have to observe the packaging of 48 (i.e. 58-10) more cartons.
Step 4: Setting the Standard
This is the final step. Here, the analyst first determines the
normal time for each work element by
judging the pace of the observed worker. Next, he assesses not only whether the pace is above or below
average, but also a performance rating
factor (RF) that describes how much above or below average the worker's performance on each element is. Note
that setting the performance rating
requires the greatest amount of judgment. Usually, only a few workers are observed during a study. If the workers
are fast, basing the standard on their
average time wouldn’t be fair, especially if a wage incentive plan is
involved. At the same time, If the
workers are slow, basing the standard on their normal time would be unfair to the company. In
addition, workers may slow pace when
they are being observed in a time study. Ironically, it is important to inform the observed worker about the study,
so as to avoid suspicion or
misunderstandings. Workers sometimes
feel uneasy about being studied and fear the changes that might result. It is therefore necessary for
the analyst to discuss these things with
the workers prior to studying the operation to allay such fears, and to enlist the cooperation of the worker. Due to
these apparent distractions, the analyst
has to make an adjustment in the average observed time to estimate the time required for a trained operator to do
the task at a normal pace. The analyst
must also factor in the frequency of occurrence, F, of a particular work element in a work cycle. The normal time
(NT) for any work element is calculated
by multiplying the select time (t), the frequency (F) of the work element per cycle, and the rating factor,
(RF) i.e.: NT = t (F) (RF)
NOTE: Use F= 1, if the work element is performed every cycle
F= 0.05, if it is performed every other cycle. etc. To find the normal time for the cycle (NTC)
the normal time for each element is
summed up. i.e. NTC = , Where = sum
of
3.3.1.1 An Assessment of Time Study Time study methods have been observed to have some limitations.
Therefore, it is suggested that they
should not be used to set standards for jobs in which the nature of the task is different each
time. Examples of the situation above
include a student solving a problem, a
professor preparing a lecture, or an automobile mechanic diagnosing the
cause of a non-routine problem.
Furthermore, an inexperienced person should not be allowed to conduct time studies because
errors in recording information or in
selecting the work elements to include can result in unrealistic standards.
Another limitation of the time study is that some workers may
object to it because of the
"subjectivity" involved. However,
in spite of the above shortcomings, time studies conducted by an experienced analyst usually provide a
satisfactory, although imperfect, tool for
setting equitable time standards.
3.3.2 Elemental Standard Data Approach S
standard elemental times are derived from a firm's own historical
time study data. For instance, a time
study department over the years might have
accumulated a file of elemental times that are com mon to many jobs.
After a certain point, many elemental
times can be simply retrieved from the file, thus eliminating the need for analysts to go
through a complete time study to obtain
them. The procedure for using
standard elemental times consists of the following steps:
(i) Analyse the job to identify the standard elements
(ii) Check the file for elements that have historical times and
record them, use time study to obtain
others, if necessary
(iii) Modify the file times if necessary. Let us look at some
cases where the file times may not
pertain exactly to a specific task. For instance, standard elemental times might be on file on
"move the tool 3 centimeters"
and "move the tool 9 centimeters", whereas the task in question involves a move of 6 centimeters.
What can possibly be done is to
interpolate between values on file to obtain the desired time estimate.
(iv) Sum the elemental times to obtain the normal time, and factor
in allowances to obtain the standard
time.
3.3.2.1 An Assessment of
the Elemental Standard Data
Approach
An obvious advantage of the elemental standard data approach is
the potential savings in cost and effort
created by not having to conduct a complete time study for each job. Secondly, there is less disruption of work,
since the analyst does not have to time
the worker. Thirdly, performance ratings do not have to be done, since they have been generally averaged in the file
times. However, the elemental standard
data approach suffers from a major limitation
in that times may not exist for enough standard elements to make it
worthwhile. In addition, the file times
may be biased or inaccurate.
3.3.3 Work Sampling Method Work sampling is a
work measurement technique that randomly samples the work of one or more employees at periodic
intervals to determine the proportion of
the total operation that is accounted for in one particular activity. These types of studies are frequently used to
estimate the percentage of employee's
time spent in such activities as:
Unavoidable delays, which are commonly called ratio-delay studies; Repairing finished products from an
operation; or Supplying material to an
operation The results of these studies
are commonly used to set allowances used in
computing labour standards, in estimating costs of certain activities,
and in investigating work methods.
Unlike time study, work sampling does not require timing an activity nor does it even involve
continuous observation of the
activity. Instead, an observer is required to make brief observations of
a worker or machine at random intervals
over a period of time and simply note the nature of the activity. For example, a machine may
be busy or idle; a secretary may be
typing, filing, talking on the phone, etc. The resulting data are counts
of the number of times each category of
activity or non-activity was observed.
Conducting a work sampling study involves the following steps:
(i) Define the activities
(ii) Design the observation form
(iii) Determine the length of the study
(iv) Determine the initial sample size
(v) Select random observation times using a random number
table (vi) Determine the observer table
schedule (vii) Observe the activities
and record the data (viii) Decide
whether additional sampling is required.
It is important to note here, that work sampling estimates include some
degree of error. For instance, the same
number of observations taken at different times
during the week will probably produce slightly different estimates, and
all estimates will usually differ from
the actual (but unknown) values. It is
therefore important to treat work sampling estimates as approximations
of the actual proportion of time devoted
to a given activity.
3.3.3.1 Sample Size The goal of work
sampling is to obtain an estimate that provides a specified confidence not differing from the true value
by more than a specified error. That is,
we want to take a sample, calculate the sample proportion, and be able to say that the following interval contains
the true proportion with a specified
degree of precision:
Where, ˆ - ˆ e e <+ ˆ < ˆ = Sample proportion (number of occurrence
divided by the sample size) e =
maximum error in the estimate. The
surface size affects the degree of precision that can be expected from
work sampling for any desired level of
statistical confidence. Since work sampling
involves estimating proportions, its sampling distribution is the
binomial distribution. However, it has
been found that since large sample sizes are
required for this approach, the normal approximation to the
binomial distribution can be used to
determine the appropriate sample size. Figure 17 shows the confidence interval for a work
sampling study. The maximum error can be
computed as: e = ˆ (1 - ˆ
) n
Where n = sample size Z =
number of standard deviations needed to achieve the desired confidence.
4.0 CONCLUSION Through this note,
you have learned that it is important for management to make design of work systems a key element of
its operations strategy. Work
measurement is the process of estimating the amount of worker time
required to generate one note of output.
Its ultimate goal is usually to develop and
controlling operations, thereby achieving high labour productivity.
5.0 SUMMARY As already mentioned
above, work measurement is concerned with specifying the length of time needed to complete a job.
Such information is vital for personnel
planning, cost estimating, budgeting, scheduling, and worker compensation. Commonly used approaches include
stopwatch time study and predetermined
times.
0 comments:
Post a Comment