Work Sampling
Work Sampling (also sometimes called ratio delay study) is a technique of getting facts about utilization of machines or human beings through a large number of instantaneous observations taken at random time intervals. The ratio of observations of a given activity to the total observations approximates the percentage of time that the process is in that state of activity. For example, if 500 instantaneous observations taken at random intervals over a few weeks show that a lathe operator was doing productive work in 365 observations and in the remaining 135 observations he was found 'idle' for miscellaneous reasons, then it can be reliably taken that the operator remains idle (135/500) x 100 = 27 % 0f the time. Obviously, the accuracy of the result depends on the number of observations. However, in most applications there is usually a limit beyond which greater accuracy of data is not economically worthwhile.
Use of Work Sampling for Standard Time Determination
Work sampling can be very useful for establishing time standards on both direct and indirect labor jobs. The procedure for conducting work sampling study for determining standard time of a job can be described step-wise.
Step 1. Define the problem.
• Describe the job for which the standard time is to be determined.
• Unambiguously state and discriminate between the two classes of activities of operator on the job: what are the activities of job that would entitle him to be in 'working" state.
This would imply that when operator will be found engaged in any activity other than those would entitle him to be in "Not Working" state.
Step 2. Design the sampling plan.
• Estimate satisfactory number of observations to be made.
• Decide on the period of study, e.g. two days, one week, etc.
• Prepare detailed plan for taking the observations.
This will include observation schedule, exact method of observing, design of observation sheet, route to be followed, particular person to be observed at the observation time, etc.
Step 3. Contact the persons concerned and take them in confidence regarding conduct of the study.
Step 4. Make the observations at the pre-decided random times about the working / not working state of the operator. When operator is in working state, determine his performance rating. Record both on the observation sheet.
Step 5. Obtain and record other information. This includes operator's starting time and quitting time of the day and total number of parts of acceptable quality produced during the day.
Step 6. Calculate the standard time per piece.
We will now briefly discuss some important issues involved in the procedure.
Number of Observations
As we know, results of study based on larger number of observations are more accurate, but taking more and more observations consumes time and thus is costly. A cost-benefit trade-off has thus to be struck. In practice, the following methods are used for estimation of the number of observations to be made.
(i) Based on judgment. The study person can decide the necessary number of observations based on his judgment. The correctness of the number may be in doubt but estimate is often quick and in many cases adequate.
(ii) Using cumulative plot of results. As the study progresses the results of the proportion of time devoted to the given state or activity, i.e. Pi from the cumulative number of observations are plotted at the end of each shift or day. Since the accuracy of the result improves with increasing number of observations, the study can be continued until the cumulative Pi appears to stabilize and collection of further data seems to have negligible effect on the value of Pi.
(iii) Use of statistics. In this method, by considering the importance of the decision to be based on the results of study, a maximum tolerable sampling error in terms of confidence level and desired accuracy in the results is specified. A pilot study is then made in which a few observations are taken to obtain a preliminary estimate of Pi. The numbers of observations N necessary are then calculated using the following expression.
The number of observations estimated from the above relation using a value of Pi obtained from a preliminary study would be only a first estimate. In actual practice, as the work sampling study proceeds, say at the end of each day, a new calculation should be made by using increasingly reliable value of Pi obtained from the cumulative number of observations made.
Determination of Observation Schedule
The number of instantaneous observations to be made each day mainly depends upon the nature of operation. For example, for non-repetitive operations or for operations in which some elements occur in-frequently, it is advisable to take observations more frequently so that the chance of obtaining all the facts improves. It also depends on the availability of time with the person making the study. In general, about 50 observations per day is a good figure. The actual random schedule of the observations is prepared by using random number table or any other technique.
Design of Observation Sheet
A sample observation sheet for recording the data with respect to whether at the pre-decided time, the specified worker on job is in 'working' state or 'non-working' state is shown in Figure. It contains the relevant information about the job, the operators on job, etc. At the end of each day, calculation can be done to estimate the percent of time workers on the job (on an average) spend on activities, which are considered as part of the job.
Conducting Work Sampling Study
At the predefined times of study, the study person appears at the work site and observes the specific worker (already randomly decided) to find out what is he doing. If he is doing activity which is part of the job, he is ticked under the column 'Working' and his performance rating is estimated and recorded. If he is found engaged in an activity which is not a part of job, he is ticked under the column 'Not Working'. At the end of day, the number of ticks in 'Working' column is totaled and average performance rating is determined.
The observed time (OT) for a given job is estimated as
The normal time (NT) is found by multiplying the observed time by the average performing index (rating factor).
where, is average rating factor to be determined as , Figure
The standard time is determined by adding allowances to the normal time.
Advantages and Disadvantages of Work Sampling in Comparison with Time Study.
Advantages
Economical
• Many operators or activities which are difficult or uneconomical to measure by time study can readily be measured by work sampling.
• Two or more studies can be simultaneously made of several operators or machines by a single study person. Ordinarily a work study engineer can study only one operator at a time when continuous time study is made.
• It usually requires fewer man-hours to make a work sampling study than to make a continuous time study. The cost may also be about a third of the cost of a continuous time study.
• No stopwatch or other time measuring device is needed for work sampling studies.
• It usually requires less time to calculate the results of work sampling study. Mark sensing cards may be used which can be fed directly to the computing machines to obtain the results just instantaneously.
Flexible
6. A work sampling study may be interrupted at any time without affecting the results.
7. Operators are not closely watched for long period of time. This decreases the chance of getting erroneous results for when a worker is observed continuously for a long period, it is probable that he will not follow his usual routine exactly during that period.
Less Erroneous
8. Observations may be taken over a period of days or weeks. This decreases the chance of day-to-day or week-to-week variations that may affect the results.
Operators like It
9. Work sampling studies are preferred to continuous time study by the operators being studied. Some people do not like to be observed continuously for long periods of time.
Observers like It
10. Work sampling studies are less fatiguing and less tedious to make on the part of time study engineer.
Disadvantages
• Work sampling is not economical for the study of a single operator or operation or machine. Also, work-sampling study may be uneconomical for studying operators or machines located over wide areas.
• Work sampling study does not provide elemental time data.
• The operator may change his work pattern when he sees the study person. For instance, he may try to look productive and make the results of study erroneous.
• No record is usually made of the method being used by the operator. Therefore, a new study has to be made when a method change occurs in any element of operation.
• Compared to stop watch time study, the statistical approach of work sampling study is difficult to understand by workers.
Computerized Work Sampling
Use of a computer can save as much as 30 to 40 percent of the total work sampling study cost. This is because too much clerical effort is involved in summarizing work sampling data, e.g. in determining the number of observations required, determining the daily observations required, determining the number of trips to the area being studied per day, determining the time of each observation, calculating the accuracy of results, plotting data on control charts and like that. Computers can be used for mechanization of the repetitive calculations, display of control charts and calculation of daily as well as cumulative results.
Services: - Work Sampling Homework | Work Sampling Homework Help | Work Sampling Homework Help Services | Live Work Sampling Homework Help | Work Sampling Homework Tutors | Online Work Sampling Homework Help | Work Sampling Tutors | Online Work Sampling Tutors | Work Sampling Homework Services | Work Sampling