What is work sampling and specify procedure for conducting a work sampling study?

With the growth of the apparel sector over a period of time, Industrial Engineering (IE) practices are gaining popularity and benefiting the organizations through providing scientific yet simple and practical solutions like time study, method study and PMTS. Work sampling is one such practical techniques which has some unique characteristics that make it immensely useful to the management, highlighting the true picture of the shop floor functioning, by measuring time consumed on non productive activities of an operator with minimal money, manpower and time involvement.

It has been observed that in India most of the organizations don’t follow work sampling as a key tool for optimizing apparel manufacturing operations, the reasons for this may be the difficulties faced in practicing it on the shop floor, involvement of statistics with which people are not comfortable and the nature of jobs being done or the kind of production orders being executed. Manoj Tiwari, Assistant Professor at NIFT, Gandhinagar discusses the basics of work sampling, procedure of conducting the study, advantage and limitations, and the factors affecting the work sampling results through exercises done on the shop floor at various production houses.

Work sampling is one of the work measurement techniques similar to time study with stop watch and data synthetics using PMTS (Pre-determined Motion and Time Systems), capable of highlighting the percentage of occurrences of time consumed in an activity, which prima-facie is non-productive or ineffective in nature and can help the management take necessary actions to curb such inefficiencies. On the other hand, work sampling also highlights the amount of time spent in productive (for example sewing), associated or support activities; thus by using work sampling, factors affecting the operator productivity can also be identified.

Conducting time study is a cumbersome and costly process which consumes a significant amount of observer’s time, while work sampling is a simple yet pragmatic approach based on the theory of probability, following the statistical sampling and random observations. Work sampling is just the “degree to which the event(s) is likely to occur”.

Work sampling observations need a significant amount of planning and preparatory work before conducting the same on the shop floor. A clear-cut identification of all the objectives is one of the primary requirements of this study which highlights why exactly the study needs to be done and what is expected from such study. Pre-planning also requires a thorough discussion of all the aspects related to the project in detail prior to conducting work sampling, with the people and authorities associated with it. The objective is to get people in confidence by explaining the pros and cons of the study as well as convincing them with clear explanation about the possible impact of the study on the organization.

Sample Size Determination

Critical to the entire work sampling method is the determining of how many times the observations need to be taken based on a confidence level and the percentage error or degree of accuracy using a direct Nomogram (refer Figure 1). Confidence level signifies the number of times the test results are within a specified range, whereas the accuracy level means degree of closeness of measurements of a quantity to that quantity’s actual (true) value. 95 per cent confidence level with 5 per cent accuracy level means in 95 per cent of the cases when test is conducted, the result will be between +/- 5 per cent of the actual value. For example, if there is a result which is 55 at 95% confidence level and +/- 5 per cent error, the actual value will swing between +/- 5 per cent of 55.

Classifying Observations

Deciding about the times, when the observer is going to take the observations on the floor is purely random in nature so that any kind of bias can be avoided while making the observation. In addition, for an error free study, various activities of an operator should be broken down to smaller visually measurable movements of body parts (elements), each of which needs to be clearly described and classified.

In order to avoid subjectivity, it is critical that the level of understanding of all the people involved in the study should be uniform while deciding the elements, as the entire results and inferences depend solely on the break-up of the activities.

To begin with the generic nature of activities needs to be decided, like productive activities, personal activities and avoidable and unavoidable delays. Productive activities are the activities which contribute directly in the productivity of the work like sewing. These may be divided into main activities and associated activities, while stitching is main activity when needle is working (that’s the real productive activity), activities like untie the bundle, taking piece to the needle point, folding, aligning, guiding, etc. are associated activities. Personal activities are individual physiological personal needs like drinking water and going to washroom, relaxing, etc. Avoidable delays are the activities, which are primarily non value added and don’t contribute in productivity enhancement. These are responsible for inefficiencies in the production process. Sitting idle without any reason, waiting for material or work, speaking to co-operators, roaming around, machine maintenances which are unplanned, etc. can be classified as avoidable delays. Unavoidable delays are the activities which may not be completely avoided, like bundle handling, changing the needle and bobbin thread, needle replacement, changing work-aids, organizing the work place, etc.

The operator on a sewing floor may be involved in various activities related to stitching, material handling, recording the production, communicating with supervisor, checking the quality of the piece stitched by him, chatting with other operators, repairing the machine, replacing the thread, replacing the needle and doing many more such activities in the various departments on the shop floor, right from fabric inspection, spreading to sewing and finishing. Apart from it, he/she may go for short breaks for drinking water or to go to the restroom or washroom or may be just sitting idle or waiting for the material. Classification of all such actual activities should be done very carefully as observation on these points will show the real picture and will guide for decision making and improvements.

Conducting Work Sampling

Once the number of observations required is statistically derived, and timings of observations, as well as the standard format (on which observations need to be taken) are ready, one is ready to proceed for the work sampling study. The observer needs to go to the floor on the particular pre-defined times, observe and mark the individual operator in one go without spending any time, and fill the information in a standardized format with clearly defined classifications of activities. The format (refer Figure 3, Work Sampling Format) may change from organization to organization depending on the situation.

This is the reason that work sampling is also called snap-reading or flash-reading. The observer just needs to take a glance and put tally marks in the work sampling format under the observed category of the activity against the respective operator code in the format. The same system needs to be followed whenever the observer revisits the floor for the study.

Analysing Results

Once the study is over, each operator’s data needs to be compiled to know the percentage share of individual activities from the total number of observations.

The results may give the idea about how much time (in percentage) is consumed in which all activities and subsequently corrective actions can be taken. The work sampling exercise is recommended for repetition every 3-4 months for checking the effectiveness of measures taken and analyse the improvements.

Benefits of Work Sampling over Time Study

• It helps in the determination of standard times that should be allowed for various activities.

• It helps in determination of percentage allowances required for personal needs, fatigue and unavoidable delays, etc.

• Work sampling is economical and capable of producing the same results as time study at 1/3rd to 1/6th of the time and cost, as it is based on operator’s view point or probably factual results from his experience.

• Longer time periods can be studied based on the convenience (does not require analyst to make continuous observations of observer) where time study needs uninterrupted time to observe. As the person in case of work sampling just walks through the line, observes each activity at a glance and goes back to his seat, whereas in time study the person needs to stand continuously in front of an operator and measure using stop watch.

• Operator is not subject to continuous observation.

• Many operations and workers can be studied by a single analyst.

Limitations of Work Sampling

• The work sampling results may not be applicable for the small order quantities, as there may be chances that the production order is over during the study. In this case the results may not be authentic and reliable.

• Time study is capable of providing a finer breakdown of activities and delays while work sampling cannot provide as much detailed information of individual operator.

• The observation timings should be purely random. In case it’s not followed and observer enters into the floor after regular intervals; operators may change the work style. Hence the result may not be realistic and useful.

• Involvement of statistics makes the work sampling difficult to understand and apply, and people hesitate in using the work sampling.

• The work sampling results may vary when the method is changed, for example level of automation while stitching, method of material handling (usage of over-head material handling systems), type of product, etc.

Conclusion

Work sampling technique is capable of providing new insights if followed religiously and scientifically. Prima facie it looks difficult to understand and conduct but in real sense the technique is fast, easy and economical. In fact, it is very useful to the people who want to get a glimpse of the activities without getting involved in the technical intricacies. The technique is very much capable of highlighting the needle time, which is the only productive activity in real sense while stitching.

There are global benchmarks available for percentage of time consumed in various activities in the garment manufacturing shop floor which are guides for standard times allocation for various delays yet these standards cannot be adopted straight forward.

There are many factors which need to be considered depending on the working conditions of that particular country or region which make it more relevant and practically applicable. Some of these factors may be cited as level of technology used, product type and volume of orders. For the advanced new generation machines, where automation level is high and human intervention is minimal, the needle time should be high as the machines are programmed for each activity and performs tasks automatically based on the commands given.

At the same time there is very minimal time consumed in material handling, positioning, data recording, thread replacement, etc. Product type is another area which affects the needle time in a factory. For example if one factory produces bed sheets where seam lengths are comparatively longer than that of garments like T-shirts or shirts. Lesser stops in the sewing of a seam are observed which may lead to increased the percentage of needle time.

At the same time if the factory produces garments where high level of skill is required (such as high fashion tops with some hand work, etc.) with precision, then there are fair chances that percentage of needle time may go down and personal, fatigue and delay levels may go high. High volume orders which have more number of units to be produced may have increased the percentage of needle time, as the system might have achieved the rhythm and overcome the start-up, losses.

While in case of small orders there are chances that by the time system achieves its peak after start-up, the order quantity is finished from stitching. In this case work sampling standards are of no use practically.

In context of Indian garment industry, so far no such study has been reported where needle times for different products (at different technology levels covering different apparel production setups) have been observed and studied. Such data may give us insights regarding the percentage of time utilized in sewing and other activities as well as the factors affecting the needle time with its logical justification.

What are the steps involved in conducting a work sampling study?

What are 5 sampling procedures?

What is sample and sampling procedure in research?

What are the 4 sampling procedure?