work sampling study
Six Sigma – iSixSigma › Forums › Old Forums › General › work sampling study
 This topic has 10 replies, 7 voices, and was last updated 15 years, 7 months ago by chainani.

AuthorPosts

September 20, 2005 at 6:47 pm #40755
HI ALL,
I am required to do a work sampling study on an operator to determine, how he is spending his shift time on a set of n activities { A, B,C,D…n}.I want to know how many observations should i take for my study. i also want to know the statistics behind this study. If you can help me with this or forward me to resources where i can look into, i will be greatly thankful to you.
Thanking you,Sincerely,Vidya0September 20, 2005 at 7:43 pm #127191
thevillageidiotMember@thevillageidiot Include @thevillageidiot in your post and this person will
be notified via email.For what purpose?
0September 20, 2005 at 10:10 pm #127198I clearly told what the purpose was …
thanks,
vidya0September 20, 2005 at 11:00 pm #127199
thevillageidiotMember@thevillageidiot Include @thevillageidiot in your post and this person will
be notified via email.Glad I could help.
0September 20, 2005 at 11:23 pm #127201
thevillageidiotMember@thevillageidiot Include @thevillageidiot in your post and this person will
be notified via email.Maybe you are new to the forum so I will extend the proverbial olive branch…the lack of responses to your question indicates it is poorly worded….yours is vague and incomplete. For example, the type of analysis (ie purpose) impacts the decision as to sample size calculation, among other things. Also, you could google an answer to your question almost as fast as you could post in the forum. Try that first, come back with questions. You will find people who will help you after you have done a little work. Good luck.
0September 21, 2005 at 12:56 am #127202
Ken FeldmanParticipant@Darth Include @Darth in your post and this person will
be notified via email.I apologize on behalf of my IdiotFriend. He was being sooo rude. It is so unlike this Forum to treat new posters this way. I take it that your purpose was to classify what proportion of the time your operator is engaged in each of the identified activities. I have done this to see if there is a capacity or utilization issue. Your first decision is over what time period will you take your study. Do you suspect that the proportion might vary day to day or week to week? If so, you will want to sample over that time period. The number of random…and I emphasize random, times you will observe can be determined as you would any other sampling for proportions. That formula is easily available with an internet search. Since you have no idea as to the proportion, I would suggest that you select .5 as your “proportion” defective and then decide on what confidence you want (maybe 95%) and what precision (possibly 5%). You would then divide up your time period into minutes and randomly select the appropriate number of observations. You would then go out and observe the operator at the exact time and record which of the activities they are engaged in. Once you have accumulated all your required observations you can compute your proportions and voila you are done.
0September 21, 2005 at 9:36 am #127213Hi Vidya,
Take a look at http://hal.lamar.edu/~ie_dept/underdown/Work_Design/INEN_3380.htm
Let me know if it is helpfull
Bernard
0September 21, 2005 at 10:13 am #127214Dear Vidya,
Assuming that this is a study for process improvement and not for spying on colleagues, my best answer would be to tell you how I would appproach this myself. I would certaintly not start with dealing with this question only from a statistical point of view. I would start with going to the workplace, just observe the area and the operators, and chat with one or two of them. Then it is time to ask yourself a few questions:
1) Can you assume that the mix of percentages between the different tasks is stabile? For instance, is the operator working on a stabile process or in an ever changing environment? In case of the latter, sampling over longer time does not necessarily make the result better. And are the operators determining what the process does, or does the process determine what the operators do. If the operators determine the process, there can be more differences between operators than in the other case.
2) How many different tasks do you define? Almost every task can be divided into numerous subtasks, but the more tasks you define, the longer it will take before you are finished. It is a good idea to discuss this carefully with whoever ‘required’ you to do this work.
3) What resolution do you need for the proportion of time? Is your client interested in portions of 1%, or 10%, or …..? Without this question you will not know when enough is enough. Again, discuss with the client.
Next step is to start collecting data during a short period, calculate percentages, collect new data in a second short period, recalculate the percentages, and so on. I would start with at least ten samples. Then, make a plot of the sample number on the xaxis and the cumulative percentages on the yaxis.
Just look at the graph, and check if the percentages still show important changes with each new sample “Important” being defined as: judged against the question of how much resolution you need.
If it doesn’t, it is now a good time to do a formal calculations of the percentages, with their confidence intervals and such. If these are not too large, you can start reporting.
If changes have a random appearance that are large compared to your desired resolution, there is still too much noise in the data and more samples are needed.
If changes show a trend, the percentages of time spent are not stabile, which is also usefull information that you can discuss with your client.
I hope this is helpfull for you!0September 27, 2005 at 9:41 pm #127528HI all,
I thank all for replying to my question. i couldnt respond soon thanks to Hurricane Rita. As Darth posted the purpose is to”your purpose was to classify what proportion of the time your operator is engaged in each of the identified activities”. Bernard, i couldnt access the link u posted. Arend, ur approach seems new for me. Anyways thanks for all the help.
Vidya0May 4, 2006 at 5:34 pm #137255
Fabio PaivaParticipant@FabioPaiva Include @FabioPaiva in your post and this person will
be notified via email.Hi, friend
Could you please give me any hint about how can I use Minitab to evaluate such sample size?
I am concerned because we have to know the sample power.
Ex. If I take the formula for work sampling I could have, for instance, 896 as the sample size alfa= 5%, absolute error = 3% and the proportion =30%
How can I evaluate such sample size by Minitab? How can I think about the sample power? I believe I should perform a power and sample size for 2 proportion
How can I make the 2 methodologies match?
Thanks0October 23, 2006 at 5:25 am #145444i want work study materiyal plz send to my email
0 
AuthorPosts
The forum ‘General’ is closed to new topics and replies.