Time Studies in manufacturing 7

[papimonte]

I am curently trying to develop a good time study to apply to low volume to medium volume (less than 200 parts per order) production process. e.g: job shop. Any suggestions?

 

[PSE]

Break the job down into various steps that may be of interest such as setup, value added steps, business value steps (such as inspection/qc) and non value added steps (wait times, material handling etc). If you can, get information from time applied against previous build orders and follow them up with an actual review/study of the work performed.

If you are looking for software etc. There are numerous programs out there to consider. A lot can also be done with setting up a spreadsheet.

Regards,

 

[tomwalz]

Parkinson - work expands to fill the time allotted for it.

Thomas J. Walz
Carbide Processors, Inc.

http://www.carbideprocessors.com

 

[tripleZ]

At 200 pieces an order, is this an easily reproducible process or a difficult to reproduce process? You need to look at both the skill level required for the task and the rate at which someone learns. For me, where my presses cycle 6-20 parts per minute, my operators get to steady state pretty quickly. For a more detailed, small-batch machining operation featuring longer process times, you may find that you need a few more iterations and production runs featuring different personnel to get an accurate depiction of your process.

Per Tom Walz's statement, you also first need to observe the process and put in some defining factors (e.g. set this part up in a Harig fixture within .0003" TIR, then dress radius on wheel, etc.). If you don't define the process, you won't get accurate measurements and will be permitting people to work against a base standard.

 

[papimonte]

The study will include control charts and performance index graphs. The processes are assembly and some machining of small parts,and testing. e.g. pressure regulators. There are some larger parts, also. I will be looking into raw material handling rather than assembly for the larger parts, since assembly has being optimized.

 

[PSE]

papimonte

After you do your work on raw material handling, I would recommend going back and checking on assembly. You indicate that it is currently optimized. That may no longer be the case as you make changes or improvements in the "upstream" processes.

Regards,

 

[ietech]

Try this thread it has some good info thread284-73360

 

[automatic2]

I've always considered the classic time study to be an exercise in reverse. Here's a procedure that we implemented to establish work standards;

we selected a unit build that represented a 'stock' unit. This drive has approximately 400 pieces delivered to the cell in sub assembly organization. We delivered two identical units in two adjacent cells. The manning was one journeyman mechanic with two apprentices to perform two simaltaniuos builds. All three individuals had 6 month product training.

Tooling, consumables, and documentation where completed. In addition, two engineers where assigned to negate part non conformance. A junour engineer was assigned to record process.

Several qualifications where established. The largest one being that the assemblers represented what we deemed 'base line standard' in regards to knowledge, trade expertise, and work ethics.

This exercise was in the end, invaluable for all. Not only did we build two of our highest quality drives, we built them in record time.

Now we are working similar exercises in the departments that support manufacturing.

 

[ietech]

"we built them in record time".

You did a very good job of stacking the deck. The good quality is great though.

I don't intend to be negative but a good time standard in a production environment is an average of many studies with a LARGE population of varied skill levels.

There will be the very fast and the very slow which will be either cancelled by the person doing the study, or if the sample is large enough the numbers alone will bring them into line. There are some folks who will try to sabotage a time study by rushing, or the opposite, dragging their feet.

Record times, under excactly ideal conditions are not how you want to establish an accurate standard. Of course unless you intend to run a sweat shop.

Just my experience -- if perfect works for you then "Go For It".

 

[tomwalz]

Are you familiar with the Western Electric study at their Hawthorne regarding light levels? I was taught it in first year management.

http://www.newagepublishers.com/samplechapter/000920.pdf

section 1.3.7

In my experience, you have to take all hours used over a week or a month and all parts produced and divide them. Then you have to use some form of multi-variable analysis to isolate various issues.

If you watch people then they will work harder and faster than is sustainable on a long term basis. You will also have more pre and post operation support, such as parts staging, than is typical. Also you may have a huge amount of conversation about how they could do the job faster or why they aren’t doing it faster. Examples might be; the machine is slow / cold today, the parts are harder to do than average, the new material isn’t very good and we have to use more of it and so on.

My favorite technique was taught to me by a supervisor many years ago. He would very carefully set up an ideal work station then sit down and work like crazy for 5 minutes. If he had started another part at the end of the 5 minutes, that part counted as finished even if it wasn’t.

He was always the best man in the shop. Employees never met his standard because they were lazy and his supervisors didn’t work hard enough. I worked there 3 months and still have vivid memories of his management style.





Thomas J. Walz
Carbide Processors, Inc.

http://www.carbideprocessors.com

 

[gerhardl]

What is the purpose of your time study? As a time study always is administrative work, costing money, it should always be stressed that the end result (earning money) is focused.

Some possible control questions depending on the purpose of the study, trying to keep the study simple and cheap follows:

Methode and work improvement purpose. Can the same result be obtained by asking/involving the operators?

Setting external job price purpose. What price will the market bear? Signals from the market instead.

Controlling internal cost and work amount purpose. Can a coarse guesstimate based on already existing data be sufficient (not less than - not more than) and then gradually buildup and improvement of a database over time?

Can other motivating methodes be utilized in stead? ('The weeks best production team' ?)

 

[trailcarnage]

Takt Time = Available Time / Customer Demand

 

[automatic2]

'Record times, under excactly ideal conditions are not how you want to establish an accurate standard. Of course unless you intend to run a sweat shop.'

Don't misunderstand the intent of such a process. I can weed through a million variables to expose the core activity, or I can just perform the activity itself, then observe how each variable impacts the effort. There is no trickery or evil intent, it's simply exposing the root activity, then analizing how well the supporting perform.

 

[BillPCMfgE]

I have never seen a customer pay for a time study. Time studies are waste in general but they offer insight into work content of a process and waste content. Eliminate or reduce the waste. Walking to get a part or a tool. Waiting for something to occur. Overproduction.