Printed Circuit Board Assembly & PCB Design SMT Electronics Assembly Manufacturing Forum

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


SMT productivity measurement

Stoney Tsai

#9693

SMT productivity measurement | 9 September, 1999

Currently, I have been requested to measure our SMT productity, capacity levels. I employed "pcs/day" as a base to come out with indicators for the top management decision-making. Anyone has other ideas to measure these two items?

What are the latest benchmarking for SMT productivity evaluation?

Best regards &

reply »

#9694

Re: SMT productivity measurement | 9 September, 1999

| Currently, I have been requested to measure our SMT productity, capacity levels. I employed "pcs/day" as a base to come out with | indicators for the top management decision-making. Anyone has | other ideas to measure these two items? | | What are the latest benchmarking for SMT productivity evaluation? | | Best regards & | Stoney: Your SPC program should be a good basis for beginning. Ta. Dave F

reply »

Scott S. Snider

#9695

Re: SMT productivity measurement | 9 September, 1999

| Currently, I have been requested to measure our SMT productity, capacity levels. I employed "pcs/day" as a base to come out with | indicators for the top management decision-making. Anyone has | other ideas to measure these two items? | | What are the latest benchmarking for SMT productivity evaluation? If you a talking about SMT using pick and place equipment the best way to measure productivity is by equipment utilization. Most equipment will give you data broken into several categories - placement time, transfer time, wait time, error time, maintenance time, etc. To get utilization take the good time (placement and transfer time)and divide by 24 (or hours people are staffing the equiment)hours per day. This gives a real picture of how much time your machine is really working. A company building a mix of products should be between 60 - 75%. If you are only buidling a single product the number should be 90 - 95%. Most people are shocked when they look at the data provided by the machine. Whenever I walk into a contract shop I always ask to see the data currently on the machine. Often the utilization is between 30 - 40%. I believe this is better measure of how the process is managed than raw parts placed because of the variations between pick and place equipment capability. As far as measuring capacity you must look at parts placed. Once again look at the machines data system. It will track the number of parts placed. Make sure to divide the number of parts placed each day by 24 hours. In theory you could run 24 hours per day. If you take this data for a month you will have a good picture of your equipment capacity. You should also add a quality measurment to help balance the other two measurements. It is always possible to max the equipment out while swamping the rework people.

reply »

Earl Moon

#9696

Re: SMT productivity measurement | 10 September, 1999

| Currently, I have been requested to measure our SMT productity, capacity levels. I employed "pcs/day" as a base to come out with | indicators for the top management decision-making. Anyone has | other ideas to measure these two items? | | What are the latest benchmarking for SMT productivity evaluation? | | Best regards & | At my last contract site, a very simple measurement process was used though it had very formal foundation. Placement numbers were all that mattered.

During my last month at that location, we targeted 16 million placements up from 12 the month previous. Does sound simple but with nearly 40 customers and 1500 board numbers, it required some thing more than the PPPP principle.

The placement concept allowed us only to focus on meeting daily numbers no matter the product or customer. We ran 90 product change overs a day on six lines. Placements were conidered no matter the operation from the front end to back.

All cost models were set up and running in the background so all labor, material, and overhead issues were "invisible" to production. Profit was clearly visible as placements met or not.

We used a simple flow chart system to describe all nine board types available to be processed. The flow charts were turned into routers and all operations were numbered that did not change as process steps were not important. An example would be kitting (10), stencil printing (20), PNP (30), and so on.

We had one half hour quality huddle meetings promptly at 0900 each day to initiate immediate corrective action for anything negatively impacting daily, or shift related, placement numbers. Each critical area was represented as process, manufacturing support, and quality engineers as well as maintenance, materials, line supervisors, etc. As an example, I owned DEK, among others, so any problem constraining placement numbers relating to stencil printing were my responsibility. Therefore, I had until the next day's meeting to effect CA. Usually, problems were corrected much sooner except for those long term. Other process ownere had the same responsibilities.

This all gets back to placements. Marketing was able to project all production requirements based on placements. This was true for all other areas and departments.

Thought you might want to see something else being done, though not unique but very efficient.

Earl Moon

reply »

PCB X-Ray Inspection

soldering station