Downtime on SMT lines

Where are the experienced operators? I'm sure they would have some good input to what it is you seek. How would this program you wrote know if the operators went to the bathroom during setup? Or took break? Or even waiting on incoming parts from shipping?If activity Sampling takes up most of your time, maybe your boss doesnt really want to know the causes, I'm surprized he doesnt know already. If it helps we have an average setup time for most jobs(which are setup offline). By the way I dont think Setup time should be considered downtime, its part of the job and should be calculated in planning. Analysing the downtime of the line proves difficult for you? What would be so hard about asking the operators to jot down how long it took to perform a given task and what problems were encountered? I really think you should grab an operator who knows the area and work with him/her on the matter. After all you are the student.

Hope it helps DL

reply »

If you have the individual times for each machine you should be done. Simply find which is your bottle-neck for each day and use that info. You can probably write a seperate program to do this step also, I assume you are viewing all the down time information in a speadsheet format( offline ).

If this is to be a new process you might want to look at some of the optimization softwares on the market. Many will do this for you.( Even with mixed vendor lines )

That's just my opinion, I could be wrong!!!

reply »

Hi Willem, you may be better off to study the archives of Universities. I know that at least the Universities in Berlin, Munich and Georgia Tech here in Atlanta are studying the manufacturing process on SMT lines. I was once exposed to a quite complicated process, which involved keypads on every machine station, in order to collect stop times caused by machines and operators. Needless to say that it was not well received by the operators, who had to press a key first than work on the problem and press another key when they were done. It actually required another operator to collect the data, although all key pads and the machines were connected to a host computer. You may collect data for a month or so, before you piss every one off and disconnect your stuff. The SMT process is not different from any other manufacturing process. In line processes may cause stand still times of the entire line due to stop times on one station and independent work cells are tried all over the world to overcome the bottleneck problem. Try to be diplomatic to your boss and the employees. Figure out what your boss wants to conclude out of this study. Out of my experience, he does not want to hear that the machines break all the time or that his employees spend too much time on the bathroom. Good luck, Stefan

reply »

Hi Friend. I recently went through this same project in my factory. Initially, I spent quite a bit of time out on the lines (6) collecting data. After getting an overall idea of where and what the activities were I created forms and work instructions for the machine opps to collect the data for me.

overall, I used 15 categories to assess down time activities. Anything greater than 5 minutes was to be loged in as down time. After 4 weeks I had suficient data to create project plans for each line and assign Process Technicians to supervise the project progress.

The most important thing is to not get swamped with data. Start small, alter course as data is reported and start all over again until your process is a fine tuned money making machine. PS it helps to have an access database and data entry person as well to keep your "feet outa the mud".

Good luck. And remember, You can not control what you do not measure!

reply »