Hello!
I have some questions,I shall be very grateful for your answers.
Whether it is necessary to apply control charts on a place of the final control (100 % control of all pieces)?
Can I use other instrument?
How manage key characteristics ( indicator lapse)?

P.S. Sorry for my English.

Welcome to The Cove! :bigwave:

SPC (control charts) are not intended to "provide" 100% control, but they are instead designed to understand when a process goes out of control. It's intent is to minimize defects, yet avoid rushing into a process and meddling when there is no good reason to do so.

The way to achieve 100% control is to remove all causes of error. We don't usually manage to remove 100% of all error. We instead prioritize and assess results of carefully designed and implemented improvements, so we can learn our lessons and progress toward excellence.

Can you please be more specific and describe for us what type of control you wish to achieve on what sort of product, during what kind of manufacturing? Maybe then we can give a better answer than I have posted.

Welcome to the Cove :bigwave:

I agree with Jennifer. A process/product characteristic that has 100% "inspection/control" will cover your "key characteristic" requirement if that is included. However - it is certainly not the most economical means to maintain control. As you gain understanding of your process, you "graduate" to SPC/SQC (the SQC term is "archaic"), you gain the benefit of not needing the 100% coverage (hopefully :rolleyes:). As you gain statistical control there is the benefit of sampling to maintain control. Starting with the 100% is OK as long as someone is "watching" the data and trying to improve the process.

I'm not sure if that makes sense, but I think I'm agreeing with Jennifer. A 100% control is, normally, not feasible for a manufacturing process to maintain a "margin" (plus 100% "inspection" guarantees nothing other than if you're process is "able" to make a nonconforming part, you are also "able" to ship it).

Just :2cents:

With 100% inspection programs, one is generally relying upon an individual inspector to judge if the final product is good or bad, and to discard (or send back and rework) bad product. The difficulty is you are generally relying on one person, who may manage to miss defect(s) for one reason or another (hey - they are human).

A better course of action may be to build monitoring of parameters during the building process using SPC. If the SPC finds out of range defects, then you have a record of when they occurred. You can also determine if this was a batch problem (affecting several parts with a common cause) or an infrequent event with a special cause. This helps with process improvement and elimination of defects. The final inspector with 100% inspection is less likely to be able to determine how to fix the source of the defects, since they are separated from where the defect was created, and a busy with the inspection task.

So, a good SPC monitoring program of in process steps should help to eliminate defects, and eliminate the need for 100% inspections as you become comfortable that the process is routinely capable of meeting specifications without creating defects.

Thank you for answers.
We produce instrument cluster for cars. Key characteristic is speed indicator lapse. Operator controls it on end control stand (100 % of output). We can’t eliminate the need for 100% inspections, because it is acceptance (approval tests). When defect appear, operator stops check and the kind of defect is automatically registered. Using p-charts was supposed. Is there any sense to spend efforts on SPC here?