Toyota SMED (Single Minute Exchange of Dies) Clarification

[duecesevenOS]

New to the forum so bear with me.

I haven't seen any topics that include information about single minute exchange of dies but I've been looking into it for my area of a plant. I have been overseeing the implemention of a lot of toyota production system / lean manufacturing ideas (kanbans, 5S, takt time...).

The company is an ammunition manufacturing company and I specifically work in an area that manufactures the very small primers that are assembled into ammunition. All of our lines are completely dedicated to one product so there are never any actual "exchanges of dies." Because of the very small tooling and the fast run rates we do have to repolish tools repetitively. This results in a very significant amount of downtime while punches and dies are brought back into spec.

Could someone with some experience clarify wether or not SMED would help me? I think some of the theories would work in my situation but I don't know if it would be worth going through all the steps with my operators.

Thanx

[triner]

If you have multiple sets of tooling, I.e. you take out one set of tooling, replace it with another, and polish the tooling offline than yes, the principles of SMED will apply.

If not, than most of the SMED tools may not apply. SMED gets its biggest bang from converting internal work (work done while the machine is down) into external work (work done outside of the machine, while it is still running).

If the machine has to be down while you polish the tooling, than simple organizational things like making sure all of the tools and material are available at the machine, and that the operator can reach them without walking or stepping will reduce the time.

[CarolX]

ducesofsevenOS - Welcome to the Cove.

If you scroll down to below this post you can find links to similar threads.

[KReynolds]

I agree with Triner to that extent regarding SMED. If the machine has to be down for the polishing, it may be a good candidate for Total Productive Maintenance (TPM). Is there anything that the user can do as a daily routine to extend the downtime periods - which would be incorporated into the Shine of 5S. And if there is variance in the machines, a good Black Belt would suggest Six Sigma.
In summary, in order to really know what to do, one needs to look at what is causing the need for polishing, is it uniform within the plant, et cetera.

[duecesevenOS]

I don't think the polishing is avoidable. This is a process that is stamping then drawing out over 6 million parts a day with specifications that have to be held to a .0003" range. Even carbide punches can wear that fast. Also, a lot of the changes that have to be made are made to accomodate the material going into the process because our vendor can't meet a higher specification.

We currently do not switch out the punches when there is product running out of spec so downtime is probably around an hour while the operator waits on reconditioning. I think we could get this down to 20 minutes by simply giving the operators a set of already finished tools that they can switch out as soon as the part goes out of spec. I'm working with production to put together a tool reconditioner (supplier) to press operator (customer) kanban right now.

I think SMED could help but I'm not sure I need to follow the 4 step process. I'm thinking it must be pretty straight forward and obvious (most everything Toyota is). My plan right now is to do some kaizen with the operators and get the same kind of results. It sounds like SMED is just a specific kind of kaizen event. Am I right? Theres a lot more information out there about how to do kaizen events than there is on SMED.

I might buy Shigeo Shingo's book to learn a little bit more on the subject though.

duece

[triner]

Your plan to have a kaizen event sounds good. I think that will allow you to capture the low hanging fruit very quickly. Productivity Press has a nice work book on SMED that is based on Shingo's book. Buying Shingo's book is not a bad idea either.

They have a good examples of time reducing ideas such as eliminating bolts or altering the bolt so only a half turn is needed to make it tight.

Good luck!

[gszekely]

Hi
I have posted some links in the below topic. Check them out.
http://elsmar.com/Forums/showthread....648#post131648
BR
György

[wmarhel]

Originally Posted by duecesevenOS

It sounds like SMED is just a specific kind of kaizen event. Am I right? Theres a lot more information out there about how to do kaizen events than there is on SMED.

I might buy Shigeo Shingo's book to learn a little bit more on the subject though.

Shingo's book is a good reference. You can "run" an event in an attempt to improve the time for die changeover, but some of the true improvements won't be completed in a week. Don't get me wrong, you could very possibly see upward of 60-80% improvement. In many cases that has less to do with the breakthroughs achieved in the event, and more to do with how disorganized the process was in the original state. Many times this very example happens, everyone pats themselves on the back and then people forget about it.

SMED is also about looking at possible ways to re-design the tooling/equipment. Maybe it's standardized die beds which are rolled in and out and then secured using 1/4 turn screws. Typically, that type of work and those items aren't available at the local Home Depot. This practice also dovetails into future equipment acquisitions. How it should look, operate, etc.

A big part is also about attaining "good" parts with the least number of set-up pieces. Reducing the time to change out a part by 50%, but still requiring 20-30 pieces for set-up (# of pieces to get the first good part) may offer only marginal benefits if the wasted material costs are eating away at the overall gains from reduced labor time. This would be especially true if there was already excess capacity at that particular machine.

Wayne