Plate Remakes

[Red_Right_Arm]

I'm running into an issue where our logged plate remakes is becoming too excessive for upper management. The problem is that no one is sure what is an acceptable tolerance to use as a baseline. Depending on whose opinion I am getting at the time the acceptable level seems to change. Someone says that they fully expect to see plate remakes completely eliminated. Someone else says that the important thing is to get them under control. Someone else says that 123 remade plates in a month is too much, but can't tell me what is acceptable. Someone else says that 40 in a month is okay. Someone else says that they should be below 10%, but can't say what the 10% is in relation to. Basically, everyone is guessing at what they think sounds good. No one is really sure what sort of remake statistics a company can use to see if we are really doing a normal amount for any company, or if we are completely screwing the pooch.

How do other prepress departments determine their acceptable level for plate remakes?

 

[gordo]

I'm running into an issue where our logged plate remakes is becoming too excessive for upper management. The problem is that no one is sure what is an acceptable tolerance to use as a baseline. Depending on whose opinion I am getting at the time the acceptable level seems to change. Someone says that they fully expect to see plate remakes completely eliminated. Someone else says that the important thing is to get them under control. Someone else says that 123 remade plates in a month is too much, but can't tell me what is acceptable. Someone else says that 40 in a month is okay. Someone else says that they should be below 10%, but can't say what the 10% is in relation to. Basically, everyone is guessing at what they think sounds good. No one is really sure what sort of remake statistics a company can use to see if we are really doing a normal amount for any company, or if we are completely screwing the pooch.

How do other prepress departments determine their acceptable level for plate remakes?

What is the nature of the remake?
- Difference in plate image content vs source file?
- Variation in tones on plate?
- Plates not delivering expected tone in presswork?
- Plates physically damaged in transit?
- Variation due to different plate batches?
- Or?

 

[Red_Right_Arm]

What is the nature of the remake?
- Difference in plate image content vs source file?
- Variation in tones on plate?
- Plates not delivering expected tone in presswork?
- Plates physically damaged in transit?
- Variation due to different plate batches?
- Or?

- What is the nature of the remake?
-- Generally they are caused by one of eight categories, ranging from human error in the digital files to the press chewing up the plates.

- Difference in plate image content vs source file?
-- Sometimes, but rarely. We found out the RIP doesn't handle live text with post script fonts too well. So we've started outlining fonts before sending them to the RIP.

- Variation in tones on plate?
-- I'm not sure what you mean. Like dot gain?

- Plates not delivering expected tone in presswork?
-- Sometimes. We have generic curves that need to be adjusted constantly for specify press variables. But we tend to not count these on the remake log as they are viewed as normal requirements to get a job up and running.

- Plates physically damaged in transit?
-- Sometimes. We log remakes when that happens, yes.

- Variation due to different plate batches?
-- We haven't run into this issue.

- Or?
-- I'll break down our tracking metrics below.

- Each department has a letter code. A = Prepress, B = Press, C = Customer Service, etc...
- Each general error has a number code. 1 = File error. 2 = Mechanical error. 3 = Damaged plate, etc...
- We only track plates that were remade due to someone being at fault. Back ups, test plates, pressmen have an idea for improvement, etc... These don't get logged.
- When a mistake happens we log the job info, how many plates are needed to be remade, the total square inches of material used, department code, error code, detailed reason.
- We can see that out of 123 plates being remade in a month, 58 of them came from errors that originated in Prepress, 23 from errors that originated in Customer Service, etc...
- We can see that Prepress (or whatever department) has a reoccurring trend of file errors. We look into what's causing that issue.
- We can see that Prepress is the cause of 44% of the issues, Set-Up is the cause of 15% of the issues, Press is the cause of 20% of the issues, and so on.
- We can see the cost each department caused to remake the plates.
- We can see the departments that caused time lost on a down press waiting for plates.

We can gather all this data, but we don't seem to have any idea if it's normal for Prepress to screw up 55 plates in a month. Or if it's acceptable for Prepress to be the cause of 38% of all plate remakes. We don't know if other companies with better quality control methods are only screwing up 1 plate per month or if they'd call 200 a good month. Basically, we've spent so much attention gathering data that we never put any thought into establishing our acceptable baselines assuming everything is running as expected. Now we are dealing with multiple guesses as to what is or isn't normal and acceptable.

 

[wonderings]

We are not a massive shop, go through maybe 100 - 200 plates for our 4 colour Komori and maybe 50 plates for our QM. We do not have to remake plates often, I would say it is around 1 - 2% a month, if that. For us, everything is checked before it is plated, I always print out a sample sheet on an 11x17. It is reduced, but I can double check visually that everything is ok.

I cannot remember the last time a press destroyed a plate, has to be a few years. It has happened, but far form a regular or normal thing.

Same with damaging plates in transit, been a long time since anyone dropped a plate or scratched one so it was unusable here. Wrecked or wrong plates are not a normal thing for us. Not sure if it is how we handle them. When making plates they go directly into a plate box, never really being transported around on their own, always in a box. In the past (10+ years ago), there used to be a lot of remakes because of human error, but that was before I started.

 

[Red_Right_Arm]

there used to be a lot of remakes because of human error, but that was before I started.

So what did you do to reduce or eliminate human error? Just a good eye for detail? Or was there a workflow change that everyone started to follow to be more accurate and reliable?

 

[gordo]

We do not have to remake plates often, I would say it is around 1 - 2% a month, if that.

That, in my experience, is quite typical irrespective of shop size.

 

[wonderings]

So what did you do to reduce or eliminate human error? Just a good eye for detail? Or was there a workflow change that everyone started to follow to be more accurate and reliable?

Eye for detail and an understanding of what I am doing. It did not start when I started, I came in with pretty much no knowledge in how pre press worked. It is my family business so I care more I am sure. The other pre press person just does not see things. No new software, apart from me shifting us to indesign from quark a few years after I started. We don't have any fancy software, we print from acrobat or indesign to our RIP and make plates.

 

[jimas67]

We have someone that does proofing after prepress does the layout. Then we have someone else ( me) that looks at the images on the pc for layout, proper color densities , bleeds, sizes and cut marks. I also make sure that everything is in it's proper order , such as the layout of a magazine. We have many eyes that look at a job before it does go to press. We have some plate makeovers but not as many as were described.

 

[dabob]

For the life of me I cannot imagine having over 1-2% remakes over any specific period unless it was a bad batch of plates . . . . and even then you should catch it before it exceeds that percentage

 

[jlee]

I think you're referrring to flexo plates based upon other posts you've made. I think everyone here is referring to litho plates. However, 38% is way too high, especially with the cost of flexo plates in comparison to offset litho. My shop produces around 500-600 flexo plates per month and I would have to say that we might remake 2-3 plates per month due to prepress error. When I worked at one of the biggest flexo plate suppliers in the world we had about a 5-7% error rate due to operator. In terms of offset, we produce around 2,000 plates per month and remake due to operator error is around 5-10 plates per month. As previous posters have mentioned, we look at our plate files before we even manufacture plates and due to the cost of flexo plates alone it's worth it.

 

[davarino]

Red_Right_Arm,

As I remember, you are in a flexo shop. So these are all flexo plates, right?

There will be a higher remake rate on flexo plates than litho plates, but not horribly high. (From my flexo days, I'd imagine 10% would be REALLY high. Then again, we were often running some really "non-critical" work as flexo 40 years ago.)

What is your overall "remake vs. total plates" percentage?

That kind of a number is needed in order to see the picture. (128 plates a month is negligible if you are hanging and banging 4000 plates a month, but it's abominable if you're running 400 plates a month.)

It's really not so much about setting a target and then hitting it, though. (Although it's always useful to know where you are on the grand continuum.)

Something that I've learned in many shops is that if you set a target, the target will usually be hit but the problem will often not be solved.

Reasons: one of the most common is under-reporting (after all, who wants to risk getting someone fired if you're in the midst of a snowstorm every morning... you need all the help you can get). Some others are shop politics, sloppy reporting, misstatement of the problem, "non-problems", and misunderstanding of the production process.​

So someone thinks that the remake ratio is too high... I would say that that is irrelevant except in a shop-politics sense unless you know the production context.

It REALLY can be that you really DO need to throw away 75% of your plates if your system is based on trying to get quality by inspection rather than planning. (There are reports of factories in underdeveloped [and under-educated] countries where products rejection rates of 90% are common, but accepted because the cost of production is so low.)

It CAN be if your equipment isn't designed to dependably make the kind of plates you need.

It CAN be if your CSRs do not get the specifications right.

It CAN be if a strategic person has a drug or alcohol problem.

These problems ALL are management problems. It will not help if the departments "work harder". It might even be harmful for them to "work harder". (A hung-over, tired press operator is worse than a hung-over press operator.)

If your biggest problem is, for example, truly in customer service: the solution is to make sure that (1) the CSRs all understand what they are supposed to do and then (2) that they do it. Anything you can do to make these tasks easier is worth considering. That is management.

Solving a remake ratio problem is much more than pinpointing department-by-department. That kind of pinpointing is like looking for all the nouns in a sentence and not paying attention to the verbs.

(Read anything by W. Edwards Deming. His writings are relevant for anyone looking to learn to control output from a system.)

 

[Red_Right_Arm]

We have someone that does proofing after prepress does the layout. Then we have someone else ( me) that looks at the images on the pc for layout, proper color densities , bleeds, sizes and cut marks. I also make sure that everything is in it's proper order , such as the layout of a magazine. We have many eyes that look at a job before it does go to press. We have some plate makeovers but not as many as were described.

As you run through these checks, re-checks, and re-re-checks, do you use sign offs to track who checked what? Or is it just built into the normal workflow that this is the way it's done and everyone follows? Basically, our department is only 3 people, and sometimes our department gets pulled in more than 3 directions at once (just like every other department, right?). So I can imagine us trying to do these various checks at first, but then slowly drifting back to not doing them because they'll slow down the workflow. Also, if there is a problem that requires a remake, it'd be nice to know how it was missed and by whom. But our current workflow already requires a lot of sign offs. I'm apprehensive about requiring my team to have even more sign offs to do on top of them. But if they are required, they are required. I'm just wondering how you handle the backtracking if it isn't done aspect to your workflow?

 

[Red_Right_Arm]

For the life of me I cannot imagine having over 1-2% remakes over any specific period unless it was a bad batch of plates . . . . and even then you should catch it before it exceeds that percentage

I think I stated the numbers in a misleading way. We don't have 38% plate remakes. It's that out of all of the remakes that are required, we can see if Prepress is the cause of 38% of them. As far as how many total plates we remake compared to how many total plates we make, I am not sure. Our system currently doesn't track how many plates we make unless we start to track and log that info manually; which we have no done currently or in the past. I can tell you that as a guess 1-2% total remakes compared to total plates is probably a figure we are not hitting.

 

[Red_Right_Arm]

I think you're referrring to flexo plates based upon other posts you've made. I think everyone here is referring to litho plates. However, 38% is way too high, especially with the cost of flexo plates in comparison to offset litho. My shop produces around 500-600 flexo plates per month and I would have to say that we might remake 2-3 plates per month due to prepress error. When I worked at one of the biggest flexo plate suppliers in the world we had about a 5-7% error rate due to operator. In terms of offset, we produce around 2,000 plates per month and remake due to operator error is around 5-10 plates per month. As previous posters have mentioned, we look at our plate files before we even manufacture plates and due to the cost of flexo plates alone it's worth it.

I think I stated the numbers in a misleading way. We don't have 38% plate remakes. It's that out of all of the remakes that are required, we can see if Prepress is the cause of 38% of them. As far as how many total plates we remake compared to how many total plates we make, I am not sure. Our system currently doesn't track how many plates we make unless we start to track and log that info manually; which we have no done currently or in the past.

Right now our system can track jobs, not plates. No metrics are built for how many plates a job has. So that info cannot be easily found by looking at the job data. The best we can do is look at a job in the ordering system, then find the proof, then find the number of plates stated on the proof, then find the number of plates we remade for that job. In order to find total plates vs total plate remakes we would have to open every proof for every job, manually total up the plates from each job, then look at the total remakes on our remake log. Not very practical or efficient way to gather the data.

 

[Red_Right_Arm]

Red_Right_Arm,
See below

These are all flexo plates, right?
- Right.

From my flexo days, I'd imagine 10% would be REALLY high.
- As a gut feeling guess, I'd say we might be around that mark.

What is your overall "remake vs. total plates" percentage?
- Unfortunately our metrics and statistical data can only be captured manually on a handwritten log, which we have not been doing ever. There's no way to know beyond a guesstimate how many total plates we made in a week, month, or year unless we start manually writing down a log of every plate we've made. So far the workflow was built to ONLY track the number of remade plates. The idea being that the totals don't matter. As I am told by upper management, "The goal is zero. 1 remade plate is just as bad as 100. So if you have any remakes at all in a month, that wasn't a good month."

128 plates a month is negligible if you are hanging and banging 4000 plates a month, but it's abominable if you're running 400 plates a month.
- That's the kind of mindset I had until recently. I was trying to find out if we were doing bad, really bad, or really really bad. But it's been explained to me recently that I need to treat 1 remade plate as bad as I'd treat 100 remade plates.

Something that I've learned in many shops is that if you set a target, the target will usually be hit but the problem will often not be solved.
- It's been recently explained to me that our target is total perfection knowing full well we'll never get there.

It REALLY can be that you really DO need to throw away 75% of your plates if your system is based on trying to get quality by inspection rather than planning.
- Although the total plates vs total remakes isn't as high as 75%, I see what you're saying. Our system is indeed based on getting quality by inspection. The planning part is harder to lock down because other departments get involved. Each department knows for sure they are doing the best they can do, but since the plates come from Prepress it is Prepress's responsibility to make them perfect every time. Someone from ordering tells us that this is a window cling, they assume we'll understand that the plates need to be made mirrored and won't mention that spec to us... doesn't matter, it's Prepress's responsibility to know or ask or find out. Someone on press is going to need to use a heavier anilox to beef up the ink density, and we can't know which anilox will work until they try it on press... doesn't matter, it's Prepress's responsibility to account for that. Someone puts the wrong die on the job specs we make plates according to that info and it causes an issue on press when they are running with the correct die and our plates won't work... doesn't matter, it's Prepress's responsibility to catch that error and not let our plates become the problem on press. If any of these issues arise the suggestion is stressed to make more notes, ask more questions, and do more double checks in the future.

It CAN be if your equipment isn't designed to dependably make the kind of plates you need.
- Could be a small part of it.

It CAN be if your CSRs do not get the specifications right.
- Sometimes. I wouldn't say "not get the specs right". Instead I'd say it's more "assume you know how to do this without all the specs fully spelled out."

These problems ALL are management problems. It will not help if the departments "work harder".
- The suggestions for us from upper management and outside departments is indeed to work harder, focus more, take more notes, and ask more questions.

If your biggest problem is, for example, truly in customer service: the solution is to make sure that (1) the CSRs all understand what they are supposed to do and then (2) that they do it. Anything you can do to make these tasks easier is worth considering. That is management.
- Good suggestion. I tried it. It's been met with "You know, we do our best but we are figuring out the job as we build it ourselves. We have to figure out the right label stock, adhesives, laminate, shipping date, press run time. And to have to also figure out for you what's needed for Prepress just makes our jobs impossible. You're Prepress. It's your job to know what you need and what you need to do. Whatever notes you have to take or references you have to make... you know... maybe you make a big poster that says in big bold letters 'EVERY JOB NEEDS THE RIGHT BEARER BARS'. Whatever it takes. But you have to start making plates reliably from now on or else..." That message coming back in the form of a non-threatening clear the air meeting, to someone yelling across a table.

Solving a remake ratio problem is much more than pinpointing department-by-department. That kind of pinpointing is like looking for all the nouns in a sentence and not paying attention to the verbs.
- While I'd tend to agree, the people who sign my paycheck have made it clear Prepress alone has to solve this problem and no other department is expected to change in order to help. They're all doing as best as they can and Prepress is on our own in making every plate perfect, every time.

(Read anything by W. Edwards Deming. His writings are relevant for anyone looking to learn to control output from a system.)
- I've been looking into him. Seems like a knowledgable guy that no one agrees with. The Red Bean and Funnel experiments went over like a lead balloon when I suggested looking into them.

 

[dabob]

I think I stated the numbers in a misleading way. We don't have 38% plate remakes. It's that out of all of the remakes that are required, we can see if Prepress is the cause of 38% of them. As far as how many total plates we remake compared to how many total plates we make, I am not sure. Our system currently doesn't track how many plates we make unless we start to track and log that info manually; which we have no done currently or in the past. I can tell you that as a guess 1-2% total remakes compared to total plates is probably a figure we are not hitting.

Well . . . you could arrive at a number pretty close to total plates by just looking at the plate invoices for a given period . . . if you buy 200 plates a month . . . . it would be a pretty good bet that you are making about 200 plates a month . .

 

[bean66]

You may need to start tracking remakes and specify where the error originated. Then you can categorize the errors and try to fix or improve on them. We have probably less than 1% remakes for our shop

 

[davarino]

Right_Red_Arm

My sympathy goes out to you.

But: recognize that you have an opportunity to become the manager that it seems no one else wants to be. Maybe not in title, but in reality.

It's not easy. There's no visible glory in it. You can lose your job because of people who feel threatened.

But maybe you can swallow your remaining pride and your disgust, bearing in mind that the livelihoods of your fellow workers require the best management. If you do this, you will feel what only the great can feel.

 

[davarino]

Looking your reply over once again, I see that you have essentially been given the job of quality assurance. Apparently it's because anyone else who could be approached with the task refuses to take the responsibility, for good reasons or bad.

You might want to look at the notion of someone (probably yourself) analyzing EVERY job that comes through pre-press and checking every process and spec against reality. That would mean walking around and asking pressmen and CSRs and whoever else to the point that you understand precisely what is wanted and how it will be accomplished.

Of course that takes a heck of a lot of time. And nobody wants to do it.

(As an aside, my Dad used to say, "If you want to get anything done, just do it and don't take any credit for any part of it." If you take no credit, you get the job done, get little blame, and get almost no resistance.)

It would probably become a full time job, depending on your shop's workload.

And after it was in place, you would be both indispensable and overworked. So long as the pay is right, that's not a bad option. (And I suspect your boss(es) would see an increase in "quality", throughput, and profitability, whatever those buzzwords mean this week.)

I suggest that you look at this as a grand opportunity.

If your employers need you, that's good. If they can depend on no one else to deliver, that's a competitive advantage you have.

If they (arbitrarily) want to have not a single plate done wrong, then the consequence of that is that there must be someone eagle-eyeing the entire production process to make sure that there is no missing information... you.

Again, tread lightly. Someone is actually handing you the keys to your future, and I think you can profit if you play it right. Don't be fanatic, and make good happen gently.

 

[baritone]

I know the feeling on this issue. The problem here at my work is they like to jump the gun and make us plate the job before the customer approves it and like an hour later they have a correction then the plates are no good. I'm sorry but to me its always the CSR that are 1 too lazy to look at the files 2 doesnt input the information on the ticket right or 3 like to rush to make everyone busy. But it might be my work lol. I'm just an IT guy that no one can help but I can help everyone else in their department.