Guest Podcast Episode 2
Manufacturing Rework Best Practices
Rework issues could eat up all of your margins.
There could be several reasons why there might be rework orders.
See Below for the full transcription of this episode!
Sam Gupta (00:23):
Hello everyone. Welcome to today's show. And if you're joining for the first time, this is part of our digital transformation series for which we meet every Thursday at 5:30 PM Eastern. We pick one topic related to digital transformation, and we always have an expert panel here for today. We are going to be talking about a very deep topic. It's manufacturing WeWork, but if you can reduce that, you can make a lot of money. Um, so we are gonna have a lot of fun discussing that. Before we do that, we are going to start with everybody's intros. I am going to start with my intro. If you don't know me, Sam Gupta, principal at Elevate iq. Elevate IQ is the independent e r P and digital transformation considering firm. On that note, I am going to move to Chris for Astra.
Chris Garrad (01:07):
Hi Sam, and if you don't know me, I'm Chris Garrad, the owner and president of Turnkey Technologies. Sam's had me on, so Tom, we're like brothers these days, but, uh, thanks for joining tonight. Uh, gen Turnkey Technologies is about a 29 year old, uh, partner organization implementing Microsoft Dynamics. So great topic tonight.
Sam Gupta (01:23):
Amazing. Thank you so much for being here, Chris. Dave, can I move to you next for your intro?
Dave Griffith (01:28):
Absolutely. Hello everyone. Uh, my name is Dave Griffith. I run a company called Capsule Solutions. We help companies, organizations go through digital transformations that pay for themselves by focusing on executable small projects that, that turn large profits. And I'm excited to have this conversation about rework.
Sam Gupta (01:46):
Okay, amazing. Thank you so much for being here. Dave Nara, can I ask you to introduce yourself next?
Nirav Sham (01:52):
Yeah, absolutely. My name is Nirav Sham, CEO of Eder, c r p. We are an Acumatica partner and we help implement, um, Acumatica and mid-market, um, industries, manufacturing being one of them. And this is a very important topic for manufacturers to get under control. Um, I'm excited to talk about it. Thank you, Sam.
Sam Gupta (02:09):
Get under control. Um, amazing. Thank you so much for being here. Chuck, can I ask you to introduce yourself next?
Chuck Coxhead (02:16):
Absolutely. Chuck Coxhead. I'm also with Turnkey Technology, so what he said, and, uh, after 35 years in the manufacturing world in various capacities, this one's near and dear to my heart. How do we make money out of the things that we have to rework?
Sam Gupta (02:31):
Okay, amazing. Thank you so much for being here, Chuck. And if you're in the audience and joining for the first time, make sure you guys, uh, post questions and comments. We typically try to cover them during the show. If you run out of time, then we'll make sure that you receive your answers. On that note, Chris, I am going to start with you. And that is going to be the whole process of rework. And I don't know how many people are going to be really familiar with what rework really is and what are going to be different trigger points for rework. Um, and from the er p perspective, if you wanna set the context, press over. You
Chris Garrad (03:04):
Sure appreciate it, Sam. So yeah, rework a lot of different ideas come to mind and you think about, okay, first of all is where, where is it occurring at? You know, what, at what point in the process that somebody said, whoops, I've got a problem. And, and if you think about what triggers the event of a rework and, and what's that mean? Well, again, it could be a customer. I got my product at home and it just doesn't work. Send it back in. That's rework. Okay? So, you know, some organizations, I gotta get an rma, I send it in and they, if they made it, then they gotta fix it, send me a new one or send me a new one. But that process alone, when it comes back from the customer, there's diagnostic initial, there's diagnostic final. There may be a work order where we put the cost, whole other conversation on that one.
So again, that's discovered after I shipped 'em. A finished good product comes back in. It's a different process to deal with returns where we're trying to repair them. That's a little different than rework. But, um, we still need the data from that event to come back to manufacturing so we can eliminate this particular defect if that is the cost. So then where's the other place it could occur? I'm in, I'm in the manufacturing line, and guess what happened? I dropped it. Oh, whoops. Whoops. The machine wasn't perfectly calibrated. Okay? So again, there's, that's another example where now I'm in the middle of a manufacturing process and it's not even a quality check that all of a sudden something happens. Could be equipment malfunction, could be an operator problem, right? Dropped it, broke, it just was doing it wrong. Um, it, it also could be a, again, in the same context, it could be a quality catch, which means quality's inspecting.
And, you know, we've got medical device manufacturers and you think about the scrutiny throughout the manufacturing process. So again, what happens is it's caught after an operation and it's a quality inspection. And they're like, Hey, this isn't, this isn't right. And again, as you think about then the diagnostic on, okay, what is the corrective action? It's a whole, and again, does it, oh, we're on the line. All right, let's just start sanding it over here a little more. Stop, right? Imagine the, imagine the complexities in the engineering. What if I'm doing an airframe piece and I discover something, right? Again, maybe I'm the worker, I find it maybe quality finds it again, maybe it came back from the customer and illuminates a bigger issue. But again, there's, uh, so now the, the, the rework, again, there's a simple rework. I got one piece that I needed.
It's a little rough, I'm gonna go sand it a little bit more. And, and again, in the simplest example, does that mean that, you know, the planner can add, add an operation to the manufacturing order? So that's one example of how we get rework is we're gonna add another operation with more labor time for additional sanding, for example. And that may be in that particular scenario, maybe it's that maybe it's not enough sanding, whether that impacts all production going forward, meaning the master bomb should change the route is forever changed. Now we've got an engineering management loop. So again, as we get feedback from customer sites should go to engineering, change management to see if there's a defect. Cuz then the goal is how do we resolve? But, uh, but again, it it's business process. And again, you heard me just talk about a whole bunch of places and you're like, okay, well how should you do it?
Well, you're always costing it to the, the production. So, and, and part of that is you need to trip the variance indicator. You need to say, Hey, why do I have a variance on this? And if I'm making 50 and I got one part, okay, that's different. If I got 50 and I gotta fix all 50, that's a bigger deal. And I'll see, we're gonna see a big variance because customer price, I guess the customer's not expecting me to come back and raise the price cuz I need more labor or, you know, I found out that the part that I ordered is the wrong part. That's, that's another one of these things that could require rework is again, it, it can't finish it. Where does it do? It has to loop. And so there, there are a lot of complexities and even how you address the transaction or create a new transaction and where's the cost? Is it all part of this production? Certainly if it's in a production run, you wanna keep the cost there cuz you illuminate variants. But again, as you heard me articulate, these feedback loops are imperative, um, just to, to fix these problems. Uh, and again, there's a lot more going on, but I thought that was a good start.
Sam Gupta (06:50):
Yeah, that's definitely a good start. In fact, I'm going to touch on some of the comments that you made. And overall from the best practices perspective, I have seen all sort of crazy scenarios, uh, when it comes to rework. And I don't know if that is the right way of doing it. That's how they may have done it, just because they didn't probably have the rework processes supported as part of their e r P system. So they had to come up with creative ideas such as creating an operation, uh, as part of your bomb. That could be one possibility, right? Uh, the other possibility could be, I find when I look at many different businesses, if your product is not really des defined, uh, not designed, you are probably gonna have design but defined. Okay? If you are not necessarily productized from the manufacturing perspective, then sometimes it could be harder because you don't know whether this is going to be a real rework or is it going to be some sort of change order. I don't know if there is gonna be, so any sort of correlation. So, Chris, any
Chris Garrad (07:46):
Point, if you think about engineering change, and again, it depends on where the feedback comes from. So if your quality guy picks it up, typically it's gonna fail your own set. Internal tolerances, it comes back from a customer. You're right, there could be decision trees. Well, that's just the way it works. Okay? We get that out of the guys upstairs in the cloud a lot, don't we? <laugh> interesting rework. But, but again, you go back to what's the type of rework and how do I collect that cost? Because in the end of the day, rework has cost on it. And then we go back to the process as to which, how do I orchestrate it? Was I done with production? I recognize that at equality, I gotta rerun the whole batch. I gotta take the whole batch. And again, how do I do I open up that MO and add more operations and run everything through again? But I think each of the different scenarios requires the conversation
Sam Gupta (08:29):
Okay, amazing insight. Thank you so much Chris for that. Uh, Dave, I'm coming to you. So obviously you bring up very, very, very important and different perspective here. That is going to be more from the shop floor perspective. And one of the things that you do as part of your day job is finding efficiencies, uh, in your manufacturing process, right? So hopefully you can talk about the best practices from the OT perspective, because now when you think from the OT perspective, then you have to think about your machines. You have to think about off your machines, you have to think about production runs. And then finally, how do you control this process? Because once you have the rework that throws off everything, uh, unless you have a process figured out for rework, that is very rare in my experience, that is all manual in general. Mm-hmm. <affirmative>. Um, so j over to you.
Dave Griffith (09:11):
No, a absolutely. And I think that there have been a lot of really good points brought up by Chris and, and also you Sam. Uh, so as you said, kind of my, my background is mostly how do we actually go implement and execute this stuff on the floor? And, and I loved how Chris laid it out. Uh, typically that, that is not my experience. Uh, typically I guess when I look at, uh, when I look at rework from, from the shop floor, it's, it's always a bad word, right? So 10, 10, 12 years ago, I, I helped design an aircraft facility, right? 200,000 square feet. We were putting together drilling riveting, the back couple of pieces of a fuselage together, and we had to have a rework station, right? And we, we, so the conversation was, well, we have to have a rework station. So the rework station was the smallest little square, uh, amongst all of our squares, because that, that typically is not a conversation, that is a happy conversation to have because we know that we're losing money and it's costing us money and, and, and it's all of, of these other issues.
So I would say kind of to Chris's point, um, doing some of that root cause analysis is extremely important. W when I look at rework, you know, there, there's the opportunity that, hey, we made a mistake. Maybe, maybe we hit something, a tool hit something with this piece, and maybe we can go sand it out. Maybe we've gotta go scrap this piece. But if you find that this happens over and over and over again, you have to do some amount of root cause analysis of the process to understand, hey, this is why 50 out of 50 items had issues. And hopefully you can find that before it reaches your customer. Because there, there's nothing more that upsets your customer when you go send them 50 of the same part and they, they don't work because they're not to spec or they're not to a variety o of other, uh, o of other, uh, ways.
Um, I will also say that I've seen the opportunity to do a bunch of rework. So I was working with a, with a beverage company, uh, who will remain nameless for, for reasons fairly, fairly soon obvious to, uh, to understand, right? So, so that they were putting their, their beverage into cartons and th this one facility, 18 lines, I don't know, hundreds of millions of dollars over the course of the year if they had an issue with carton crushing, right? It doesn't pass quality inspection through the end of the line. If it didn't, they had no ability to rework, right? And so for that organization, rework can be as simple as taking it, kind of pouring it into a bunch of food safe, uh, 55 barrel drums, if you will, and then moving it back to, to the, the front side of the line, going and pouring 'em back into the tank and going to put it back into, going to put it back into, you know, different cartons.
And so that was a huge opportunity for that particular organization to put in a nearly, nearly no cost to them rework station to be able to go rework the, the items. And so that becomes near a hundred percent, uh, opportunity to recoup the profit, right? I think there was probably 60 cents, uh, 60 cents a carner thereabouts that was, you know, scrap. But we were able to rework the most expensive parts of the process in order to save them a bunch of money. And that all, that all came from kind of the root cause of, Hey, why are these getting poured out? These are very expensive. Uh, some beverages are, are difficult to, to get and to continue to, uh, to make anex expensive to make. So let, let's go through the process of understanding where the problem is and then going to solve that problem. So typically in my experience, that becomes a let's go solve the problem before we worry about maybe how we get it into the system, be it in e r p or an mes or kind of any a any of these other issues that a lot of the time becomes an engineering process that we have to go solve.
And then we've got the ability to go take those learnings, push them in as, as Chris had talked about, and hopefully we don't get to the point of a customer returns a hundred percent of a run because it, it wasn't a spec or there weren't, there were opportunities that we missed to go solve these problems.
Sam Gupta (12:51):
So that's very interesting. And, uh, I, I'm, I especially like your story to be honest. Okay, so when you are talking about, and I think you had a little layer there in terms of the expected rework and the unexpected rework, I guess, I don't know if that is the right term. Yeah. Uh, in terms of defining the rework, <laugh>, uh, but I don't even know if you can really configure your rework process as part of your quality processes. If you already know, uh, what could be the issues in your production line and because of which, uh, you might have your rework. So I don't know if you can configure that as part of your quality process, but in your story, uh, you sort of add that, you know, that you are getting these unexpected rework, but for some reason you don't know. So there are always going to be the variables that are going to drive whether it is a rework or not. Uh, but who sort of makes the decision in the rework process whether there is going to be a rework, uh, you know, it's going to be throwaway product, or how is that process? Dave, do you wanna touch on that?
Dave Griffith (13:48):
So, yes. So, so typically I've found if we've got the ability to rework it to, to sell it as a normal finished good, we'll absolutely rework it. Um, in, in some instances, be it like food and beverage or, or some consumer goods, we've got the opportunity to sell things as seconds. And th this is very factory dependent, so maybe not up to the normal quality, but we can recapture 70 or 80% of the, the, the typical cost by selling it as a second. So we don't lose the, the cost of all of the time and materials that, that we, that we put in. I would say when we talk about who is the correct person to kind of make that call, I think that that's super dependent. A lot of times it gets to quality, right? So there, there are a, or a number of quality checks along these processes.
And as we go through a number of these quality checks along these processes, it becomes, Hey, we failed. We send it to rework, or we send it to, to somewhere else. Um, and I think that the more often an organization has this style problem, the, the more kind of decision tree that they have internally as to what we do with all of these particular items, uh, what I find is most of the time, if it's a one or a two off, it becomes, maybe it gets set over to the side and maybe we think about doing it at the end of the run. Or I know some organizations that'll run 20 or 30% extra because they, they just expect there's a number to fail quality, which I should just point out is absolutely not what you should do. We shouldn't run 20 or 30% extra because we assume some of these are, that many are going to to fail quality, or our customers will buy more if we have more at the end.
Sam Gupta (15:20):
Okay. Amazing insights there. Thank you so much, uh, Dave for that. Uh, so Nara, I'm coming over to you and the space that you operate in, that's very interesting as well, uh, you know, make to order. Uh, typically in my experience, there is always going to be a very thin line, uh, in terms of what is a real service versus what is a product. And typically when you have really defined products, for example, bombs, uh, then you are probably gonna have easy life in defining, uh, the rework. And I don't know, uh, if when you talk about, because see in the make to order space, you might have the field service as well. And I don't know if field field service could, could result into the rework processes as well. There's going to be some sort of overlap there overall, uh, you know, in terms of how field service is going to result into rework processes. So I don't know if you wanna touch on that layer and whatever else, uh, you may have, uh, too often enough.
Nirav Sham (16:08):
Yeah, absolutely. I think you touched on something I was gonna touch on there. Uh, but I wanna start at saying, you know, rework is just doesn't, isn't only, you know, from manufacturing, right? Rework actually come from many different places. I think Chris pointed that out, you know, nicely earlier, uh, rework is where somebody has let something gone through and it didn't meet a specification specs, right? Engineering customer, right? It could be the machine machinery, right? The machinery failed and it didn't proves the right output on it. So really there has to be a lot of checks and balances, right? When it, when it comes to rework, because it's, it's sometimes you, you, you have to try to prevent it from, you know, measuring twice, cutting once, right? Sometimes hard. What happened? Machine fail. Machine fail, you measured twice, you, you, you loaded, you know, the coal into the machine, but the machine just failed.
What do you do about that? Right? Well, in, in situations like that, you have to be very careful, like, you know, was it rework caused by a breakdown in the process, right? Or was it uncontrollable, right? At the end of the day. So, you know, there's, there's a couple different, uh, avenues there. And then the other side of it is outside the manufacturing floor is the fuel service side. So fuel service, everything could have been great. Now, between the transportation, you know, installation at the customer site, somebody bent the wrong corner, right? Yeah, something happened, right? And now you gotta take this whole assembly, bring it back and rework it, right? There's, there's processes and there's steps and there's a lot of cost and rework, right? What I think what gets missed is what happens to your profit margin, right? Rework comes back and typically, you know, make to order.
They don't apply it back to a job. They'll try to go ahead and quickly, but reprocessed it real quick and send it right back out the door. And now everyone thinking we made a margin on this particular project, but we really did not at the end of the day, right? You could really go from being profit to all the way to being a loss on a particular job because, you know, the margins could shift that easily from rework perspective. You have to be able to control that. Um, you know, the other indirect impacts that rework could have, right? Is rescheduling on the floor, right? To have too much rework. You're throwing off the standard work that's going on on the capacity, right? On the floor. R p and nps. You know, you're not really running rework through r p and nps. You're just basically on the fly trying to throw in a production order at this particular place in the plant to go through.
So you can now what? Now what happened? Now all those orders that were scheduled to be there are automatically linked, right? So that throws off capacity, that throws off your whole planning process, right? We had a really good discussion on that last week. Um, that has to be carefully monitored. You can't just take rework and throw it back into a plan and expect that, you know, it's not gonna have any other impact. It has a huge impact on capacity, capacity planning. And what about handling, right? Material handling, you have restocking fees, potentially you're gonna decon deconstruct something. Maybe you salvage what you could salvage out of that assembly that you already used. Now there's additional cost, indirect cost for handling that material, be putting it back to stock, maybe bringing some other additional materials out, uh, onto, onto the floor, right? You have employee morale, you have frustration, right?
There's people that, that, that are doing this, right? Uh, you know, you, you continuously have rework issues, right? What's happening to, to everyone that has to deal with this, right? They're like, we don't wanna deal with this. This is, this is hard for us to work. We don't even know where the rework, where the process is feeling, right? Documentation is important. You gotta be able to provide pro proper quality procedures, right? So in order to not be, what I would say reactive and be proactive on, on, on rework is to have a good quality process, right? Start from the beginning. Where are you gonna measure? Where are you gonna make sure that you're gonna inspect, right? Don't inspect, if you know that there's a high tendency of rework in a specific part or a specific cell or a specific work center, try to inspect that sooner before it goes down to the next step and the next step.
And before you complete it, all of a sudden you know that there was a big failure upfront in the product, you know, up in the manufacturing process, right? Try to catch it sooner. Try to isolate where it's coming from, right? What you're doing is constant quality analysis to reduce your rework, right? And you want to prevent it by building some sort of structure around your rework processes, right? Is where does it start? Start documentation preparing for where rework could potentially occur in the process. You gotta measure right? When it does occur, why is it occurring? What is a root cause of it, right? Is Chicago, right? Analysis root, you know, all these different, you know, QC terms out there, total quality management to, you know, all that stuff and communicate, right? You gotta communicate between departments. A good quality system is where you have open communication. Um, you're communicating between departments, you're focusing on process improvement, right? If we, if we look at rework, it's a lot of it is preventable, but a lot of it is, you know, some of it is not. But even then, I think if you focus hard enough on it, I think you can try to, you know, kind of minimize, minimize as much as possible and become more kind of, uh, proactive instead of reactive on a day-to-day basis.
Sam Gupta (21:03):
Okay? So amazing insights there. And I'm actually making notes as you are talking and I have the is mc John isolate structure, uh, measure as well as communicate. So great job there. Uh, now the point I really wanna touch on your commentary is going to be related to the rework. Does not go through your M R P as well as, um, a p s process, right? Um, so now the people who might be sort of doing the M R P, but not doing the m p or maybe doing aps, but not doing aps, um, sometimes, you know, they are probably going to be doing the m p at least for the procurement, but when it goes to the shop floor, good luck with your scheduling. You know, whatever happens on the shop floor, only shop floor knows about it, right? Um, <laugh>. So, uh, you know, when you look at rework, right? Rework is definitely probably not gonna be part of that. So do you wanna touch a little bit in terms of if the people they might not be familiar with, you know, how the APS and M R P operates and how your rework sort of correlates with that? There are over to you again.
Nirav Sham (22:04):
Yeah, absolutely. I'm a big proponent of color coded type of documentation, okay? Right? Depending on what's being released to the floor, right? You have green, that means it's, you know, at the right schedule and it's being released. If there's a rework, have a red traveler out there, right? Saying, Hey, this is urgent potential, there's a rework. So everyone understands, right? That this has come in because of a specific reason that it's defective. Instead of just kind of someone tell, tell somebody back office, Hey, create a production order for me here and I'm gonna release and throw it into the floor. It should be no on the plan that this is a rework that's coming in. Is it coming in from a manufacturing defect? Is it a customer return order that's causing this, right? What's causing this rework? So there's all eyes on it. They're actually taking this document, they're making notes on it, trying to figure out how the re you know, how the defect happened in the first place, trying to identify exactly where the defect is.
And now you're using that as a quality measure to in the back office. Understand, okay, why did this happen? Right? Here are all our documentation. Here's our notes on this red traveler that we released out there, and now let's use that for our analysis. Let's take it to engineering. Let's real, let's kind of talk, talk through this. Now let's figure out, was there an issue with the bomb? Was there an issue with the routing, right? To make sure this doesn't happen again, instead of it just being kind of going through the shop as a regular production order. Cuz it's really not, at the end of the day, it really isn't right? That it starts there. You have to treat this as a very unique entity on the floor that where all eyes are gonna be on it in order to make the right outcome, uh, for this rework order, it doesn't happen again.
Sam Gupta (23:36):
Okay? Amazing insights there. Thank you so much Nara for that. So Chuck, I'm, uh, coming over to you and I think love made a little controversial comment there, to be honest. Okay? And that is going to be in the world paperless <laugh> in two, he's talking about paper. So I don't know Missy, everybody's gonna have probably different opinion. Uh, okay, so Chuck <laugh>, uh, in your case, what do you think? I mean, should we go paper paperless? Uh, and where does overall, uh, rework fits
Chuck Coxhead (24:00):
Regardless? I mean, I, I think we, we don't need to talk about the travel. You can certainly do identification in the absence of a traveler, but quarantine, quarantine, quarantine, <laugh>, okay? It needs to come out of the product flow and it needs to go into another product flow. Uh, we put it off to the side. Oh, we took it over there. Okay, now it comes back to the product workstation. Where is it in the product flow? Ah, we'll get to it. Well, what's the priority? Truth be told, an e r p world that ought to be injected in by some sort of rework work order that's give it that some sort, that give it that priority once it comes back in how you identify it. I mean, depending on the size of the product, there might be a different handling, uh, material handling container or something to that effect.
So quarantine, quarantine, quarantine, I think is the, is the issue there, regardless of whether your paper or paperless. So both can work. What comes to my mind in this? I mean, all the building on the other comments is interior, uh, inventory planning and engineering is a direct labor cost as opposed to an indirect labor cost. Okay? Depending on the scope of the work, typically you're gonna classify, you might classify those employees as indirect labor. And, and those hours are, you know, brought in that way. Now you're bringing him to a timekeeping function, okay? Depending on the failure mode analysis that they have to perform the testing that they have to perform different skill sets. You may not have technicians, manufacturing workers do that. Okay? You have that accounting you have to care for. And my really big concern in rework is inventory planning. It's super easy when you've got 50,000 parts in inventory and you've got a stake, uh, uh, uh, uh, safety stock and you're planning for that and you're using some percentage of those for rework. Not a problem. You consume faster, you cost it against rework jobs, end of problem. What happens when you've got one? What happens when you're make to order buys one part. You know how I know did it, did some C A C C machining earlier in my career. Great big stainless steel forging. This was really expensive and we bought fricking one and guess who messed it up? <laugh>.
Now that's an extreme example. The good news is, the good news of the story is that we were able to take it back to engineering, fix the mistake, and I remachine it and everything was good. But that's an extreme example of one. But when you look at your supply and your material planning, when you have very, very high inventory quantities, it's easy. When you're stocking things in small quantity, it's hard to plan for rework. And as was mentioned, that will heavily impact your mat, your production schedule as well. It's not just a labor issue, it's also material issue. You could have a 12 re 12 week replenishment, replenishment time and orders due in four weeks. Now what? So the question is, how do you plan for these things? And it requires some foresight. It requires really when you take a look at your requirements, say, okay, what is my real situation?
No, I don't wanna plan for 20% rework. That's really, really bad. I will tell you we've spoke in another earlier panel in a semiconductor business that actually might be a pretty valid thing because it's just simply not achievable, right? So you do plan for additional wafer material and time and planning to do that. But no, in general, we don't wanna do that. We want a hundred percent yield, a hundred percent output. But how do you plan your inventory if you're kidding yourself and you're not considering, what are your impact part the parts you're gonna be impacted, particularly the lower quantity stock parts, where are you gonna be? And we just gloss over that. We'll, we'll deal with it when we get there. No folks, there's customer at the other end. You've gotta do that in your inventory planning.
Sam Gupta (27:37):
So some very interesting insights there. So number one, I mean, inventory planning. I'm glad that you brought it up because obviously that's where the real plays of rework because uh, that is going to throw off the entire equation. Uh, overall from that. So you did mention some strategy there in terms of how to plan. Obviously you need to have foresight, but I don't know what you are going to do when you are gonna have Dave's situation, which is going to be just unexpected, uh, <laugh>. And in some cases we could have, uh, you know, a lot of things that are going to go unexpected. But the point that I really want to touch is going to be related to your direct versus indirect labor. Mm-hmm. <affirmative>, and I don't know how many people are really going to be familiar with this whole direct versus indirect, and how many people do the costing appropriately to be honest, okay? Mm-hmm. <affirmative>, uh, if you are going to be doing your scheduling on paper, then, you know, scheduling and costing both are probably going to be all over the place in my mind. Mm-hmm. <affirmative>, uh, so check over to you. Any, any follow up comment based on my commentary?
Chuck Coxhead (28:35):
Well, I, I think the, the, the issue is how do you classify those workers? Is your system lock you into the, you know, this engineer is indirect labor, this engineers direct labor. After that, we just, oh, well just send you over near, I'll just have Dave go over or I'll have Cheryl go over there, she'll retest it and it's gonna be no problem. And we'll just, but are we actually accounting for that? Does that actually, you know, to Chris's point earlier, does that actually get captured in a, in a a rework work order? We're actually costing that. Okay? And capturing that variance from engineering. If we just sync engineering costs and everything, we burden the entire product line. But in truth, one of the reasons we affect continuous improvement is by recognizing the variances. Where are we having these cost variances? And if we just never capture that labor, we spread it over the product line, we don't know that a product or a product line is causing a particular problem.
So what do we do? We have to capture that labor up until the point mean every manufacturing situation is different. Sometimes the, the cost of capturing the labor exceeds the amount of labor. Okay? If you're, if, if you're, if that cost of data entry is a cost that, okay, forget everything I just said. Yeah. All right. Spread it across your entire product line. But most of the time that's not the case. Make sure that you're capturing that direct labor because it can really go crazy. I mean, I've seen failure mode analyses that just go on for hours and hours and hours. At what point, what's your decision? You know, what's your decision process? Chuck it over your shoulder and move on. Right? You know, how many hours am I gonna plan for in this work order? How many am I schedule? Okay, it's gone more than an hour.
Do I stop? Okay. And that's part of your planning. What is your routing for that change order? What is your default routing? Are you do have a check-in, you know, of red flag that goes off when you've exceeded the planned number of hours so that someone can make a decision? You know? And then from a people management standpoint, do you empower the people on the floor to make that decision? Okay. Or in an hour and a half, move this one on and then, you know, it goes back to production and you feed back through e r p, the need for, you know, you scrap it and then that sends a signal back for somebody to create a new manufacturing work order. Um, you know, that's, you know, so there's a people aspect of that as well.
Sam Gupta (30:50):
Yeah. Could not agree more. There is always people aspect in everything that you do related to e R P. Okay. <laugh>,
Chuck Coxhead (30:56):
Thank you so much Absolutely. To that,
Sam Gupta (30:57):
Um, Chris coming over to you, um, you know, the next round that we agreed to, um, any stories, any comments over comments, Chris?
Chris Garrad (31:05):
Sure. And you know, and, and I love Chuck's comment that this one's needs rework, move it to quarantine. Okay, now I'm building an aircraft, okay, <laugh>, I'm building one and I got a hanger and I got a problem where an airframe, and again, if you think about rework and, and so it, again, in the scenarios we're implementing manufacturing, we have projects over manufacturing. Cuz an aircraft is not one bomb, right? It's lots of bombs and it's a project with a work breakdown structure and a lot of different to, to configure and build. So there's a much more complex structure for manufacturing and airframe, for example, it's projects, hand bombs and manufacturing. And so as you think about, you know, the project level task, it's, again, it's back to services again. And I think it's very interesting as you think about when do we trigger change management?
And, and, and depending on the manufacturing landscape, right? You know, if I'm, if I'm building a part, I'm selling it out the door through Amazon at, you know, $2 and 25 cents, you know, you're, you're eating the change management, frankly, because you've already set a retail price and the expectation is engineer is clean, still gotta cost it, and Rob's right. You, you, you, we, we lose a dollar on everyone, but we do volume. Sorry, that does not work. You will give away all your profitability and you'll find out 90 days later that you lost a lot of money. And I think that because we talk about the, the lost resources, people aggregate, they congregate, they talk. It's not captured. Nobody's clocking that stuff. It's not like I'm in a structure where I have a project in addition to manufacturing where, hey, if I've got r and d, I've got an r and d project.
Well I got, so again, that whole costing thing is, is really tricky there. But again, in the context of big, big operational manufacturing, again, rework could be triggered again, when you're building an aircraft, okay? Is it, is it engineering related to a specific particular run? Is it designed that'll unpack the 13 of these I'm making over the next 24 months, for example. But as you think about triggering a process around an in, in process, I was gonna say in-flight, but airframe, and again, I'm, what, what's my problem? Hey, the wing assembly doesn't fit. What's wrong? Okay. You know, oh, it's temperature, it's, it, it, it, it's got cold, it contracted oh, oh oh. But again, you think about the guys that are on the shuttles and some of that kind of stuff where again, they got a, a unit one and there's, there's no quarantine, but it is a rework has a whole different meaning and large complex manufacturing scenarios versus small.
So I think I heard something, throw it over your shoulder, right? Penny wise and pound foolish. It goes a few different ways. So I've had people spend weeks reconciling accounting to find a few pennies. It's like, <laugh>, make an adjustment. You just cost me $10,000 to find a nickel. So again, there's an 80 20 rule, and I think you gotta rationalize that, yes, you want the engineering correct, you need, you need the process in place to capture, to measure. People go, oh, we don't do quality. You should have quality. Everybody should have quality just to avoid downstream implications. So again, there's a lot of best practices that you can glean from what you're hearing from everybody in this room, in this con whatever, this virtual room. Because it, it can save you a lot of money if you're thoughtful about it. And again, you put in some good processes to capture, just to anticipate you hope you don't have it. Rework in our industry is a dirty word. It really is. Cuz you lose, you know, and who owns rework? And again, is it because of something unanticipated? Meaning a change in scope, a change in circumstances, weather, logistics. Hey, we're putting together an, it, it 18,000 feet elevation, okay, that wasn't in the plan. You get it. But lots of things that could cause rework.
Sam Gupta (34:18):
So very interesting points there. And uh, guys, I mean, since we have a smaller group, if you guys have any sort of stories, and, uh, if you wanna chime in, please, uh, feel free. Um, Dave, do you have anything you unmuted yourself?
Dave Griffith (34:30):
Yeah, no, no. I, I was actually going to say, Sam, I, I, to, to Chris's point, have a story of an air airframe, uh, that required rework, right? So, um, I, I, at one point in my career spent some time working for a company that will remain nameless, that helped, uh, kind of build these, these aircraft and, uh, we, we built the machines that helped build the aircraft. And so typically every Friday the operators didn't wanna work overtime or over the weekends. And so we'd have issues, right? That that was a very frequent occurrence. Well, one Friday we got an issue that an operator whose name luckily I don't know, crashed this machine into an almost done, uh, almost done air aircraft. Um, and the, the cost, right? The, the, the, the cost was somewhere north of a million dollars. I never heard the, the actual number.
All I heard is that it was somewhere north of a million dollars. And so the immediacy was, well, who's gonna pay for it? Because the manufacturer didn't wanna ma pay for it cuz there's like, oh, the machine shouldn't be able to crash into it despite the fact that we didn't train our, our operators. Um, a as, as we go onto that, the, the, to what I heard, it became a, a multi-million dollar write off to the end user buying the aircraft at a, you know, significantly lower price in, in order to keep the aircraft and keep their, their place in line and, and a, a rework or a repair done that was as acceptable to anyone as as possible a a, uh, for a, for a brand new aircraft. So, uh, Chris, Chris had brought that up and, uh, there are very few times in which I get to tell that story and I think it's been a, a more than a decade, uh, since that happened.
So I, I can probably, uh, tell that story to, uh, to, to the point, uh, to the point that I did. Uh, but, but if, if I may continue, Sam, I, I like, I like the, I like the comment of paper versus paperless. That paperless, that nirav started. And for a person who spent a number of years trying to convince, uh, shop floors to go paperless, I, I can tell you that my, my new opinion as of about two or three years ago is we should use as little paper as humanly possible, right? Um, I, I, we've done a lot of work on digital travelers, right? So there is absolutely value to capture that information digitally. Um, I worked with an organization that every production run they had, someone had to go print out 26 sheets of paper. Of those 26 sheets, they were all supposed to be filled out of the ones that were all supposed to be filled out, typically about four were.
So we, we immediately know what is, uh, what is the most important, uh, there, there was one very important one that the production lead had to go type it up, um, into the Excel spreadsheet and you're like, oh, okay, then gonna go send it over. No, they printed it off as a PDF and physically handed the piece of paper to someone else. So, um, th there is a, there there is a, there is a wide gap between, uh, paper and paperless, but we, we should absolutely use as little paper a as possible capture as much digitally as, uh, as possible as, as we go through there. And, and I really like the, uh, the standard, uh, kind of accounting, uh, the, the cost accounting that, that we were talking about. I think it, it's really important. I think we kind of hit, I, I typically look like I, is it a a standard issue that we have or is it a special case?
Is it a v is it a different variance, right? Like, is this something that we allot 20% of these things to happen, or one in a hundred to happen that then maybe it goes and gets costed to that particular run. But if it's a special cost and we go do a, we go do a FEMA and a bunch of analysis and we go try to be better as an organization, I don't know how anyone could reasonably cost that to a particular run. Cuz if we go cost $20,000 of cost to a $5 part, there's no way we'll be competitive. The the next time we go to well, we go to look to, uh, to build that, uh, in, into the future. And, uh, and, and if I may finish with, with one, one other story, I, I have lots of rework stories, um, in, in case anyone missed that, uh, lot, lots of, uh, lots of ones, at least I don't, uh, I'm not kept up, uh, awake at night, um, most of the time.
So, so I, I did some work with, with a company that's kind of within the, the semi-conductor, uh, realm. It is one of the, the worst quality organizations that I've ever seen. And so that was kind of the initial request that they, they're kind of, they're throughput kind of based upon expected was something, well they said it was about 5%, but you, you go and you do a little bit of research and you realize they only get about 25% throughput of each of their runs because they, they continue to steal, right? So, so they go run acts and by the time it's done that they have a sub, uh, sub appropriate amount of, of final output. And so, so they're just stealing from, from, uh, earlier orders, which is really bad. I imagine it would cause anyone in the e r P side to absolutely completely lose their mind because we're we're just stealing from one order to the other. Um, and, and my favorite part about that organization is when you ask them how long something took and they, they have, they've got two timeframes. Sam one is a new order. So if you have a new order that you want to come in, they're gonna quote you 20, uh, 20 weeks. If you have rework or a reorder, they can do it in three weeks.
Sam Gupta (39:27):
<laugh>, very interesting stories. Uh, Dave, thank you so much. And, uh, check, uh, semiconductor. I don't know if you're gonna have any follow up comments there. You always have a lot of stories there.
Chuck Coxhead (39:37):
Oh, well, it, i, I think that the reason one of the worst quality organizations is because semiconductors is a, it's a relatively low yield, uh, production situation. And so it is one of those things that since you cannot possibly achieve 100% in theory, right? It's, it's asymptotic, will you ever reach a hundred percent in manufacturing semiconductors? I guess it depends on the situation. Some are worse than others. Um, by design it's a, it, it's a difficult quality organization because you're literally writing it off. This is just gonna happen, it's just our life and it's just, uh, you know, that's the way it is. Well, most of the quality people that I know, they're rather particular picky yoon and, you know, they're those guys and girls that are gonna make sure that they're keeping everybody in check because they're extraordinarily detail oriented. Yeah, that is counter con, uh, you know, contradicting to the semiconductor quality model, which don't get me wrong, I mean, these were amazing people I worked with and, and, and, and I'm not impugning anyone, but there becomes a little shift in, okay, acceptable quality level, yeah, acceptable loss, tighten it back to e R p.
I mean, if you are not using that data and capturing that data and feeding it back into your system, failure's always traceable to the feedback loop. Okay? If your yield loss is changing over time and you're not feeding that back into bill's material as a possible feed loss field, uh, excuse me, yield loss, boom, material planning's out the window again, okay? Labor planning's out the window again cuz it's not just lost peace parts, it's also lost and spent labor. So
Sam Gupta (41:23):
Yeah, great point. Thank you so much. Uh, so go ahead Chris, please. Yeah, we
Chris Garrad (41:26):
Had a, we had a field service project. So the big deal, they're a major equipment manufacturer and so the field service teams totally different systems, right? Yeah. But we needed the field service data to get it to manufacturing quickly, quickly. I mean, this is big, big, millions and millions of dollars in warranty. So that was the other we didn't even talk about is when it leaves the building and there's, and there's a problem with it, it's warranty work. So if you think about cost, I mean, again, I think we go back to where's the cost impact. If we can fix the problem quickly and, and in proximity to the beginning of the process, it's great. But when you're way downstream at the end of the process, cost goes up and, and the problem is to truck's anybody's point is the data. You don't get the data back.
And they're like, they don't know. They keep making these broken widgets and they're, Hey, why are we fig? Wait, we got increasing hire more service people, more dynamics, licensing. No, it's not the right answer <laugh> anyway, <laugh>, but it is, it is. If you, if you're listening and you're thinking about, you're right, the feedback loop, I always tell people the engineering, I got a box, it's got an input and output, it has a loop that comes back and says, adjust it, refine the output. That little simple model with a box and lines in a little loop is, is applicable to everything you do in life. You gotta take feedback and put it back in. If it goes, it drops. It's like the airplane toilet. If it goes no good, no good at that point, you need to get it.
Sam Gupta (42:46):
Okay, amazing insights there. Thank you so much Chris for that. So off, uh, you know, any stories, comments over comments since you have not spoken so far? So <laugh>,
Nirav Sham (42:53):
<laugh>, no, I love the examples. I think there's a lot of context there when we talk about rework. But you know, going back to Chris's example, now you know what happens after the sale and there's an issue. Yes, it's warranty, but is it rework or is it a repair? Yep, there's a difference. There is a two difference there, right? And that's a whole different type of analysis, right? You can't consider everything rework, essentially rework is when you're building it back to compliance, right? Where repair, you're making a minor adjustments for it to function the same way potentially, right? There's a higher risk involved there versus a rework, right? Rework, you're redoing the whole thing potentially. But
Chris Garrad (43:30):
If they all have to be repaired
Nirav Sham (43:33):
<laugh>, right? And, and so, right, that is a clear dis distinction, right? You, you get a, you get a tire, right? The tire doesn't have enough air in it, you're gonna rework that and you're gonna put enough air in it first as you got the tire, you sold it, you know, there's a puncture, you're gonna patch it. You're not reworking that whole tire, right? So there's a difference, right? It is, you have to clearly understand at one point, is this considered a rework? At what point is it considered a repair, right? And what are you doing with this data at the end of the day in your e r P system, right? What decisions are you making and what are you striving to do? Are you looking to be ISO 9,001 certified, right? All this, if it's properly documented, properly presented, and you have corrective action measures for anything that could potentially happen, right?
It's a continuous loop that that needs to be implemented, right? If something, something's gonna happen. But what should you do to fix it and repair it, right? That's the important part about this, right? You could talk all long about repair reworks, right? How often does it happen? What are the yield factors? All this stuff, but where is this gonna affect? How are you gonna take this information in your e r P system, turn around and say, if this happens, this is our corrective steps, right? That's where the value is at the end of the day.
Dave Griffith (44:49):
I, I just want to ask the question, I think Nirav brings up a bunch of good points is at some point between warranty or rework and or repair or extended service, where's the liabilities? Is it the li are the liabilities on the initial, you know, process And what we did is the liabilities on the end user who has done something to break this to, to cause the need for repair? Where are the liabilities when we go there? And that goes back to kind of the, the costing and accounting within the system that I'm sure causes a whole bunch of people to, uh, to pull even more hair out as we try to figure this out.
Chuck Coxhead (45:24):
<laugh>, well, the liability when it gets out into the field ist always on the customer. If that were the case, then we wouldn't have lot traceability in part traceability to go back and find that there's a failure that did make it out to the field and then we can go back with a recall process. So not a hundred percent, but clearly your point is, you know, valid. Absolutely.
Nirav Sham (45:41):
Sam Gupta (45:42):
Okay. So very interesting point. Go ahead and Rob, you had a comment. Go ahead.
Nirav Sham (45:46):
Yeah, no, I mean, you know, just, just, just, you know, when you talk about the rework process, you have to look at it from the customer standpoint and internal standpoint. Sure. Um, and exactly what does that mean and how do you correct it?
Sam Gupta (45:57):
Okay. Very interesting points guys. So I definitely wanna touch one point that Chuck made. The liability is going to be on the customer, right? Uh, so I don't know if customers are always going to be paying for these things. I don't know if that is what you are implying. I don't know how cu how many customers are going to be ready for that. And in fact, let me see, there is gonna be a transfer of ownership, uh, argument that I don't think we have discussed, especially when you are going to be talking about the product leaving your dock, then it goes to your carrier, then there is gonna be a little insurance play there. So I don't know who really takes the hit. So anybody who wanna touch on this, um, anyone from the room?
Dave Griffith (46:35):
Chuck Coxhead (46:36):
I'll touch on it a little bit. I mean, there's other reasons why things can happen. We we're talking about conformance and compliance to specifications as Niro said before, but now you're gonna get into, you know, touch on Inco terms here, right? At what point do you release Yeah. Reliability to, you know, or the liability to the customer. I is it from who packs it? Okay. <laugh> Freight on board X works fca, what is that? Uh, I mean that's where, and that's why honestly so many small companies, they just use F O B because that's what everybody uses and they're thinking, you know, they're thinking less than load, you know, uh, flatbed truck or something that's some, ah, everybody just uses F O b is that really true in a, in a UPS world? Okay. Exactly. Perhaps not. It might not be the best choice. Um, but you know, the liability is really comes back to that warranty issue as well.
You have to understand what's the reason for it, and that's where the liability really comes into the case. Was it damaged once it left my hands? Okay. Liability, potentially third party shipper. That's why you have claims, right? Um, was it because it never functioned right in the first place? Did something burn up because there was a short that came down because there was a small bur in a part, and that developed over time as that burr moved. Manufacturing beef defect liability falls back on me. And that's why we pay engineers, right? So we can understand these things and do these failure mode analyses and you know, and then of course there's the sales side, all right? And that is, well, you know what? We've decided to absorb that as a cost of doing business. Okay? So there's an artificial liability that's sometimes assumed by the organization. So liability, Hmm, it's a big fat, depends.
Sam Gupta (48:21):
Yeah. Could not agree more. Uh, great points there, son. Rob, I'm actually going to give back the question to you. And that is going to be, uh, I know that you had offered some examples related to your repair and rework, but I don't know how many people really understand and where do you draw the line between your repair and rework and what are going to be the financial implications, uh, between your, uh, repair and rework? And I don't even know how many, uh, p systems are going to support these process, uh, differently. You know, some may not have any of those <laugh>.
Nirav Sham (48:50):
Yeah, no, you, you, you're, you're exactly right, right? It's, it's those, those organizations that are actually looking to start investigating, you know, this, this, this, these high cost, right? Manufacturing costs. You know, exactly where are they coming from? Um, you know, is it all related to rework, true rework that it, through our manufacturing line, we were able to identify before releasing to the customer that, you know, we have to rework something, right? Essentially, versus a sales return order right now, a sales return order. There's other process supporting process outside of just the material. It's the validation, right? That you can return something, right? It's the series of asking questions by the sales representative. You know, what is going on with the product, you know, how is it not functioning based on how it should function on the specifications, right? Capturing that detail and that data.
So when it comes in, right, you have something to go off of and start reviewing, okay? Is this not, not automatically, assuming it's a rework, comes our process. Is this a repair? Because the way it was designed didn't really fit the way it should have functioned because a customer used it a different way than we initially thought right now, right? You have to take that information, it's an additional cost, but you're getting market feedback from that, right? You're now adjusting it, fine tuning it, especially make to order environments, right? You're, you know, you make something, but now you don't realize that there's, there's a low ceiling there, someone forgot to, you know, measure it and this thing doesn't fit properly and sometimes jam it in there and it, it bet all these different jaws on there, whatever it is, right? So now when you bring that in, yeah, it's kind of a repair.
You made it to the specifications that was initially there, but what failed in that process, the design team, when they went out there and looked at the space, when they looked at how they should design this thing, didn't take that into account. Well now let's put that into our questionnaire. So now we go out to a similar plan in the future, or we're trying to sell this in the future. We know exactly how to measure this so when we do, it doesn't go through our production line and everyone's happy, but we didn't do something here, right? It's that feedback loop again, that, okay, we understood what happened, we did our due diligence, now it came in, we repaired it, right? Uh, maybe you repair it or you scrapped it, right? You could scrap it and remake it at the end of the day, right? You scrap it, there's some cost there, scrap it against a project, remake it, issue another production order, and now send it out there.
Now it's actually gonna be something that fits. Now with service, you could actually swap out the bill of material. You could swap out that component. Say initially we had this in history, now you could say that. Now this is a new component that's in there because of so-and-so reasons, right? You're compiling all that history against that specific, uh, bill of material that's actually deployed to the, to, to the site now. And, uh, you know, keep that for archive and, and you can make sense out of it, right? At the end of the day, you gotta make sense outta your e r p data, right? What would, what, you know, what happened? What's the postmortem around this repair, around this rep repair, uh, replacement around this rework, right? And someone should be able to analyze it. And if somebody should, you should have a task maybe every six months. Let's analyze our reworks, right? Let's analyze, you know, which parts are going through there, right? A lot of times these reworks go through and no one is looking at it. No one is gonna make heads, you know, go back and look at that one production or the house for rework or repair. No one's looking at it, right? They just know at the time there was repair of rework or what good does that do? Is this data flowing through the system without any action?
Sam Gupta (52:09):
Okay? Amazing insight. Thank you so much, Nirav. Uh, any other comments, guys? Open floor, um, anybody? No, uh,
Nirav Sham (52:19):
I got, I got thing right? Uh, when implementing quality, right? It's important to have a connected quality system in your E R P, right? And the reason is, as you're going through quality, you could quickly create re rework production orders, right? Directly through quality link those production orders back to the quality records in your E R p, so you could get that audit trail, right? If you're taking quality right now and that's outside your E R P and you're filing that away somewhere, that's never gonna be connected back to that production order, right? You've got to look at that analysis. Keep, try to keep quality inside your E R P as much as possible. I know a lot of times quality gets implemented, second phase, third phase, right? But you know, you can start rejecting next day you go live, right? Consider that a phase one as part of your implementation cuz you're gonna, you're gonna get, uh, you know, some good data out of it.
Sam Gupta (53:08):
Agree. Go ahead Chris. Yeah, so
Chris Garrad (53:10):
He, he hit some good points though, you know, in terms of the quality piece of the, of the solution here. But not everybody has those features. And you go back to, oh, is there a we a rework type transaction or is there an RMA repair? So even in some of the legacy products, you know, I had a depot management, well, great, that was an RMA that came in to disposition. Do I repair it, generate a work order, send it back out, do I back the inventory it, do I scrap back to the vendor? Yeah, not everybody has that functionality. And I think one of the, one of the comments would be is how do you augment the process in your current system without that capability? Meaning what do you do? How do you capture it? What's your, what's your I provisional approach to handling this stuff?
And I think it, it still, we talked about a lot of things like where's it discovered, you know, I hear the example, okay, is it a one-off or is it a more systemic problem that needs to go through engineering? Change management. Guess what? A lot of systems don't have an ecm. So, uh, again, I think that's is the audience. If there's, you know, group of people listening is you gotta look at your e r P system and look at the processes it supports. And if it doesn't directly support it, you need to come up with one, right? Whether that's, uh, hey, I'm gonna use a sales order or I don't know. I don't know what the right answer is because, you know, I'm, I'm lucky that we've, we've got some different options there. But the point is, don't expect to have a perfect transaction to handle this.
And again, the example where we said add, add an operation to the manufacturing, that means you catch it there and you're doing it there. But if it's every part going out the door, it's a bigger problem. If it's a one-off because of something that happened in the process, just that one time, little different scenario, but you've got so many different scenarios. But as I would indicate, you're gonna find, and I've got, you know, an SMB product that I don't know that it has all those transaction types and you're like, where do I put it? Excel? I don't know. That's a bigger conversation as you think about how do you capture this thing and you're like, do I have a GL account? That's, that's, but again, to Niro's point, you, somebody's gotta be looking at variants and if you don't capture data and a lot of manufacturers back flush, they back flush labor and material.
We use more material, we use more labor. Nobody knows where. Do you know, how do you know the wages don't change? The standard consumed into the manufacturing doesn't change. Well, what happened? Well, they made less pieces this day. Why was that? But the point is, where does the variant show up? And what you just heard from me is you may not see it. Labor looks good, parts looks good. Oh, they use more stuff. Did they issue that? Is inventory off? You find a variance at the end of the month doing a physical inventory, you go in and find, right? And so again, the systems that aren't really set up to handle it really challenge yourself as to how to capture that. And you may have an offline, an off-book process, but the goal is that the transaction volume of these fixes is so small and you want it, you want it to be a difficult process, so you fix the process problem. Okay? So there's my advice.
Sam Gupta (55:53):
Okay, interesting. Thank you so much Chris, uh, for that. So, uh, closing advice? Anymore closing advice, Chris, or are we good? Good. Okay. Uh, Dave, closing advice please.
Dave Griffith (56:02):
Absolutely. Just to jump on the quality train for a moment, I would say please don't do human-based visual quality inspection. There is, it's only marginally better than just shipping everything out without looking. Absolutely. Go ahead and close the, uh, clo close the loop, pull your data together, be able to have those conversations and understand what your, your larger problems have. Uh, to, to the rework and the rework best practices. No one likes rework, as we've all said. To some extent it's always a bad term, but in the imperfect world that we live in, rework is a thing that happens and we need to find ways to, to take that, capture that information, make our process better, and make the organization better moving forward.
Sam Gupta (56:40):
Could not agree more. Thank you so much, uh, Dave for that. Uh, <inaudible> closing advice, please.
Nirav Sham (56:45):
Yeah, measure twice, cut ones have keep, keep looking and uh, and, uh, you know, take, take, take it seriously. Uh, rework happens. Take it seriously and don't just kind of brush it under the rug cuz there's dollars there and you don't wanna, those dollars are hiding, but you, you know, take them out and um, they'll, they'll improve your bottom line.
Sam Gupta (57:06):
Okay, amazing. Thank you so much Nara for that. Chuck. Closing advice please.
Chuck Coxhead (57:09):
Yes, I wanna talk about opportunities. The best way to deal with rework is to prevent it in the first place. Okay? And you've got a couple of opportunities. One, we talk about machines going down. Does your quality, does your e r P system have at a quality module that helps you to address pm stay on target, you know, get your PM work orders, get a schedule out there, follow up with that. Does it have it? If it does, that'll help you. Another is, you know, training your people. You know, there are opportunities nowadays to integrate training platforms that you know in with your E R P system that's amazing. You know, you, and it, it helps to keep peop get people trained and keep people trained in a way that works for them, you know, now and in a more real time manner. And lastly, sometimes it happens.
Okay? And now what's one of the big cost drivers in get getting your rework. Somebody's gotta go dig up drawings, somebody's gotta go dig up documentation. It takes extra time and effort to do that document presentation integrated with your E R P, right? You can use it not only at the production, but you could also use it at rework. And they may need more documentation than is, than they typically will have in a, you know, in a regular production run. So document presentation at the point of the work using these fancy computers that we've all invested in can be an amazing time saver, which can reduce your variance and reduce your costs for rework. So you've got some opportunity to do things better and improve your process by design.
Sam Gupta (58:30):
Could not agree more. Thank you so much Chuck for that. So that's it for today. Uh, and if you join for the first time, this was part of our digital transformation series for which we meet every Thursday at 5:30 PM Eastern. So make sure you are going to be there in the new year because we are not gonna be here for next, uh, four or five weeks. Uh, we'll come back, uh, on in ten third week. Um, so happy holidays and merry Christmas everybody. I'll see you in the next year
Chuck Coxhead (58:56):
And happy, happy holiday everybody. Take care. Thanks,
Nirav Sham (58:58):
Everybody. Care. Bye-bye bye.