Transcription
So let's start with the agenda. We're going to cover six things today. First, lean cost reduction basics and why it's so much more. Then, practical tips to find where and how to start reducing costs. Next, actionable tips to engage your people in the fight, getting help so that it's not just you or a single person doing this. Then, 17 practical steps to reduce your labor and materials costs. Finally, a case study I did with an injection mold company. We'll get to this; don't worry. Before we get to the case study, I want to give you a bit of background.
Senpi, our business, was formed in 2007 to help manufacturers who are serious about reducing their costs or improving their businesses. We work with very small to very large manufacturers; our client roster includes £2 million to £30 billion businesses. Fundamentally, we transfer TR lean capability to hidden heroes in manufacturing. I'm a co-founder; my experience is with operations and lean manufacturing across four continents. I've worked with, trained, and coached about a thousand people in 100 companies. We generally do courses, coaching, and consultancy, and we have a number of things we do (you can see some of them here). I'll keep this punchy; don't worry. We work from shop floor to boardroom, all levels throughout the business. We have an award-winning lean app that we use to help people improve their capability, set up two global lean initiatives for billion-dollar businesses, and that's a picture of me in Japan with a team that was voted best factory after five years of lean transformation, out of 170 factories around the world. I'm an awards judge and a conference speaker. There's my Sensei from Toyota there, who trained on the Toono line—a very savage man, but passes on a lot of knowledge—and I write a few books.
Right, so I want to start with lean and the background of lean, going very quickly through this because I think we have to go back to roots on cost reduction, as it can be misunderstood. On this graph here, money is the vertical axis, and time is the horizontal axis. In the Cost Plus environment years ago, you added up all the costs of your product—your materials, your labor, your overheads—you banged on top your profit, and you said to the market, "Take it or leave it," that's the asking price. As we know, things changed. We're all in a cost-down environment, and the price is set not by us, or by our quotation or our salespeople; it's set by the market now, and every day of the week and twice on Sunday, it's going to be below what we want them to pay, which gives us a choice: a palatable choice and an unpalatable choice. The unpalatable one is to reduce our profits, and we can't do that because our shareholders will cry, and we can't reinvest in the business. So if we maintain our profits, we really have only one choice: to reduce costs.
I appreciate that that may be preaching to the choir for some of you, but I firmly believe that if you're in manufacturing, everybody in your business should be able to sketch this out on the back of a napkin, because this is the fundamental reason why we have to pursue a lean path. But it's not the only reason; it's necessary, but it's insufficient. There is, of course, the lead time part of the equation, and the fact that yes, it's about cost reduction, but it's also about lead time reduction, and the fact that whatever you make, somebody else makes something similar, and if they're able to offer half your lead time—two weeks rather than four weeks—they're likely to take your lunch. But beyond that, cost reduction gives you the base so that you can think about flowing value better. My favorite definition of lean is the ability to flow value to the customer using fewer resources, and that really connects us to the customer. But the start point is cost reduction, which is where we are today.
Now I'm going to spend the next three slides—and there's a bit of an example on here as well—talking about where and how to target in terms of cost reduction. The first is about the top level; the second is about what everybody in the business should be able to do and know and help you with; and the third level is operational cost reduction. At a top level, my interest is in the four big loss drivers for any manufacturing business: number one, revenue loss from unhappy customers (some of it you'll know about because they just won't renew contracts; some of it you won't know about; they'll just go elsewhere); number two, dealing with internal quality issues (quality drives a ludicrous amount of cost within businesses); number three, cash tied up as inventory; and number four, the poor use of capital (capitalizing kit, machinery, investing, having overcapacity on some which is sat idle, or monuments of capacity where you have large increments that aren't flexible if demand changes), and then there's the constant seeping loss of changeover for your productivity, for your OEE, part of your operation. Those four big loss drivers speak to something really important in manufacturing. People—whether Ops managers, Ops directors, or CEOs—there needs to be a mindset shift. The mindset is: yes, it's okay to think like a production, like a professional manager, and have an ROI approach to the business, but also you need to think like an owner and have an ROA (return on assets) versus return on investment.
The top one there, your professional manager will be thinking, "I've been bought into this business; I've got all sorts of problems; what sexy new technology or kit can I invest in to help me become more productive?" The bottom one there, the owner—and I'm encouraging you to think more like an owner—will be thinking, "Okay, I've already sunk a lot of money into this business and resources; am I getting the best for my buck out of it? Where can I reduce waste, look for issues, and improve the performance on the assets I currently have?" Now that's critical because if you're in a position of authority, you're looking to improve your performance. I would urge you to be very, very tight on the justification for capital and expect a full understanding of the losses to the decimal point of the current kit you have. So that's top level, strategic level, if you like. All levels of the business, in my experience, need to understand the three types of cost, and we mentioned these a little bit ago. The question is: does everybody know? We know that your costs are likely to be made up of materials, labor, and overhead. The question is: does everybody in your business know this—from the quality manager to the finance manager to HR to your operators to your team leader? Let's take a team leader example here. This person's going to need to know two things: firstly, what are the percentage splits between materials, labor, and overhead for your given products in your given market? Because then they'll always have in the front of their mind what they can and need to affect most to improve stuff. If it's heavy on materials, then they should be always thinking about materials; same for labor. But then they should be thinking—and you can help them to understand—what can I control?
So let's take our team leader here. On the materials front, he can control how much he's ordering to keep the line stocked, how much inventory he's holding between stages on the line, how much buffer he's got, and how much scrap he's producing. On the labor front, he can be thinking—and he can affect—possibly how many people he has online, definitely how productive they are (whether we're getting back from breaks or not and we're hitting hourly targets), but also things like are we working overtime? On overheads, the team leader can start to think about things like oil consumption (where am I losing oil?), the amount of glasses and gloves I'm handing out, the small things, turning off fans and lights at the end of the shift—the small things that they can affect. Those three levels of cost give you a good idea of how to target and where to target throughout the business. Sorry, that's the first two levels. The third level is the operational level, and this is where I didn't start with SQCP because I wanted to think strategically first, and then I want you to think in terms of your cost makeup for your products, but operationally, particularly quality, delivery, and cost—some kind of pie and Pareto analysis. So this is a defect pie chart showing that 20% of what we make is defects, and then we know that our number one line for creating the most defects in the factory, and then we also know our number one product for creating defects or scrap on our number one line. That level of 80/20 thought and understanding where our specific biggest drivers are is useful because time is scarce to improve once you know that, and once you have a return on assets mindset, it then comes down to: can you see? Are you able to go to the point in the shop where you've identified it? And a quick recap on the seven wastes, and I want to show you an example of the power of finding small wastes and stacking them up before I do that.
So you watch anybody working, and you'll see three types of work: value-added stuff that transforms the nature, shape, or characteristics of what someone's making or doing—the product (they might be welding two parts together, they might be assembling two bits together, they might be painting something); then there's non-value-added (this is stuff that has to be done currently but doesn't actually add value—things like loading and unloading parts into jigs because parts can't fly, things like changeover of a machine from part A to part B because you can sell both but you need to change the machine, and things like inspection); but then if it isn't value-added strictly speaking, changing the product, and if it isn't something that you absolutely have to do, then it has to be waste, and the definition of waste is two words: everything else. And that takes you—if you've forgotten, or you don't know, or you don't consider it—that takes you back to considering the seven wastes, which is a bedrock of being able to see in the shop. Now let me give you an example of that. When I set up JCB's Academy some years ago, we wanted to set up a lean academy that was actually a training academy plus had a real factory part to it so that people could apply what they learned in a real environment. So we had to pick a product, and this was a VI—an arm for a vibratory compactor. And this here is Jim; I videoed him doing a 19-minute assembly process. This is before; this is when we first found this product. In the first two minutes, I found 24 wastes, and I just want to play you the first 40 seconds or so and just give you a feel for the kind of wastes I found and then show you where it led.
So here he is wrapping up a plastic bag to put a piston in. That piston he's now going to put in the vice after he's wrapped it like that. He hasn't got any kind of jig to just slot it into; there's a quality risk of damaging that piston, as well as the time it takes. He's got to tighten that vice up. He's now gone hunting for his drill, which is more precious. And he's about to get his attachment; reach over there, he could have left that on. He's turning around; that's a waste to go and attach his airline. He's got to yank it out, and he's about to do an operation here that's interesting; he's actually about to change the insert inside this piston, which there's an argument—and we actually stopped doing it in the end—so that's a processing waste. And actually, as he's doing it, he's having to glue it back in with Loctite; Loctite it back in. If his Loctite works—in this case his Loctite is clogged—so now he's got to go and get hold of a pin and he's got to unclog his Loctite. And if you look down to the right-hand side while it was playing there, you'd have seen there was a pallet he could have tripped over. And right in front of him there's a coffee cup—mind you—that has a piping he's going to use in a bit with oil in that he's softening up now. All of these are pretty small wastes, but within the 19 minutes I found 91 wastes with Jim. And with the team leader, we categorized the short-term low cost, short-term high cost, etc., and we managed to reduce them, to eliminate them, change it around. And when we brought it into the academy, this operation, we got a 63% reduction in cycle time. My point being: don't sniff at what we've all learned years ago and we continue to practice—or not practice—and that's the importance of going and seeing and finding the waste on the shop floor, which takes us on to the next bit: getting help. In that, normally that kind of activity is put in the hands of industrial engineers or the lean team. A fundamentally important thing in reducing your costs is engaging. If you have 100 people, 100 brains, rather than just one. One of the biggest mistakes I see is relying on specialists alone.
Let's take this approach. Imagine, as a business, you decide strategically you want to go north, there, in the direction of the gray arrow; that's the direction you want to go. Now imagine that you have people in the factory actually who are trying to improve things, but they're improving things in different directions; they're just trying to do some good wherever they see bad. You're going to get some improvement, but your overall motion is going to be slower than if you had everybody pointing the same way. My point being: it's not enough just to get people finding kaizen, even if you can get 100 or 10 people doing it; you need to give them a direction to go in. There's the direction there; you'll move faster, which takes us to the importance of the leader, which is creating or clarifying the crisis to rally around. So cost reduction—a crisis. Why should people get involved? Why should they feel the need to help? Years ago, my father gave me a book—35 years ago—*Zen in the Martial Arts*. That's the actual battered cover of it; lots of stories in it. This one particularly gives us the reason why that's important. A Zen master walking with a student…oh, hang on, I've got a question. Okay, thank you. Quite right, quite right, Eduardo. A Zen master walking with a student, and they see a fox chasing a rabbit. The master says, "According to the fable, the rabbit will get away." The student said, "No, the fox is faster." But the rabbit will get away, insists the master. "Student, why are you so certain?" "Because the fox is running for his dinner, and the rabbit is running for his life." I love that phrase. If people believe that you, as a collective in the business, are running for your life, they are likely to run a better, a better qualifying time; they will give a bit more if there's purpose. And I'm going to come on to three or four nice examples of how to create purpose and frame the crisis in a minute before I get on to the cost savings. This takes us on to Hoshin Kanri, sometimes called policy deployment—a strategy to improve. It's basically: you're busy enough trying to run the business; there's very little time to kaizen or improve the business, so whatever time you have, best be focused on the vital few things. And Hoshin Kanri can make sure you have the vital few things—unless you overcomplicate it. And in manufacturing across the world, we've overcomplicated it left, right, and center.
What I mean by that is there's a Hoshin Kanri X matrix, cross matrix, that's used. It can be valuable if you're a mature business; you're used to using it, fine; if it floats your boat. For a lot of businesses, it's too complicated, too unwieldy, even if you have it on one piece of paper, like this business where they've got…it's nice and colorful; they've got safety, quality, delivery, cost, growth, and manufacturing or management systems on there. I had a long-running debate with the MD of this business saying, "Look, you've got 51 things on this sheet; you're thinking that more is more. If I keep putting things up there, if we throw mud at the wall and half of it sticks, we'll get half of it done." Well, actually, mine—and most of the lean profession's—experience is that less is more. The real skill in Hoshin is deselection: choose the vital few—three, four, five things to focus on that people can get some depth on and really make some progress. That's the skill. And when you do it, have something for the head—a logical reason: we need to improve quality by X or Y—but also have something for the heart; we need to connect those two things so people want to go after it for a reason. And that reason quite often starts with WIIFM: What's in it for me? Particularly, and I'll give you a Chinese example in a minute of a factory I was in where you can connect the pain of the shop to the pain of the leadership and the business. By the way, if you want a simpler Hoshin Kanri process, go on a website, and you'll find it there under lean resources.
Before I get into the 17 steps and the case study, I want to go through simple as best again, and I want to give you some examples of setting the cost reduction task, getting the troops rallied around it. This factory is an injection molding factory with some other processes as well—probably four factories in one, actually. The point here is that bit of paper on the floor, which is actually a label. The fundamental issue was that there was a bank of injection mold machines—up to 50, actually—and then there were nine assembly lines that they were feeding, but there was a disconnect. The molders molded what they wanted to a push schedule, and the assembly lines took what they were taking to a customer manifest, and the intermediate stock was out of control—too big and too much of some, too little of others. We could have set the lean team and the Ops team a target to by-stream map it, then look for where you can do a future state, and you can put pull systems in, but actually we didn't. The question that we decided to ask collectively, including the MD, was: "What if we had no paper on the floor? What would that look like?" And sent the team away to look at that and think about that in a catch-ball style. Now what that does is it leads them to the same place, which is a place of…right, to do that we need to have some kind of Kanban in place, some kind of pull system, but then we need to connect both parts of the factory, and it leads to the same point anyway, in a broadly similar fashion. Same factory, you'll recognize, same trolleys. They had a problem with quality where lots of little tabs were being snapped off of these molded cases. Tabs are important for securing them. You could put a Pareto pie chart and Pareto in front of the molding people and say, "Can you improve that?" Or you can say to them, "How are you going to get to the point where there are no clips on the floor?" And that sets them running again; it stops you specifying and dictating what to do and asks the question. Another example: this is my Sensei, Moran. This was in a China mobile phone factory. He spent three days with me walking the factory and with a team in different places looking at pallet trucks and trying to improve pallet trucks. It took me a while to figure out why before I asked him; I didn't want to ask him too early. The reason being: when he first went in, he started, as he always does, in finished goods, and he saw in finished goods that there were pallets of finished goods—all the value-added material and labor—that had fallen over in the warehouse. Now that's a waste. He then noticed that the pallets were a bit battered, and he noticed that the pallet trucks weren't in good condition; they weren't responding well; they also were a bit battered. So his decision was to get everybody in the factory looking at pallet trucks, to get quality sensitivity throughout the levels. He got the pallet trucks; he got them assessing them; he got them in, giving them a 5S home; we did some basic PM, TPM on them, and everybody started looking after pallet trucks so that they wouldn't damage pallets, so that we wouldn't damage parts. And that's also how he started to teach them about 5S and TPM. He went in a different way rather than telling them to do lean tools. Final one: Plymouth. I spent the best part of an hour putting this together and trying to persuade the MD of a medical business—medical syringes business—that if he carried on as he was, he was going to run out of space because their capacity was going to run out…oh, one sec, because their capacity was going to run out. Now he understood it intellectually; his eyes didn't come alight; I didn't see a fire in him when I was explaining that: okay, you have current state; you're getting by now; extrapolated current state, you'll have tears; and in future state, hey, here's a smarter way of doing it. But when I took him for a walk in the factory and we started—and this is the molding area—there's a door there; you can't see it; it's behind those; there's a door that leads to the warehouse. Whenever they molded anything, 90% of the time they were taking the bag, the pallet of molded parts into the warehouse and storing it there because, once again, there was a disconnect between the molding and what was being needed for the final stage of the process. As soon as I told him that every time you send a pallet through that door you failed—you personally have failed because you're just building inventory and you'll have to build a bigger warehouse—he got it; the penny dropped. So framing as a story really helps.
Now I'm going to move on to the 17 steps, but before I do, I'm going to pause; I'm going to stop sharing and I'm going to ask if there's any questions. In fact, no, I'll pause it there. If you have a question, please either put your hand up, or put something in the chat, or by all means come off mute. Okay, there is a question. Okay, no problem. Thank you. Yeah, thank you, Alardo. So no questions? Okay, I'll crack on now onto the 17 steps. Now, before we do, a word about your finance team. If your finance team are happy and comfortable to look in the rearview mirror, to give you reports and data—lagging metrics—and to tell you how things went the last week or month, please expect more…I've worked in a couple of businesses where a really strong finance team are engaged in operations; they want to get out there; they want to see what's happening, and they can give really, really sound advice on where to focus, what to focus on—apex stuff like that. So don't forget the finance team; get them engaged if you can. Now the 17 steps. So we're back to materials and labor, two of the biggest cost areas. Here's the eight ideas for labor, and here's the nine ideas for materials. Fear not; I'm not going to go through those individually on this slide; that's there so that you can just see where they all are. What I am going to do is spend the next three or four minutes going through these and to give you an idea of what I mean by them; these will help you to save money.
On labor cost, the first one: imagine you're a pigeon and you're flying in the rafters of your factory; look down and look at where the density of labor is in your factory, and then start to focus there, particularly if labor cost is a large proportion of your overall costs. While you're up there, take a look at the amount of space that is given over to storage—inventory tied up as cash. Also, do you have any assembly lines—two of them, line by side—where you've worked out you need 5.3 people running one and 6.3 running the other, by calculation that would make 11.6, but you have to man up at the moment with 13 across the two because you can't have three of a person, but you can have five people on one line, six on the other, and one shuttling between the two. If you arrange the work well, put some little C-bands in, or some little sub-assemblies, because not one of a person is still one person. Do you have machine minders looking after each machine as a habit, or as problems that you've forgotten to solve, or you can't get round to? Anytime you have a machine minder, there is a very expensive cost there, and quite often it's because it's been accepted—abnormal has become normal. Are you using operator waiting time while the machine is cycling? Can you use it for something else, something close by—another part of the process? And those two, three, and four I'm going to use in the case study in a minute. Have you standardized, or at least sized, the line feeding work in your factory and ensure that forklifts don't ride around the warehouse and factory with empty forks? Forklift trucks, as well as being the thing most likely to hurt you in a factory, they are also a huge cost. Anytime you see empty forks, think…Eddie Stobart lorry, or a transport lorry, will always be down in its own mares; it won't go anywhere unless it has to drop off unless it has a pickup. Very often walking around a factory—and I encourage you to look at this—forklift truck drivers, because they haven't been asked to do it a different way, will act like a taxi and not a bus—i.e., they're going out looking for a fare rather than a bus going around on a route picking up fares, which would be less costly. Which takes us into a deeper understanding of standardized work. You probably have some experience in type one standardized work where somebody is doing an operation—first, second, third, fourth, fifth element—and then they start from the beginning, making something. There is something called type three standardized work, which is for long-cycle indirect work where you can look at what a team leader or a setter or a materials feeder does, and you can find waste in that. Indirect labor cost is sometimes hidden. And have you standardized the regular planned downtime elements of the day—your startup time, your 5S and TPM time? This is a commissive It Card system I put in for a business across the factory where the small chunks of time needed to do things like 5S and TPM are standardized for a certain time and a certain length. And consider ways of using short periods of unplanned downtime to complete non-value-added tasks. So you can have a rack where it says: if we have a five-minute breakdown or a 10-minute changeover, please do this, this, and this. This one I would encourage you to do today or tomorrow: shift changeover, or the start of shift. Look at the shift changeover discipline. So if you have people drifting into their cell or drifting into the factory, bear in mind that if everybody starts at three when they should, or at two when they should, that's a lot more cost-sensitive than if 100 people start one minute late; that's 100 minutes you've lost—an hour and 40 minutes of productive time.
Moving on to material cost: how well are you procuring material, and are you making the most of your negotiating power? We've always got to start there. How's your negotiating? Look at old suppliers you've been with for a long time on old contracts. And then the yield and defects. Now I'll come to that bin in a minute. I'd encourage you to do a Pareto of your high-cost material scrap types by quantity and cost—yield and scrap. You should make any scrap bins you have somewhere prominent in the factory, on a gangway, facing outwards, no lid, not covered, and they should be small to fill quickly, especially if you have expensive, exotic materials. If your material cost is high, and your engineer supervisor needs some basic problem-solving skills…is your make/buy in good shape? Are you so vertically integrated as a business that you're doing some very basic things that just tie people up? Sometimes it's not worth doing it for you, but if you do outsource, remember the effects on contribution to fixed cost and profit. Number four: how well are your materials planners bringing in material only when required, not before? Have they got used to bringing in stuff early because they don't want to get a shoeing, they don't want to get told off? Is there some noise in your production planning information flow problems, or are there production issues in certain parts of the factory that force frequent replanning—replanning multiple times a day? Have we got used to running with too much fat in goods inward stock? Again, Excel is your friend here. And then finally on these: how quickly you get material through the business with lean is about compressing the timeline as well as cost reduction. So where is inventory and cash therefore stagnating? A rough-cut value stream map helps here. How smart are you in ensuring that incoming deliveries are optimized? This one I often see a lot of potential—use Excel again. Bought-out part numbers, multiply the cost per unit by the current inventory, add two columns on the end for the supplier location and the frequency of the inbound delivery, and find those suppliers who are 10 miles away but are delivering fortnightly, so that you're having to hold the inventory. How are you not communicating well with your suppliers? Where are you giving interface waste to your supply chain—either batching or you're just sending schedules late? Ask them where you're not helping. And finally, look at stock losses by part number, by value—overs, unders from stock takes—to show possible BOM errors. There's nothing less interesting than a BOM error, but they drive a huge amount of waste throughout the business.
Now onto the case study. I'll pause this for a second. Has anybody got any they want to ask me about or challenge? Okay, I'll keep going. So this case study was for a factory—a factory in the Midlands. Their focus was on labor and scrap cost saving, and it's actually mentioned in my book, *Vent in England*. Small clients, £6 million pound turnover—every penny counted at this point in their growth. Revenue growth and cost control to survive was their focus. They had a need to stabilize quality to get repeat business because they were letting themselves down; they weren't getting repeat business because their quality…and we saved £50,000 in one hour and 17 minutes, and here is how we did it. Now the case study I'm going to show you is based off the email I sent to the MD that afternoon, so I'm going to dissect that. Here's the thing: anybody on this call today could have done this; it's basic industrial engineering—seeing, not just looking—and a bit of attention to detail, being willing to stand in the same spot for five, 10 minutes and look. The labor cost things I mentioned earlier—this is two, three, and four of those combined. So here's the email I sent him afterwards—bullet point summary. I spent today looking at a mold shop, quick-hit productivity to support labor cost reduction. I was there for an hour and a half, and they were running six, eight machines, or seven machines. I focused on four machines—three, four, six, and seven—and this is a reason why I focused on those four machines. Let's start with machines three and four. I looked at them, and a quick look showed that they were being man-marked—so one operator on each machine with a lot of idle time in between machine cycles—semi-auto operation. Each person had about 21 seconds of cyclical work plus two seconds of non-cyclical work. Non-cyclical work is work they have to do every so often, like fetching empties or taking full pallets away or quality checks. Where you time how long it takes to do that thing and then divide it by how the frequency with which they do it. So 23 seconds in total. I set up a trial with a team leader to run both machines with one person walking between them for an hour, and we worked out to get exactly the same output with one person as with two. That's a saving of 600 plus per week by just not man-marking machines. However, at the moment the boxes he uses to put the finished products in would need to be made up by the laborer. My calculation—again, a bit of industrial engineering here—is that it'll be half an hour per shift for the labor, and with some work on the bench size and orientation, they could maybe squeeze us into the operator's work as well without flogging James there to death. Turning the machine around would have been ideal, so that loading side with a loading side, but not possible at this point—too expensive—and I pointed out they need to write an SOP for it. If leg A is an issue because of the walking—that's why we tried it for an hour—then they could rotate every half shift. And here's a video afterwards just showing briefly. So this is the first machine; take the part out, spark it off again, then add a bit of packing; this is a dry verge that goes on the side of your house; clip off the sprue and then walk around to the next machine. It's a walk, but it's not that big a walk, and it's a lesser of two wastes, and he arrives just in time for the second part without rushing.
Now on to machine six. Machine six: noticed the material was clear; a lot of scrap around the machine when I looked at it suggested a quality issue. Looked at the cycle time—46 seconds—and he only had 25 seconds of work content, but there were lots of spare parts hanging around, which set alarm bells ringing. I counted with him that he had 10 scrap out of 160 by 9:00—that's a 6% scrap rate for damage to one hand. It was a one-plus-one tool; observing it, it looked to be after ejection, and either the part was dropping on other parts that were already there or it was bouncing up and hitting the casting of the machine. Because the team leader made a shoot…I nipped out to Homebase, to a DIY store, bought a net, tensioned it, put it under there; we very nearly eliminated it there and then, but then they eliminate it with tensioning it a bit more—more experiments—and that one saved 314 per run per week, 5,000 a run at a time. And this next slide just shows you the math—set the maths I put there because we have to know what the true costs are. In this case, it's the parts lost, but also the extra four-hour run for the matching hand blanked off, labor cost, etc. So machine seven—final machine here—two people sat down; looked like a simple job; I know foam has a knack to it; I couldn't see any real quality issues, and they were keeping up okay. It looked here like somebody who'd been trained and had got to the training standard, but there were still two people on there, and they were concerned about a persistent scrap issue. So my point was: get your best tech focused on that, and then you can speed the machine back up to 40 seconds from 48 and you get a 17% productivity boost. So that all added up to £50,000 in one hour and 17 minutes. Now almost done now. In this session, we've covered six things—five of them so far. What I want to do is just take a couple of minutes before I do Q&A, excuse me, and tell you about the trickier bit behind this—the bit I left them with. So actually being able to do all that does take some training and some skill. The trickier bit was leaving the four next most useful things that that factory could do. And they were: having defect bins at the end of each shift in a home; making sure that the team leader and