#21: Why your MRP is insufficient in the real world of S&OP and what to do about it

Søren Hammer Pedersen:

Hello everybody. Warm welcome to this S&OP masterclass from Roima. My name is Soren Hammer Pedersen. On a daily basis, I work with supply chain planning and I’ll be your host for this podcast here today. The purpose of these S&OP masterclasses is that we dive into trending interesting topics within supply chain planning, find out what is happening within areas, give you our perspective on the matter, and hopefully give you something that you can use in your daily life.

 

And today’s topic is no different. We are talking about the MRP. So we are going a bit nerdy. So my first question to you would be, do you trust the output you get on a daily basis from your MRP in your company or do you find yourself with a lot of unrealistic plan and endless firefighting on a daily basis? That is what we are dialing into.

Why is the MRP so difficult to handle in many companies and what can you do about it from a supply chain planning perspective? Hope you enjoy this. And you’re in luck, as always, I have invited my good friend and colleague Benjamin Obling into the studio to help shed a light on this area. Welcome, Benjamin.

 

Benjamin Obling:

Thank you.

 

Søren Hammer Pedersen:

As always, we need the Benjamin introduction just so people know. Give us a few points on who is Benjamin.

 

Benjamin Obling:

Yeah. So very briefly, Benjamin Obling, been working with the Prita IVP product for the last 16 years working in Roima. Various different projects, but always in the inventory planning, demand planning, supply planning, and the full integrated business planning.

 

Søren Hammer Pedersen:

Perfect. Thank you. You’ve been good at that introduction from all these times now.

 

Benjamin Obling:

We’re brief, let’s get to the topic. Yes.

 

Søren Hammer Pedersen:

Exactly. Exactly. So first question, right out of the bat, Benjamin, why is this something that we see within what we do is highly relevant and highly interesting for people?

 

Benjamin Obling:

Yeah, because the MRP being a very, very, very nerdy, you could say element of the full supply chain planning and so on, but it’s just super critical. Obviously, that is where we plan or most companies almost all plan their purchases, the production, semi-finished product through the supply chain, both in terms of multiple factories, in terms of within the factory, the bill of material, et cetera. That’s how we convert the forecast, the inventory planning, et cetera, the inventory balancing. We convert that into real actions, real life actions. And that’s really where we need to have a grip around the MRP to make sure that everything we agree on in the S&OP meetings and so on, that they actually come to life that is in through the MRP in almost all companies.

 

Søren Hammer Pedersen:

Yeah. So check mark. Done. That sounds good. Good input. Then we just run the MRP. Why is this a problem then for many of the companies that we talk to?

 

Benjamin Obling:

Yeah. If you have a more simple setup and you have something that sells fairly frequently, you’ll set a safety stock. If that is good, if your forecast is pretty good, your lead times are correct and your suppliers are actually meeting the lead times. Maybe you don’t even have a production you just purchase. Then it can be very simple and easy and it will work. The problem is when it gets very complicated, when things starts to change a lot, when we have variations in lead time, we have variations in demand. We have variations in MOQs. We have detailed production scheduling going on and so on. Then we have a lot of different elements and we have that at play and we have that throughout the bill of material and maybe even throughout the sourcing across different factories or even across different companies or ERP systems within the same client.

And then at the end, understanding why do I need to purchase or produce the amount that is proposed becomes very, very complicated to understand. And it also changes quite a lot. We have quite big differences in the output from the MRP, which is very hard to explain.

 

Søren Hammer Pedersen:

So how do companies see if they have this issue? What are the tell signs that they would see in their planning or in their organization if this is something that they should pay attention to?

 

Benjamin Obling:

Yeah, you could say one is, of course, when you have the production scheduling, if the proposals that comes out of the MRP are pretty useless in the sense that they’re not detailed enough, they don’t have the right sequence, they’re not obeying to the capacity that you have, et cetera. So it’s basically unrealistic proposals. That would be one red flag. It could also be the purchases from your suppliers. If you’re not taking the holiday into account, so if you have Chinese New Year or you have summer holiday in Europe, et cetera, you’re not taking that into account. You’re not pre-loading your suppliers with purchase proposals because you’re still expecting the MRP would be expecting that you can just purchase that during the holiday, for example. Or it doesn’t tell you how to fill up a container, et cetera. Then those would be the signs that you’re spending a lot of time then probably on the manual corrections of the MRP.

And another element could be down the supplier side that the supply forecast that you’re giving your suppliers are changing dramatically every time. And your suppliers would get back to you or your planner, purchasers or planners are getting back to you and saying, ”The plans are changing constantly.” Don’t you have a grip of that?

 

Søren Hammer Pedersen:

Yeah.

 

Benjamin Obling:

And then you would go back to your SNLP material and you would see, yeah, the total demand actually looks pretty stable on total level, but because your MRP holds a lot of different rules and MOQs and lead times and different sourcings and so on, then everything changes dramatically at the end, you could say all the way upstream when you purchase or produce.

 

Søren Hammer Pedersen:

And I guess I think that’s a lot of very good, mostly internal indicators, but I also think if you don’t have all that in place, you’re just trying to get to Friday and stay alive basically within your organization. At a later point, you will hear from sales and you will hear from customers. So guys, the end result of this, if you don’t get it under control is less sales and bad service levels and things like that.

 

Benjamin Obling:

Absolutely. The business impact would be lost sales or too high inventories because you need to purchase a bit beforehand or you have large air freight builds coming in, et cetera. So rush planning and production, you have too many changeovers in production because you need to change your production constantly because you don’t have that calm output.

 

Søren Hammer Pedersen:

Okay. So let’s agree this is important. At least we think though, and of course also some of the dialogues we have with many companies, some of the pain points that we start with comes from this imbalance, you can say, and how do we work with that? So let’s dive into how to work with it in the sense that maybe start by talking a bit about also understanding the issue. What is the MRP designed for and what is it being asked to do basically that it’s not designed for? What’s your take on that?

 

Benjamin Obling:

Yeah, yeah, you could say the MRP is, you could say, as opposed to AI driven algorithms, then here we actually have a completely rule-based setup. So it will look at, okay, what is your forecast? What do you have on inventory, et cetera? So you can say designed to a deterministic world where we have a certain demand forecast, we have safety stocks, et cetera. It’s in most cases, not designed to handle all the different constraints that you’ll see in production and also in purchasing. For example, this like Chinese New Year, you need to front load something or if you have a supplier which has a certain constrained capacity as well. It’s normally not designed to that unless you have a very advanced MRP. You can then build that in using IBP tools, for example, or in some cases you can build it into your ERP system. One of the problems with building it into the ERP system is then you don’t have the output, you don’t have the, you say the impact of changes before you actually load it.

So you don’t have the simulation capabilities, which is pretty nice to have because otherwise you’re just uploading a new forecast, a new safety stocks, et cetera, and then you’ll just run with it and then you hope for the best that it will turn out and it will start to create purchase productions and stock transfer orders and production orders and so on.

 

Søren Hammer Pedersen:

But you still, I guess, have the black box element of the MRP here, even though it’s not AI yet, maybe, in the sense that, at least in my experience, what-

 

Søren Hammer Pedersen:

... in the sense that, at least in my experience, what companies struggle with is that the MRP has developed over time and we are not aware of exactly what you’re saying that what it’s made for, what it’s not made for. We had some problem, let’s say three years ago, we customized something and handled that problem and now nobody knows why it’s doing what it’s doing because we can’t remember anymore.

 

Benjamin Obling:

Yeah. So you could say actually it shouldn’t be a black box because it is deterministic, it’s rule based and so on. You can say you can recalculate everything if you have the time for it. And that’s probably the problem here. We don’t have the time for it and we don’t have the transparency around why it’s doing what it’s doing, and we don’t recall the different rules. You can customize that a lot in the RP systems and so on. And all those customizations are of course made for a good reason at the time. But the problem is that it ends up being a lot of different rules, which then means that the MRP will get, we call it like nervous.

It will do changes and you would see your proposals, both production and purchase proposals would be jumping up and down in size. It would back and forth in time. And the more rules you put around it, that could be things like the MOQ or you have a certain capacity where you move it back and forth, if it has below that capacity, it can only start on certain days, et cetera. And all of that is correct, but that ends up then being, when I look at the capacity for July, for example, next July, okay, then it’s completely different than what I saw just two days ago, for example, when we had the S&OP meeting. And that is sort of the risk by putting in these more advanced constraints, that it will jump up and down quite a lot.

 

Søren Hammer Pedersen:

Yeah. But I think it’s, at least to my perception, it is a big issue out there because I know, of course, there’s a general tendency that in the ERP development that’s out there, that, okay, let’s try to keep it as standard as possible to avoid all of these customization. But looking at the landscape now, many companies will have what you call a nervous MRP at the moment developed over time still.

 

Benjamin Obling:

Yeah. And one of the interesting ways that you can analyze it is if you start to take time locks, or let’s say locks or timestamps, of your different MRP runs and then try to compare that week by week, month by month, or even day by day sometimes, and then you will see. So we take all of your proposals out and you do that week by week and then do that for two months, for example. Then put it all together and then look at how does that change?

And you’ll see it’s all over the place, jumping up and down, back and forth and so on. And you could say for the short term, and that’s one of the important things here, for the short term planning, you can say the production plannings proposals right now or the purchases, so if all of the rules we have set up are good rules. So for example, the Chinese New Year, okay, then we are pre-loading that before so we get the goods that we need, then that’s of course, super important and that can be very good for the short term. But when you look at the long term planning, both in terms of your supplier forecast and your capacity, then it’s probably not the best forecast that you’ll have for your capacity utilization because you’ll see huge problems in July, for example, and then you run it next Monday and then some things have changed throughout, let’s say a bill of material with seven layers and four different factories and so on. And then July is not a problem at all. You actually have idle capacity. You have a huge problem in August or something like that.

So you can say, I think, one element is there to only impose the complexity if you need it and then only use it on the shorter term planning, and on the longer term planning, don’t use the MRP or at least have it as a reference instead because it will create false alerts, you could say, on something that you cannot ... It’s like trying to detail, should I come in on Tuesday at eight or nine if I’m in production, seven months from now, and which production should I start with? We don’t know, and it’s going to change seven times before we get there. So don’t try to solve a problem that we don’t have yet.

 

Søren Hammer Pedersen:

Yeah. And also a perspective of that is, of course, the question of should we have the alerts, shouldn’t we have them before they hit the MRP so we actually can move things around or ...?

 

Benjamin Obling:

Yeah.

 

Søren Hammer Pedersen:

So it’s also a timing issue. And that leads me to the next, because I think, of course, we established what to look for here in terms of you have this problem and there will be a lot of telltale signs for sure, but maybe also give a bit of perspective of why it is that these MRPs struggle in the real life production environments out there. And maybe mostly the things we feed into the MRP here, is it that’s causing all the parameters, all that, is that the main problems or is it more what we have done with all the customizations down in the belly of the beast?

 

Benjamin Obling:

Yeah. And that obviously depends completely on the company and how many customizations and so on. But you could say in many cases, actually what we see is that if you fix the forecast, if you fix the safety stocks, the MOQs, the lead times, et cetera, for example, throughout the chain, then you would improve the proposals dramatically in the MRP. And then you could say the MRP logic as it is, might actually work. You would probably like to see a simulation before you put it into a ERP system so you know what is going to happen, but fixing the input so it’s basically garbage in, garbage out, obviously. So certainly fix that first. And then you could take the next step and then look at the different rules you have when you’re grouping proposals, when you have capacity constraints, et cetera. So all of these.

And the MOQs, do you need this large MOQs, for example, which will also cause some nervousness in the proposals. Remove as much as you can, but of course the ones that are important and make sense, keep them, but then note and make sure that you know that that means that your MRP on the mid and long term is going to be impacted by all of these different constraints. And we basically don’t know yet how it’s going to look in July. We don’t have a clue yet. So let’s not try to fix a problem that we don’t have, but certainly fix the inputs first.

 

Søren Hammer Pedersen:

Yeah. And not to throw anyone under the bus here, but what you just said there, if I said that to some of the companies, new companies that we talked to, I think one of the first thing that will pop up is, yeah, but our master data quality is not in a state where we can do anything about it. Bad excuse or valid point?

 

Benjamin Obling:

Yeah. The master data is always the problem. And then you can say in the quest for automization that we want, we want to remove as much of the manual work as possible. Getting right master data is absolutely, absolutely key in this because otherwise you can’t automate. If your MOQs, your lead times, et cetera, if they are wrong, it’s going to be dramatically wrong if we just put an AI agent to put out the proposals, et cetera. And our approach there is to say, instead of running a large master data project and then you start to fix your demand plan, so your forecast and your safety stocks, et cetera, instead, let’s start to fix the master data, but let’s do it in a way, so we chop it up and then we realize that we are not going to solve all, we’re not going to have a crisp master data set ever.

That’s a hard ... Okay, it’s not going to be solved. So, okay, but how do we then get by? Because we don’t want to spend that much money on fixing it and it doesn’t pay off. The business cases isn’t there. So instead, let’s look at how many of the master data here are we actually using? Which ones are super important? Okay, MOQ, lead time, and so on, obvious ones. Okay, let’s map out all the ones that we have are super important. Step two, let’s look at, okay, how are we then actually performing? So are they likely to be correct or not correct? Here we can use AI algorithms to make predictions. What should the master data be? Okay, then we have the large discrepancy. Okay, now we have come from a list of let’s say 8,000 different components down to 1,000 where we know it’s wrong.

Okay, next step is still 1,000. We don’t want to fix that. It’s still too much. Okay, let’s look at the ones that we are going to use in the coming period. So let’s say the next month, for example, where do we actually have proposals? So where are we going to use the MOQ if that’s the one we’re looking at, or the lead time? Okay, let’s start to fix those. Now we’re then down to say 200. Okay, final check is then let’s sort it by the biggest dollar impact. So what is the biggest, again, if it’s-

 

Benjamin Obling:

It’s the dollar impact, so what is the biggest... Again, if it’s o-ring, something inexpensive, let’s not bother. It’s a tail. We’ll probably buy 200 too many. Who cares? It doesn’t cost anything. But the ones that are really expensive, let’s start by fixing those, and then we have maybe, I don’t know, 30 or something master data field that we need to solve. We actually have a prediction of what it should be, then we chop it up and we do it prioritized.

 

Søren Hammer Pedersen:

Yeah. Okay.

 

Benjamin Obling:

Because we’ll never fix the full tail. Forget it.

 

Søren Hammer Pedersen:

No, of course. I think the interesting point, of course, here is we talked about some of the issues with this, but of course, the most interesting is how to at least work in direction of fixing this in a good way. If it was easy, let’s be clear about that, then we wouldn’t have this discussion because people have done it already, so there is a lot of work coming to this, but one thing I think could be very interesting to talk a bit more about is how to analyze this situation and gain the understanding of what is actually hurting your MRP runs at the moment. Could you talk us through the steps in how would you go about analyzing the current situation and how we can improve it?

 

Benjamin Obling:

Yeah, yeah. Yeah. You could say we touched a bit upon it previously, because I think first step is really to lock and then see how much does it change, so start to lock the results of the MRP and then look at it. You could say, how much does this change? You basically have a period zero and then you have the proposals into the future on the different factories, and different item numbers, and so on, and then you have the periods out and you have the number of proposals, because then you can really see when you put, let’s say, 10 different locks on top of each other, then you could see how much is it actually moving back and forth.

Also, if you then lock what is the level of the BOM that we’re looking at, this could be finished good product, this could be at our decentral warehouse, it could be a central warehouse, this could be across ERP systems, so let’s look at the full chain. How much does that actually change? You’ll probably see that that doesn’t change very much because the forecast, if you have a good forecast, that probably also doesn’t change that much. It does change a bit, of course, and of course, on the lumby and medium ones, it will change more. But then, when you start to get into the next levels of the production, the semi-finished, the simply and so on, and the components, then you’ll start to see level one, two, three, four, five, or three, four, five, then you’ll see the nervousness will really start and then you can set in where you can see the problem.

Because when we have the lock, then we can see, is it the end demand that is our problem? Is it because we changed the safety stocks, the MOQ? Is that causing it at the finished good level? No, it’s not. Okay, is it the semi-finished elements? Is that the one? Yeah. Actually there, we have some huge lot sizes on that, or maybe on the finished good in the production. Okay, that is really causing it, and we have some constraints in the production, for example. Going through that, that would be how you sort of identify where does it change a lot.

 

Søren Hammer Pedersen:

Yeah. I guess also, at least for the fewer analysis I’ve been close to, there is also somewhat of a trial-and-error aspective over this when you gain that understanding. I’m talking about maybe the parameters we put into the MRP. Some companies have, through all the goodwill in the world, put in these customization over years. That means that the MRP takes certain data into account that it maybe shouldn’t for some reason within, and that can be a bit hard to find out, so I guess also really looking data point by data point and see what happens, again, back to the locking and the change.

 

Benjamin Obling:

Yeah, yeah. That’s one of the issues here, that it is very time-consuming to find out why do we have 10,000 as a purchase or a production proposal, because you need to go maybe four steps up the chain to find out why is that. You can say another element here is the transparency, so creating reports where it’s easy to see... Here, I have this purchase of 10,000, what is the outbound? Is that a forecast? Is it a derived demand from a production order? Is it from which factory and so on? Okay, on that factory, so being able to easily jump from step to step in the requirements, creating that reporting and transparency makes that more easy. It’s not super easy.

 

Søren Hammer Pedersen:

No.

 

Benjamin Obling:

It does take time, but certainly, the transparency will help. Another thing is, also, here, when we say into the mid-term and long-term, don’t use the MRP, then we say, ”Okay, so what should we use then?” Also, how do we see what is the stable demand? Here, another trick can be at the same time as you’re running the MRP, you’re also making what we call a direct-derived forecast. What that means is you take the end demand, so your forecast at the end product, so that’s really your finished goods, what is all the outbound that you have there as your forecast? And then you look at the total derived demand throughout the bill of material, but without taking into account the lead time, you don’t take the MOQs, you don’t take the stock levels, et cetera. You basically say, ”If I’m going to sell 100 of these in Germany, that means I will get the 100 from my Czech factory.”

For example, in Czech, I need this semi-finished, I need four of those, so that’s four now, and it was 100 so that’s 400, and then I need 10 more going into that at the component level. That was, what was that, 4,000 as the purchase, which is then the direct consumption derived by the end forecast, because then you know that at this component here, you might have a purchase proposal of 80,000, but that is because of the MOQ, or you might have a purchase proposal at that component of zero for the next half year, for example. But actually, when you have this direct-derived demand, then you can see, but the consumption we’ll expect on an ongoing basis is actually 4,000 per month, for example. It’s completely stable. When you look and you sort of bypass all the MRP, then it’s completely stable, and that’s really critical information, because otherwise, you’ll have a full stop or a complete start jump off.

 

Søren Hammer Pedersen:

I think an excellent point, that the MRP actually makes what could be quite predictable unpredictable-

 

Benjamin Obling:

Yes, it does.

 

Søren Hammer Pedersen:

... and a bit scary, but that’s another discussion. Also highlighting the point you said, I think we’re not trying to make this look like it’s going to be a snap of the fingers and fix it. This is complex stop and it will take time to fix, but I think it’s worth the effort, and if you need to highlight a business case in any shape or where you can start calculating what it costs to do all the rescheduling, or the firefighting, or what you do as a consequence of not fixing this, but if it was easy, everybody would have done it now, of course. Yeah.

 

Benjamin Obling:

Yeah. But you could say, with these tricks, then you can certainly improve it here, and also maybe talking to this part on the direct-derived, because another element here is in the supplier forecast where we can really see that if you send a supplier forecast based completely on your MRP, that is going to be all over the place. Just like we talked about, you can timestamp it and so on, and that’s basically what your suppliers will receive. They are timestamping. At least you can say, mentally, they’re looking at it and then saying, ”Okay, this is completely different from the supplier forecast we got from you last month or two months ago. It’s all over the place, jumping up and down. I mean, do you have control at all over your supply chain? What is going on?” You have a very hard time explaining why this because there’s seven levels in the BOM, and sourcing flows, and so on.

One trick there is also to combine it with this direct-derived forecast, so you actually give your suppliers, say, the 4,000. In this, maybe it’s a bit difficult to follow verbally here, but you say this example where we were selling 100 to our German customer there, that’s our forecast, and then through the chain, that means we need 4,000 of this component per month, every month, in order to... Sometimes it’s 4,200, et cetera, based on the end forecast, but bypassing the MRP completely. If we then give that, provide that to our-

 

Benjamin Obling:

... passing the MRP completely.

 

Søren Hammer Pedersen:

Yeah.

 

Benjamin Obling:

If we then provide that to our supplier, if we only supply that, then they would say, ”Okay, so then you need 4,000 next month.” But no, actually not, because we don’t need anything until half year from now, because right now we have too much in the chain.

 

Søren Hammer Pedersen:

Yeah.

 

Benjamin Obling:

It’s not on that component. It could be on two, four, five different components in the chain. It could be on the warehouse in Germany, on the finished good, it could be on the semifinished good, it could be... et cetera. But somewhere in the chain we have too much, and that’s why we’re not going to purchase anything for the next six months. But instead of just showing you, we’ll not purchase anything more. Let’s say imagine we send them a forecast only for the next six months, then we would actually tell them, ”You don’t need to produce this anymore. We’ll not purchase it.”

And then they’ll start to, ”Okay, we can reduce our purchase, we can reduce the capacity.” And then you’ll come back, let’s say two months from now, now it’s open. Okay, now we need 80,000.” So instead, let’s provide them actually with both and then say, ”Okay, the steady demand bypassing this full MRP monster that we need because it’s doing good stuff as well, but let’s bypass that and then say the stable demand is 4,000 to 5,000 per month, for example. And our best guess is right now we’ll have a purchase of let’s say 50,000 in month seven from now.” But that is going to change and the prediction on that is pretty poor, but it’s going to be somewhere in that area. But don’t worry, the consumption, the sales of this is actually stable. I hope it makes sense without an example.

 

Søren Hammer Pedersen:

For me, it makes perfect sense. Let’s hope the listeners are following as well. And actually a perfect bridge over to... I think the last thing I would like to discuss is because of course, as I said a few times now, it is complex stuff, but are there some shortcuts here? How can modern supply chain solutions help the MRP here? So we don’t need to do everything down there. How can we improve what they get to eat on a daily basis and also these workarounds that you talk, all the areas that you would like to highlight there?

 

Benjamin Obling:

I think those are the most important ones that we have touched. Really improving the input, that will fix a lot. So you can say a new MRP, a new ERP system might not be the solution because the MRPs are pretty similar across ERP systems, for example. So fixing the inputs is certainly the first step. And then making these tricks around finding out, do I have a problem? That’s time stamp it. Let’s look at how volatile or nervous is it actually.

And then making this direct derived demand as a reference case where you can see, okay, I have this MRP proposal. I think it’s going to be here, but my steady flow of purchase of components is something in this area here because then we tell our suppliers, for example, and also the capacity, very important element there, that also means the capacity will be more stable if we use the direct derived demand. Because that means we know we’re not going to be in big trouble in July, but then you can show that when we look at it on a steady base, so directly derived, then we know, okay, then it’s actually going to be smoothened out quite nicely.

We’ll have some MOQs. We have different logics and so on going on, meaning that we might have an issue either in July or August, for example. We can see it’s jumping a bit back and forth, but you can say the general level is okay. And that’s super valuable input as compared to just July is going to be a disaster. And then next month it’s not. It’s actually the other way around.

 

Søren Hammer Pedersen:

Okay. Excellent input. But I think just to catch up on one thing you said about the capacity, I think also an interesting point here that could help to some extent is let’s see some of the issues a good time before they hit the MRP and the rough cut capacity planning then to some extent becomes interesting to move it away from the MRP and move it earlier in the state so we can start to fix some of the bottlenecks, some of the things that we have before we actually start the actual operational planning.

 

Benjamin Obling:

Absolutely. And then being able during the S&OP process to see directly, okay, if I increase demand in Germany by 20% on revenue and this brand, how is that going to impact the utilization of production line for in our check factory, for example, or full across the network, being able to see that right away without having to wait until you upload the forecast, then you start all the production proposals in the ERP system and then you can see, okay, what is then the capacity? Okay, we have a huge problem here. Okay, how do we handle that? Instead, do that earlier and then have the rough cut estimate, which will probably be quite accurate and actually less nervous than the one that you get from the MRP.

 

Søren Hammer Pedersen:

Could you also utilize maybe as a last thing, the scenario planning more to understand what is that we send in? So again, something that we move around, understanding what kind of bets do we send into the engine in the end a bit better before you actually commit anything.

 

Benjamin Obling:

Absolutely. And you could say, again, the capability of simulating the full MRP and actually doing that even across ERP systems. So let’s imagine you have a company where you have four different ERP systems because you have a buy and build, et cetera. Okay. Then it takes quite a long time before you can actually then see the capacity impact on this check factory, for example.

Okay. Instead, you can actually simulate that before, make that as a scenario high, low, et cetera, and then compare, okay, how is that going to impact across the full supply chain, across factories, across ERP systems before you upload it? And then you could go ahead and upload it once you’ve leveled out and you have increased the availability or you know there will be a problem, but we have time to fix it. So it’s okay. Or the problem we’ll face is something that will hit us either in August or in July. We don’t know exactly, but you can say we actually have the overall capacity there. So we’ll be able to fix it when we get closer because the picture is going to change, but we know the total capacity is okay.

 

Søren Hammer Pedersen:

Perfect. Benjamin, we are running out of time for today’s session. I think it was excellent advice for people and I think the biggest takeaway from me anyway is that there’s no reason to continue to accept how things are. And we have always done like this. I think it’s elephant that we have to start chopping up in pieces and finding out how we handle because the benefits can be quite dramatic if you get this somewhat more right than many companies have today.

So thanks a lot, Benjamin, for excellent advice out there. And thanks for you out there for tuning in today. I hope you excuse us that this was a bit more nerdy topic here today, especially if you’re not in a production company. Then sorry, we’ll find a good topic for you in other companies as well. But I think this is something that in a production environment is filling up a lot of space, both in terms of money, but also in terms of resources. So I think it’s something that you should look into.

Thanks for turning up today. Highly appreciated. As always, Benjamin and I would love to hear from you if you have topics within supply chain planning that you would like us to give our perspective on. Otherwise, if you want to talk or not go into Roima’s homepage, have a look, reach out if there’s something you find interesting. Other than that, we hope to see you again soon in the S&OP masterclasses and have a great day out there.