Guest Podcast

Mark Lilly (25:08):
Yeah, I think it might even go back to the sixties and seventies, Sam. Yeah. So, um, please <laugh>, uh, what we're talk, look, um, you know, uh, compared to how things were done in the fifties and forties and before then, I mean, M R P was definitely an improvement, right? So, yeah. Um, we, especially with the, the ability to traverse multi-levels of a bill of material and, and come up with some idea anyways of when you might need those sub-assemblies or, or down to the lowest level of components. So, so by all means, it was, uh, it was an improvement, perhaps a dramatic improvement to the way, you know, things were being done manually. And for a lot of companies today who are doing things manually, it, it is an improvement. On the other hand, um, folks who have experience with M R P and the folks who, who, um, have used r p a lot, um, experience, um, a lot of the, the side effects and the, and the, um, the, the headaches you get with, with M R P, um, one of the most, um, uh, that, that we see most often is, uh, the, the bull whip effect, right?
(26:13):
Is, is having, having too much of certain items, right? Um, uh, and, and just too little having, having stockouts of a number of items, either it's, it, it, the mr the M R P functionality tends to, um, encourage kind of a feast or famine type of thing. And this is, this is true through even in the four walls of an organization. And certainly when you bring in the, in the supply chain, and we, we, you heard some of the evidence of this in, in this conversation, right? You have a, you have a need for three parts, right? But, but no, we have a policy here where, where we don't want to order just three for, for cost purposes, and that's where we're planning and costing can often butt heads, but for, for costing purposes, we don't want to order less than 20 of those, right?
(26:59):
And, but then when we get to the supplier, the supplier's insisting that, that we only order in lots of 50, for example, of this sort of thing. And that, that's just a simple example of how, you know, a little butterfly flapping its wings turns into this tsunami of inventory that that ends up in your plant, or, or, or not, right? So, um, when you get down to the tactics of it, um, where, where Abu is going, right, right at the end, that he's, he's right. You, you want something that's dynamic, and this is one of the, um, you, you want to, you want to be on top of this information almost constantly. And this is one of the other problems folks have with M R P is, is that a lot of the parameters that you've set are quite static in nature. Um, lead time, for example, okay?
(27:49):
Just take, take the example of traversing that bill of material, right? To, in order to try to figure out or calculate when you're going to need, uh, a certain item, sub-assembly or raw material, you know, six levels deep, right? Well, what are you doing? You're, you're looking to the item master at each level to look for a static number, a static lead time to say, how long is it gonna take me to get this sub-assembly? Right? It has no idea of how many you're producing. It has no intelligence about the fact that if I produce two of these, it's only gonna take me three hours, but if I produce 200, it's gonna take me three weeks, right? It, it has no, it's just one static lead time value in your database for that part. So, so this is another area that skews Mr. P's results tremendously.
(28:41):
So, um, a a few years ago, um, a a couple people took a really hard look at this, uh, Carol Patak, who was a, um, an apex, uh, national president for a number of years in the two thousands, uh, together with a fellow named Chad Smith, who was a T O C lean type expert. Um, and they came up with an entirely different approach. Um, that kind of was the, is the antithesis kind of an opposite view of traditional M R P, and they call it demand-driven Mr. P. Um, and what demand-driven M R P is, instead of, instead of trying to guess the timing, and that's one of the things with M R p, you're, you're always trying to guess the timing and if things change, you've gotta rerun that program, and then the timing is different again. And so you're constantly doing pullins and push outs of, of work order dates or purchase order dates and this sort of thing, okay?
(29:37):
So instead of looking at the timing of it, um, what they decided, what, what they, the model they put together is to look primarily in the past for usage, for, for historical usage, and based upon the historical usage and certain variability factors. And this is, this is key today too, cuz we can talk about M R P and ex, the, the static parameters and the, and the fact that that M R P is driving you towards zero inventory, right? That is the M r P ideal as free, uh, aside from safety stock, the ideal in, in M R P is that your supply matches your demand perfectly and you end up with zero inventory. Well, you don't want zero inventory, right? You want enough inventory in for any SKU that you have. Okay? So, so you can introduce the topic of safety stock in that regard, fine.
(30:33):
Um, but there again, it's a static value. It's not changing with the, with the changing nature of your demand, uh, or supply. So, so that's what these folks put together for, for every sku you establish what's called a planning buffer. So you're looking at not only what you have on hand, but also what you have coming in, okay? And when that coming in may be purchase orders could be work orders, right? And then there's a reasonable order quantity on top of that to establish what's called a planning buffer. And the what's neat about this planning buffer is it's like a shock absorber, and it literally flexes and changes as your demand changes. So if you get a, so it's the, the, the nature of the values, there's a number of values that, that support the, the size of these buffers, right? And, and you can look into the, look into the e R p, anybody's E R P has these values, their usage values, some of its variability values.
(31:34):
And so these buffers are established, and then as time goes on and demand changes up or down, these buffers flex accordingly. So if a, uh, so if usage goes up, for example, then the buffer's gonna get bigger. If usage goes down, the buffer's gonna get smaller, and these buffers also trigger replenishment orders. So it's, it's really simple. The, the entire buffer is looking into the future for you. Okay? So it's looking in the past for historical usage, but it's also looking into the future in terms of what you have coming in. Um, and, and possibly forecast too. You, you can incorporate forecasts in this model as well. Um, and in some cases you have to like, for, uh, a new product introduction, for example. Um, but we've been, um, there's, there's a, a number of case studies. So you can go to the Demand driven Institute. They're kind of a, you know, a, a generic site where you can learn about these concepts, lots of videos, lots of case studies. Um, and it, it's, it's really fascinating some of the results that, um, that folks are getting with this approach to replenishment.

Sam Gupta (32:46):
So, awesome details there. Love it. Uh, you know, obviously for the customers who don't even understand M R P and I included, I guess, you know, it's probably going to take one more life, uh, to understand the M R P completely. And now you have introduced some more variables there. Um, so I don't know when I will be able to understand those, but the point, I mean, uh, there are some very, very, very interesting layers. And I wanna touch on one thing, which is, uh, the assumption that you are going to have historical data that you are sort of using to come up with your planning buffers. Now, the challenge with that is, unless the system is sitting outside of the LP system, okay, then you are probably gonna be okay, maybe it sit in the SNOP P system, because typically in the SNOP P system, these things are gonna be easier. Your historical data is going to be easier to maintain. If you had to get your historical data inside the LP system, good luck with that. Unless you are using the same e r P system for 20 years, which nobody uses, you know, you are not gonna have historical data. So how is the altham going to work again, mark, uh, let's say if they don't have the historical data, what enough historical data.

Mark Lilly (33:53):
Yeah, so great question. I mean, yeah, yeah. You're, you're in a bind there. If you don't, if you don't have any historical data, if they haven't been on any e r p ever at all, then yeah, you're kind of, you're starting out a little bit blind with this method for sure. Um, I think typically the, you know, the, the average span of usage you wanna look at is probably minimally three months. And in some cases it kind of depends upon the, the lead time for the part, right? And the, and the demand for the part, how far back you want to go to look at the usage window. Um, I think nine months is probably as far out as it as you typically want to go. So consider, you know, yeah, once you're on your, your E R P and you have, you have that usage information for at least nine months, you'd be good to go on an approach like this.

Sam Gupta (34:37):
Okay? Amazing details there. Thank you so much, mark for that. Uh, so Chris, I'm coming to you. Uh, comments or comments? Any sort of follow up stories? Sure,

Chris  (34:44):
Lots. So one of the things that everybody's hovering around in, you know, the static dynamic values in the r p, right? Well, these aren't working anymore. Business intelligence, so we haven't said it bi. So the analytics on top of what we're doing is imperative. And, and the, the real cool thing is if you could take the analytics and update the values in your system, so look for it because that stuff is showing up, but that's a big deal. And if you think about fill rates and, you know, you talk about warehouse architecture, heat maps, do I have things in the right place? Well, how does that help? Well, that's velocity, but the Rob nailed it. You know, this replenished the fulfillment bins and the two tier warehouse structures where, you know, and again, what's r p do? It tells you what to buy, when to release the po, what to make, when to release the production order, what to transfer from where to where, when to transfer it, what to cancel.
(35:34):
Oh, that's important too, isn't it? What to cancel. So a lot of stuff M R P can do for you and the bend to bin transfer is, and again, as you get more elegant, and, and this is velocity, right? So not everybody needs this. Not everybody has a hundred thousand or more SKUs. And so the manageability of this, right? And again, you're still segmenting and a heat map is, well, what do I need to manage first? Meaning, what are your fast moving items? So if you're looking at your top 50, right? Mr. P, start there. That's the most important stuff to manage the old 80 20 rule. So again, how do you cart, you know, crawl, walk, run, but analytics are gonna help, you know, what are those top items? Focus on those. And again, the refinement process, everybody said it. What are my lead times?
(36:10):
Does that impact cost? Is you get it, you gotta get analytics. You gotta say, okay, if I do this, how much more can I save? How much room do I have? How much cash do I have in the bank? What's the economy doing? I've got guys in the H V A C industry that said my inventory went from 12 to 38 million. Okay, now they're trying to bring it down. What's that tell you? They bought everything they could. They've, they violated all the replenishment rules and it's a grab. So that takes it off the systems capability, right? So now human intervention, right? We're in a precarious supply chain world, so that kind of throws replenishment, you know, somebody said that the other day, just in time wasn't meant for this time folks. So j i t doesn't fit anymore. Nobody said that either, but it's very interesting.
(36:47):
So the analytics depends all good stuff. What else did that jot down here? Again, the electronic updates. So again, you think about how you get started. Again, I'm a technical guy. I've got a client now and I do SQL updates to their r p data cuz they don't use it today. Actually they do. They make all their appeals manually. I had a session with 'em last week. But they have, you know, very complex bills and materials, probably 50,000 SKUs, okay? More than you can even, hey go update all the lead times. Well, what we do is, again, technique list of vendors, lead time by vendor, update every item for that vendor with that lead time. There's a quick way to paint lead time across a vendor. And to Mark's point is you're analyzing, right? Average, average performance, you gotta look at the performance. Maybe the number you put in there isn't, right?
(37:31):
Okay, that's gonna make a big deal because these lead times offsets, right? It's all about what's the capital float that's on balance sheet inventory versus cash. You know, we have conversation about how do we get more cash? Will you manage inventory better? That's, that's a big part of it. So again, I'm not gonna under simplify the process, but analytics are imperative. And again, pick, pick what you can start with and get that going. Is there anything else? Uh, I think that's, oh, warehouse architecture. We started talking about that. And as you look at fulfillment, again, you're figuring out how much do I need? How much room do I need? Yeah, where's it at? So again, we talked about logistics, we talked about heat map, but again, even replenishment. What are the movements receiving? Okay? Is that part of replenishment? Yeah. Okay, so <laugh>, does the truck come in?
(38:16):
Does it go to the dock? Where does it go? Do I do a putaway? Yes. Can I fulfill off the dock? So there's a replenishment can. And again, now we go back to the sophistication of your system. And again, I didn't mention so warehouse management, not just warehouse architecture, but the elegance of the warehouse management. Most systems don't elegantly create transfer orders. I mean, Mark's probably mine does. And Dynamics f and o does BC doesn't. So some do, some don't. But, but again, the sophistication of your, your mobility device, right? Because again, you can have the system and tell you to do this, but how much does it automate the work? Does it create the pos, does it create the production orders? Does it update automatically? This update automatically? That, and again, we're in your high volume numbers of skews. You can't manage that by people. So again, the sophistication of that mobility absence, hey, I need to go here and get this and move it over here. I gotta go in manufacturing. I got a picking order to pick to a cart. Okay, that's a replenishment in essence, but you get it. So

Sam Gupta (39:09):
Yeah, let missing, uh, details there. Uh, and the, uh, one thing that I want you to touch on is going to be what to cancel. Uh, I guess that's a very interesting layer. It is interesting. So how, how, how does that work? I mean, describe the process a little bit there.

Chris  (39:21):
Okay, customer calls in, ah, canceled my sales order. It was a make to order. Oops, okay. The ripple effect of a make to order. I don't stock that item. I don't build that unless I need it. You think I wanna build this a hundred thousand dollars product and put on the shelf, said okay, hopefully I'll sell that thing he wanted. Okay. That's an example of a customer cancels an order. First example to trigger it. I mean, that's the best case that I have is the customer order is canceled or something, something happens on the sales side where I don't need those or it was a mistake. Whoops. And so what happens? I I, you know, I had a fat finger in the, the sales forecast and I had a hundred thousand instead of 10,000 <laugh>. Sorry guys. There's a PO out there for a hundred grand of these and I don't want it.
(40:01):
I had a client, I kid you not, they ordered a half a million pieces and this was a total fat finger. Cause again, you go back to, oh, they were looking at historical usage, it was wrong. There was a bad entry in historical usage and they picked it up and said, oh my god, use a lot order half million. They didn't need a half million, they needed 5,000 <laugh>. And so there's a perfect example where, and again, the other thing we're talking about is the analytics on watching this stuff has gotta be more real time. So that you're looking for exceptions. But two horror stories were, I needed a cancel order. Like real quick it's like, oh, you already shipped it, refuse it at the dock. I mean, there's a lot going on there. We don't take receipt of that shipment. So cancel order.

Sam Gupta (40:37):
Okay. Amazing details there. Thank you so much Chris for that. <inaudible>, I'm coming to you. Uh, comments or comments? Any follow up stories?

Nirav Shah (40:43):
Yeah, no, I have a, actually a pretty interesting story and I think, I think Mark touched on it a little bit, um, talking about it doesn't forward plan, right? Yeah. That, that is one of the issues with traditional M R P and nps. So what we did for a customer, actually there were a big make order, uh, customer. They had, you know, 10 or 15 level deep build materials. A lot of the lower level components were made to stocks. So they're common across a lot of their items. The top level, maybe the first four or five levels were made to order. So when they knew they had a project, they're getting close to kind of, you know, uh, that project being approved, but it's not really approved yet, right? Things had r p once they get into a true demand line demand, which is either a job project, sales order, right component lines, that becomes true demand, right?
(41:27):
Or you got the independent demand like I talked about, but this was none of that yet, right? We had engineers still creating the bill material for the top floor, but they knew what the lower level assemblies looked like. There a lot of those make to stock apps. So we then leverage, uh, as uh, we call simulated production order or a what if scenario type of production order where we brought in the bill of materials, right? The top level, all the way to the bottom level, even though maybe it was like 80% ready to go, but we're concentrating on the lower level stuff, right? Mainly maybe, maybe one or two of the make to order items. But essentially we're able to check off which assemblies that we wanted to go ahead and see or prioritize in the planning run to bring bubble up to the surface.
(42:08):
We could stay ahead of the project, right? Um, and get those things on the production floor to start manufacturing, start purchasing because there was a high likelihood that the project was gonna come, you know, into fruition, right? And get approved. Versus if we didn't do that, we waited until the project came in, we would've already been late based on the due date that you know, that the sales reps were already trying to hit. Right? Uh, cause that never gets communicated back, right? Remember the due dates is whatever the sales rep thinks that they're gonna, we're gonna be able to deliver without a lot of internal conversation with the manufacturing team. Exactly. What is our capacity? What are our Bob bottlenecks, right? When actually are we gonna be able to go ahead and, and uh, get, get this item completed? So, you know, we were pretty creative there and we're able to identify, you know, without having it look at the full structure of a bill of material where we pick and choose assemblies from a full bomb and integrate that into the Mr.
(42:59):
P and NPS run and identify those specific sub assemblies or make to stock from a priority level, bring 'em up and re release them directly from Mr. P to go and get a headstart on it. So I thought that was a big, big win for this customer. They're able to kind of control their manufacturing process a little better and have better visibility on those projects coming in, right? Uh, on the manufacturing floor. So I thought, I thought that was good. And I always tell my customers when we're, when we're trying to implement M R P and NPSs, um, M R P and NPS is more of an art than a science, right? Every company's gonna be different. You, you're the way you interact with your vendors, the type of materials you buy, right? How far you are from your vendors, right? What's the seasonality of your products, right?
(43:40):
We're, you know, all these different things come into play and creating that perfect Mr. P engine, lemme tell you, it's never perfect, right? We can get close enough and we get close enough where we manage that 80 20 rule that Chris was talking about, right? Where we are able to manage a business with the high likelihood that we're gonna be able to complete on time, get our inventory and on time, um, go through the standard routine of manufacturing and, you know, stocking inventory and putting inventory and shipping to the customer. Cuz then that comes down to the warehousing lead times and all that stuff. But it, it's more of an art to the science and iterative. You know, you have to have patience. You have to be as, you have to have the continuous loop feedback to make sure it's humming. Um, and that the proper, proper departments have visibility into the r p thing.
(44:21):
That's a big mis mis misconception that only the mps, the master production schedule, only holds onto that schedule. That's not true. That should be visible to everybody. That should be visible to the buyer. That should be visible to the sales rep that's putting in the orders, right? They should be able to see, oh wait, you know what Mr. Customer, we have this coming through on our planning side, we might think that might be ready on this date because we see it being planned out. You know, let's put a tentative date, date in there that should also be visible right upstream and downstream. It just shouldn't be a silo. That information is very beneficial throughout the organization if yous properly and maintained.

Sam Gupta (44:55):
Okay. So some very, very, very interesting details there. So I am going to have one follow up question for you. And I don't know how many people really understand the difference between, you know, the role of the forecast, the role of mps, and how that is going to impact your, uh, M R P processes. It could be sometimes all over the place. Uh, you know, because, uh, all of them are sort of going to be impacting each other. So let's say if you were to describe the difference between mps as well as, you know, forecast and what the boundaries are going to be of each of them when you are sort of planning, uh, for your supply chain over to you.

Nirav Shah (45:31):
Yeah, I have a very simple example for that. Think of a seasonal business. Think about a toy manufacturer, right? Toy manufacturers will not have customer pos January, February, March, April, may, June, right? They're gonna get their POS at the end of the year. They're gonna get their pos that they have to, uh, produce to the different toy retailers, right? Amazon, whatever it is, it's gonna be October and over over December. Right? But they need to put a forecast in because they need to start making the millions of toys, right? Beanie bags was, was a big one, I think in one year. Uh, but that's the only one that's coming to my mind. But they had to start making that upfront. They had to start buying material January or February, manufacture them March, april, may, right? Get the containers ready, wherever they're gonna ship it to so they could start shipping with anticipation of those pos coming in at the end of the year.
(46:13):
Now, if they don't have the actual POS at the beginning of the year, so they're using a forecast with the high likelihood that they're gonna be able to sell the material right? Through historical sales. So they're gonna implement a forecast to get the planning engine to run so they could get the components right, get the inventory stocked so when the POS come in in October, November, December, they're able to just quickly ship out the door. Right? That is a, a real good example between a forecast and nps. NPS is gonna be telling the system, Hey, you got the, you got the, you got the purchase order now, but we have it in inventory, let's ship it out. Versus the forecast for the same example is gonna be telling the planner, Hey, we don't have a a PO but we have a forecast or we have to make it. Either we're gonna make the customer orders or we're gonna make the forecast.

Sam Gupta (46:55):
Okay. Amazing. Thank you so much Ara for that. Uh, abbu, I'm coming to you. Um, comments over comments? Any stories.

Abu Asif (47:03):
Yeah. I mean, you know, we had a client where, you know, they were at excess inventory and, um, you know, they were, you know, blaming the MRP system. And what we found were they had over enthusiastic sales, uh, uh, demand forecast, right? The sales that's always thought they can sell more, you know, hunt, they can always beat their quotas, but they never ended up beating it. So, you know, that's a very important factor is, you know, especially when sales side of the forecast comes in, you know, how can you accurately, you know, predict your sales, right? And that, you know, even if you're off like five, 10%, it can just throw off your entire material requirement planning, right? So, uh, again, forecast, you know, as Nira was saying, can, you know, vary by the season? You know, you know, you may have business model where you sell more in Christmas and then less afterwards, and then you have to build all those trend profiles and, and all of that into the system.
(47:56):
Um, but you know, I just common one of the comments, you know, um, Chris was mentioning, you know, about BI systems and the ability to, you know, automate the data piece. So in my opinion, that's the easier part, right? Because you can build some sort of system to automate some, you know, automatically update something. But the point I was making is just to gather the updated information, right? So if you have a thousand suppliers, and certainly world market condition has changed, that's going to affect five, you know, uh, hundred suppliers. Even just the effort of reaching out to them and figuring out what's the new lead time going to be is, is a huge effort on its own, right? And that's where the MRP systems often lack, is that dynamic data feedback from the market forces back into your own system, right? And that's where the most complexity of the, you know, MRP system, whether it's on the demand forecast, you know, summer went longer than usual, right?
(48:53):
So I worked for a big company, um, in the apparel business, and they spent a lot of money forecasting how, you know, how the, you know, how mild the winter is going to be, how strong the Canadian based company, right? How strong the winter is going to be, and based on those forecasts, then they order the inventory, the number of jackets, when is the winter going to start, right? But guess what, you know, the forecast change, you know, instead of, you know, we had a very mild October here, no one was even looking for, you know, winter jackets. So it certainly just changes everything, right? And it's that market forces, you know, now the winter can say, you know, either you take up the order or you don't. And you know, it's getting that market data into your MRP system, which I find is the most challenging aspects of development of good MRP systems. And where a lot of companies, uh, you know, fail, right? Uh, BI system, again, they're mostly focused on historical data, right? The reports built up on your e r P system, on your databases. But the BI systems, very few. I mean, even if it's out there just getting runtime market intelligence, right? They're not connected to any intelligence databases or things like that, right? So it's a process, you know, it's a complicated process. It requires, uh, you know, a lot of manpower to maintain it, uh, and you know, to run it efficiently be my point.

Sam Gupta (50:18):
Yeah. So some amazing details there. And I like a bit about, you know, uh, correlating your market data, obviously with your, uh, your internal data. Uh, but the challenge that you are gonna get, obviously with e R P system, as we know, as I commented, uh, about the tightness of the E R P data model, uh, any data that you are gonna store inside the E R P system is always going to be harder. Uh, the e r p data itself is very, very, very hard to manage. Now, if you are trying to bring additional data elements that sort of don't belong to E L P, uh, it could be even more problematic in general if you are going to keep that. And that's why, and you are absolutely right, uh, about the BI system work system of the other system. And that's why we have BS and OP system.
(51:01):
They are really designed for that purpose, and they typically, the output of that is gonna come to your e R P system. And they, you know, they manage all of their complexity where they are going to be overlaying a lot of different data sets. Uh, and they are also be going to be correlating with your data, but obviously your data has to be right. Your integration flows have to be right, otherwise your planning is going to be probably all over the place. Um, so I don't know if, if you have any sort of follow up comments there, agree, disagree, I

Abu Asif (51:27):
Dunno. Yeah. I mean, so I agree with you. I think, again, you're, you're pointing to one is a technical challenge of maintaining the data, you know, system, system type and how to manage integration. And the other challenge is getting the market data from your own market, right? Yeah, yeah. Getting actual data from a thousand suppliers, right? So that's, those are two different things, right? So it's getting that real time market data is, is a harder challenge in my opinion. Yep. Rather than figuring out the technical piece, uh, to get it into your system. Yep. And that's where, you know, people get frustrated with Mr. P is it's that market data. It's not a feedback from the market into the system. So,

Sam Gupta (52:08):
Okay. Could not agree more. Thank you so much Abu for that. Uh, mark comments? Over comments? Any stories?

Mark Lilly (52:13):
Yeah. Um, I want to talk about a, uh, a a subtle point. Um, Chris actually talked about in his opening comments and he was describing, you know, kind of down into a level of bill of materials and, you know, the idea of, Hey, I've got this sub-assembly, um, you know, it takes me, it's gonna take me four weeks to build it, or, or do I want to decide to put that in stock so I can save that four weeks of lead time, right? And, uh, and be able to get my end product out, what, however many levels up that is out, out to my, to my client, um, my customer much sooner. So that, um, that's a whole topic within demand driven m r p and, and they call it strategically positioning inventory, um, and strategically positioning, not in the sense of, Hey, I'm gonna have a warehouse in Milwaukee and one in la Right?
(53:02):
It's more, it's more positioning it in, in terms of the structure of the bill of material, right? So you, you do some analysis with your, with your bill of material and see, you know, how many, how many days of lead time it takes across through, through and down each level, right? And then you can, can kind of see and, and get an i uh, an idea of where, where you have the opportunities, if I s stocked this part, how much would the, would the entire lead time shrink, right? In order to offer? So, um, we, we recently, um, engaged with, uh, a couple, um, D D M R P experts, uh, that brought a, brought a client, um, to us and, uh, working, working together in a situation where they have, uh, they, they have an end product that their, their full bill of material structure and their full manufacturing process is about a 12 to 14 week lead time today.
(53:59):
And because they understand the demand driven principles and concepts, they are, they are so confident they're going to be able to have enough of certain sub-assembly in stock, they're going to be able to reduce this entire lead time down to a guarantee of less than two weeks to have that product. Okay? Now, obviously the balance here is right, you, I mean, if you have a gazillion dollars and, and as much space as you need, you can, you can go ahead and, um, and stock as much as you want, right? So, so the, the challenge is of course, yes, having enough of those sub assemblies at all times, cuz it'll, it'll cost you dearly if you have a stock out cuz that lead time's gonna come right back at you. Um, but at the same time, not too much, right? So you're, you're controlling how much, so you're making sure you don't have excess as well. And that's one of the big, the big principles and the balance points of, uh, of this demand driven, demand-driven r p approach I've been speaking of is having enough of all, any SKUs that you need. So almost 99% sure that you're going to have that, but yet at the same time, not having excessive and not having too much, so you're not tying up all that cash, um, in, uh, in what could be very expensive sub-assemblies.

Sam Gupta (55:26):
So, so there are some very interesting details and I guess, you know, one of the details that you just mentioned, there are subtle nuance there and, uh, that can answer a lot of questions and solve a lot of problems that we have been talking about. So are you saying that, let's say if I have my data off and, uh, you know, I maybe doing make to order for one of PSK right now, but let's say if I want to know whether they, that should be probably stored in the inventory as opposed to making when I get the order, uh, can the Mr. P system tell me that rather than make to order, you should be treating, treating that as more off made to stock? Can the M Mr P and then make those recommendations as well, mark?

Mark Lilly (56:06):
No, no, there's not that I know of. Um, that I know of. There's not a computer program that's going to, that's gonna do that for you and give you a definitive analysis. Someone may have it. I don't, I don't know of it, but typically it's, it's a more, um, it, it's a, it's, it's frankly a, a spreadsheet analysis of looking at the structure of the bombs and their overall products to see, to see where those opportunities are.

Sam Gupta (56:29):
Okay. Amazing. Thank you so much, mark for that. So the only thing we can take right now is going to be closing advice. Chris, what would be your closing advice, please?

Chris  (56:36):
So we're not gonna be afraid of this topic, and if we're manually generating POS and production orders today, and you have a tool in your system, look at your inventory, find your most important revenue profitable items, you can put settings in there and run master planning and see what it tells you to do. You don't have to do it. Go manually, create your stuff, run it again and see if it you missed anything. But that's a way to climb into this thing and start getting comfortable. And a lot of the old school guys were like, eh, well when you show 'em the report, he's like, okay, that's what I would've done. That's how you build adoption and you move this thing forward. So,

Sam Gupta (57:06):
Okay. I'm amazing advice. Thank you so much, uh, Chris for that. And then closing advice, please. Uh, you're on mute enough. Uh, 11 time

Nirav Shah (57:15):
<laugh>. 11 time today. Yeah, sorry about that. Yeah, I would say, I would say, uh, companies that are using r p and NPS are kind of, you know, feeling a little down and, and low about that's not working well for them. Look at your data, right? You need that continuous feedback loop. Let's look at, you know, try to create some BI around that to update that data. You know, what we haven't even talked about is other technologies within the warehouse that could also make your r p runs even a much more efficient R F I D tags on inventory, right? That the feedback into M R P and NPSs. So there's a lot of opportunity there. Don't shy away from it, right? Um, you know, o o open, open the door there and you're gonna see a lot more visibility into your business as you thought you knew.

Sam Gupta (57:52):
Okay. Amazing advice. Thank you so much na for that. Abu, what is going to be your closing advice, please?

Abu Asif (57:56):
Uh, yeah, I'm going to act for Chris's comments, right? Start small, look at your, you know, most impactful items and then start building Mr. P system, you know, from, from that point onwards. So

Sam Gupta (58:09):
Could not agree more. Thank you so much, uh, Abu for that advice. Mark, closing advice please.

Mark Lilly (58:13):
Yeah, if your, uh, experience with Mr. P, if you're not happy with the results for whatever reason, take a look at demand-driven Mr. P, demand-driven, r p uh, I'm sorry. Demand driven institute.com is the generic site that'll, um, give you a lot of education, a lot of videos about the concepts.

Sam Gupta (58:30):
Okay. Amazing. Thank you so much, mark for that. So that's it today. If you join for the first time, this was one of our digital transformation series for which we meet every Thursday at 5:30 PM Eastern. So make sure you guys are gonna be here next week. We are gonna come back with another topic on that note. Thanks everyone for tuning in tonight.

Chris  (58:47):
Thanks everybody.

Mark Lilly (58:48):
Thanks all. Thank you. Thanks Sam.