Strategies to Fix the "Late Problem" in Manufacturing with Mark Lilly

Lisa Ryan: Hey, it's Lisa Ryan. Welcome to the Manufacturer's Network Podcast. I'm here today with mark Lilly. Mark is president and CEO of Lilly Works, and he helps manufacturers to solve the late problem. I'm going to let Mark explain a little bit more about that. In the meantime, Mark, welcome to the show.

Mark Lilly: Thanks very much, Lisa. Appreciate you inviting me on. 

Lisa Ryan: Please share a little about your background and what led you to work with manufacturers and create Lilly Works. 

Mark Lilly: Sure. We've been doing this for a long time and referred to the team behind Lilly Works. It's a family endeavor.

My dad started several manufacturing E R P systems, including one in the 1980s called Profit Key. Another called Visual Manufacturing in the nineties is currently owned by Inform. They're both designed for make-to-order high mix-manufacturing types of companies used by thousands of manufacturers.

And they both also had very strong, traditional finite scheduling embedded in them. However, despite this great functionality in either of these products and what we've since learned, any E R P system has this traditional approach to scheduling and managing production that most manufacturers struggle with for several reasons.

So a few years ago, the team got together in the 2014-15 timeframe, and we started a third manufacturing ERP system up in the cloud. And as we were designing it, recognizing that while that scheduling functionality was good and the previous folks struggled with it, we said, what can we do differently?

We came up with an entirely different approach to managing production. So, from a material and a scheduling standpoint, we called this approach, or the software part protected flow manufacturing. The process is the dynamic production method.

And we would go out into the marketplace and show folks the E R P liked this approach to managing production, but they didn't necessarily want to replace their E R P right now. So that's because that's so much work. So we extracted that part of the software. We made it its own product offering called protected flow manufacturing that ties into any E R P system, whatever you're using today.

So what's nice about that is we can go into a company and help them in very short order. We're talking six to eight weeks and solving the late problem, right? Give them visibility and production of their true priorities of when and if the material is here or not to execute the prior.

And so everybody can see and know what they should be working on for future visibility. So that is when my customer will be able to get their order based on my capacity and material availability. 

Lisa Ryan: So what exactly is the late problem, and what is the extent of the problem you're seeing?

Mark Lilly: Sure. Many companies are struggling with this today, and very simply, it's not being able to get their orders out on time when they want to. I did a presentation at IMTS in Chicago last week, and I pulled together some macro data. In reality, there are over a trillion dollars in unfill fulfilled orders across American manufacturing in July. That was the latest statistics that came out of basically the fed. And that's about half of the entire manufacturing GDP. To put that into context, the manufacturing GDP is approximately just over 10% of the national GDP.

So there's a tremendous amount of opportunity. And I think anyone who works in a manufacturing company realizes an experience is this, whether it's for material or the supply chain issues we're having. Internationally, whether it's the workforce or not being able to find people, most companies are struggling because they would be able to ship if they had the capacity. If they had better management of what's happening in their production environment and better visibility of what's happening.

Lisa Ryan: Sure. I think about this, the late problem, and the first thing that came to mind was, okay, supply chain, everybody's struggling with the supply chain. We can't get parts. They're sitting in ports in LA or whatever is go that whatever we've been seeing the last couple of years.

How can people turn that around when many think that's just part of the process? It seems that if somebody could turn around, that would make them stand out from everybody else too. 

Mark Lilly: Yeah. I think it's making folks look because even before all that happened, folks still struggled with getting their orders out on time.

People realize this. So I think it's just bringing the shining a light even more. A lot of it can be helped internally. If they genuinely had visibility of when materials are coming in or even if they had the materials right now and also what their priorities should be based on the dependencies of what they're manufacturing and what jobs or work orders are through.

Most are in danger of being late rather than trying to schedule things based on a due date. However, this is one of the things we find in a high-mix environment, and most scheduling programs or tools will attempt to prioritize your work on the due date, especially in a high-mix environment.

You will have jobs that are due earlier. But that is less in danger of being late, just by how much work needs to happen on a job and whether it needs to go outside or back. So you could have a job that's due two months from now. That's much more in danger of being late today than if you don't get started on it than a job due even a week from now. 

Lisa Ryan: So how would you even figure that out?

Mark Lilly: That's a great question. And that's what we do. We work with whatever data they have, whether in an E R P system or not.

And we bring that into an approach where we create an execution plan for every work order. So you can see how much work needs to be done or for the quantity you're making when the due date is. And then there's actually, and whether folks realize it or not, they're using a lot of buffer time.

So when you ask somebody, Hey, what's, what lead time are you quoting your customers? They may say six weeks. And then you ask this, the follow-up question to that. Okay. Of that six weeks. How much time is actual production time, right? Touch time. That will be on a machine, or somebody will be explicitly adding value to it.

And that's typically their estimates of the routing from this operation steps that need to get done. So you often find that on a six-week lead time job, the actual touch time is a fraction of that. Sometimes only a week, sometimes days. Sometimes even hours. So companies are using a fair amount of buffer time between the actual time that's going to be needed and the due date.

So we consciously look at that. We work with the clients that say, is that, say, five weeks of buffer time? Is that working for you? Is that, is it too much? In which case, maybe we have an opportunity to reduce that, or perhaps it's even too little, and maybe you need to expand that a little bit, be conscious about how much buffer time you're putting in there.

The dynamic is then that the execution plan will flex based on quantity. Which lead times, don't lead times a static type of information but should be flexing based upon the quantity. And then that also designates when you need materials upfront. And when that job should start.

Okay. Now when the job should start is important because what the other thing we often find. Manufacturers have this idea in production that, you know, gee, as soon as we have material, let's get it out into production. Because then it'll come out the other side as quickly. And no, you don't want a flood WIP with too much work.

There's a principle called Littles Law, which says that there are more things or items in a system. So the longer the time, any of those items will be in the system. A good example is simply a highway. I live in the DFW area. And as you can imagine, we've got certain five-lane highways during rush hour, and you come over the, over the on-ramp, right?

All are brake lights. And, you're going to be on that strip of highway, say it's, for five miles for 20, 25 minutes because there are so many cars on that highway. The flip side is you come up in the middle of the day or at night. And everybody's flying by you these fewer cars; you're going to be on that strip of the highway for five minutes or less.

The same way in a production environment. The more work orders, the more jobs, and material you send into production. There's so much traffic. The longer the wait times are going to be on each of the work centers. A more extended period each it's just going to take to get through. That's why there's a lot of expediting at the end of the month.

So you don't want to starve production, but you want to control how much you send out. And when you do that, whatever is in production has the best chance to flow through its work centers as quickly as possible to shorten your lead time and get out faster. 

Lisa Ryan: So that certainly sounds like some of the struggles manufacturers have with that traditional method of managing their production. But what are some of the other struggles that you're seeing that they're having when they're doing it the old way? 

Mark Lilly: Sure. So one of the things with the traditional scheduling model is that you are running a computer program. So you have to run a computer algorithm to get your plan.

So you try to set up your parameters as best you can. You run this computer program that you, some folks, may understand precisely what it's doing. Others may not. And then you get this list then of priorities from the program. And even if it's, there's no such thing as a perfect plan, but let's suppose you have a good plan. That gets to the reality of production, especially in a high-mix environment where things constantly change. Then something does change. Whether a customer calls and changes a due date, a tool breaks your best setup person doesn't show up.

The beautiful plan you had now doesn't make sense anymore. So it's probably not worth very much, so we realized there needs to be a dynamic production method. It recognizes, acknowledges, and anticipates that change will constantly happen in a production environment.

So instead of devising a plan and executing it based on that, we do what we do and use those. So an individual execution plan for every work order and job also calculates a real-time priority. And it's very simple using where we are today.

And how much is remaining on the execution plan, whether it's all the work or just a portion? And then, how much time protection or buffer do we have left to our due date? We call that a threat level. So based on that ratio, How much in danger of being late is each one of your jobs?

And we publish that in real time, out in production, so that everybody in every department and every work center on every machine can see and know precisely which job they need to work on. So what is the top priority regardless of the due date, right? Like because, as we described, there may be jobs that need your attention now that are due later rather than other jobs.

Lisa Ryan: So how much do you think this is costing the average manufacturer to have this late problem happening? 

Mark Lilly: That's a great question. As I mentioned, it's in the trillions on a macro level. When you ask most individual manufacturing companies, and if they're being honest, most will recognize that.

Maybe in the eighties percentile in terms of on-time delivery. And many struggles with being in the sixties or seventies, even lower. So it's not easy, but you can imagine. And actually, we can pull their data in and show them what it would mean.

To be able to ship, on the extreme, if they were, and they could do this math as well, certainly by looking at their backlog, if they were able to ship everything on time, what that would mean from a revenue standpoint. But knowing that they're not going to be, that 30% or even 20% of their orders will not make that deadline mark, they're going to fall into a later period.

What is the financial impact of doing that? So you can imagine there's a tremendous opportunity if you can shift the bar even a little bit, so if you can go from 70 to 80 90 percentile, that change. So you're now shifting when you're going to be able to recognize that revenue, and it's a tremendous financial impact. 

Lisa Ryan: So when manufacturers are shifting and going to that dynamic production method, what are some of the results they're seeing? 

Mark Lilly: That's a great question. So there are three components, and I've touched on all of those in different ways. One is anticipating the variability. So have to make sure there's, you're acknowledging you're using a buffer today and talking about that. Where should that be? In terms of when we start jobs and when we need materials. So anticipating that variability using the proper priority.

So instead of using the due date to schedule, use a risk level of how much in danger of being late. Or of having a stockout, if you're a make-to-stock tech company, are you that's the second one? And then the third is controlling WIP. So how much WIP do you see is once the method is in place and you're doing all three of those things, there's almost an immediate reduction in work and process.

So if you are in a situation where, you know, and many manufacturers do just by walking out on the floor, they realize we've just got too much whip out here. Just so much an immediate reduction happens in just a few weeks. And that's a tremendous cost saving, right?

So you can go to the cost accountant or the CFO of the company and ask them how much WIP are you carrying right now. And with this approach, we can reduce that in very short order. So do the math 30%, even 40% of your whip gone within a few weeks.

That's a tremendous cost saving. But the bigger part is getting work out faster with the resources, the machines, and the people you're already using or paying for today. And now that becomes a contribution margin in the equation or some call that a throughput dollars equation. With our toolset, you can see this as well. Financially. You can see if I was, if I'm doing things the way I'm doing, now I can see how many dollars I will be able to ship. But if I could make these improvements or get things out faster, how much more in a given period would I be able to ship boats from a revenue standpoint, but even more powerful contribution? The neat thing about contribution margin is that any additive contribution margin you can bring in a given period goes right to the bottom line. So it's a direct profit increase for the company. So it's a very powerful lever to help a company grow and improve its finances.

Lisa Ryan: When you have a high mix manufacturer and face it, nobody has enough time to do anything today. Everybody's plates are overflowing. So how would you even start prioritizing as far as figuring out your most significant risks and knowing where to start? 

Mark Lilly: That's a great question. And companies are in different stages of being able to implement something like this, right? So we go into some companies that already have an E R P in all their routings and data or are pretty, pretty accurate. So they reflect how production is happening, and we can tie in, but they may still struggle with its scheduling components or material visibility.

We can tie in very quickly. And within a few short weeks, they can get those priorities. We talked about other companies in the future, and I've been in several of them even recently where they don't. They may have a system, but whatever data is in there, they don't believe it either. So here's how production happens.

Maybe those routings were set up from a costing standpoint, accounting used, but when you get out to the shop floor, that's not how material flows. So they need to rework those. And then there are dependencies, often manufacturers who make the complex product where sub-assemblies are involved.

So are those sub-assemblies specifically for a parent? So they need to be aligned very tightly, or are they common parts, common sub-assemblies that maybe even need to be stocked, in which case they also need to be aligned, but a little bit differently to make sure they're going to match whatever the first parent is, that's going to need them.

So, that does need to be looked at. The good news is. Now we can't create data out of magic, but most folks have it in their heads or often have it on spreadsheets. That's what we usually see where folks have struggled with the traditional scheduling model. They have scheduled in their E R P.

They may have even purchased a bolt-on. But, unfortunately, they've just struggled with it, and they revert to spreadsheets. So the good news is we can use those spreadsheets as a data source to help them move forward in this direction. Is there some trigger or sign or something to pay attention to that? 

Lisa Ryan: Wow. This is an issue that we need. Take care of it right away. Is there a timeframe or anything again that would be like that siren that, Hey, this is something you should be paying attention to right now? What does that look like? 

Mark Lilly: Yeah. If companies aren't already monitoring or measuring, they're on time delivery percentage. So they're not formally doing that. There are likely a few folks in the company who know how good or not that is undoubtedly sales customer service. And that's where the issue arises with customer relationship management, whether it's an existing customer complaining about not meeting the due dates they had established. Or not being able to win new business because of the lead time. If you're being honest and quoting a lead time, it may be longer than some of your competitors.

Lisa Ryan: And face it, nobody on the planet likes change. So if you're bringing this kind of change into a manufacturing plant, how do you get buy-in from the employees to let them know that this isn't one more program, that it will fail in the next couple of weeks, or that this is going to help them? How do you incorporate the system so everybody sees the value and benefit? 

Mark Lilly: That is a great question. And a critical question. So typically, we don't get to the point where there's a financial commitment from the client to move forward without at least introducing this to production and getting some in that regard.

But to your point, We have encountered situations. For example, we've got a modern machine shop article about keeping schedules on track. About that, that very point. We implemented this. We got management's buy-in. We had the big screen TVs and the workstations in production, showing the real-time priorities.

Everything was wonderful. And eight weeks went by, and management said nothing had changed. We're still on time. Delivery is the same. Our WIP seems to be the same. So we did a deeper dive and realized we hadn't gotten buy-in from the production folks.

They were still using their old message on a piece of paper,...