Amir Schragenheim, President of Inherent Simplicity, is a software firm specializing in TOC software for Production & Distribution environments. Inherent Simplicity is the exclusive software supplier for Production & Distribution software to Goldratt Consulting, Eli Goldratt’s consulting firm, in their Viable Vision strategic projects. Mr. Schragenheim is a regular speaker at the TOCICO conferences and is a TOCICO certified expert in Supply Chain Logistics, Finance & Measures, Project Management, and Business Strategy. Before Inherent Simplicity, Mr. Schragenheim worked as a chief analyst for Attunity Ltd., where he was in charge of defining the technical requirements of a software tool to handle business processes.
Download PDF Transcript of Podcast
Note: This is a transcription of a podcast. It has not gone through a professional editing process and may contain grammatical errors or incorrect formatting.
Related Podcast: Implementing the TOC Supply Chain Solution
Transcription of the Podcast
Joe Dager: Participating in the program today is Amir Schragenheim President of Inherent Simplicity, a software firm specializing in TOC software for production and distribution environments. I was introduced to Amir through the recently published book that the Theory of Constraints Handbook and a contributor for the supply chain management system section of it. Would you introduce yourself and tell us what Theory of Constraints Software does?
Amir Schragenheim: Sure as you mentioned my name is Amir Schragenheim I am the President of Inherent Simplicity. And what we do is software that is an add-on for a ERP system of the organization or any order enter system they have and we supply the TOC functionality meaning the people who have read “The Goal” or the recent “It’s Not Obvious” book by Eliyahu M. Goldratt and would like to have TOC priorities on the floor and would like to manage their inventories using TOC principles they could use our software in order to do that.
Joe: I was very intrigued in your chapter in your book there because it says that your software using the TOC method it’s not about forecasting. You kind of disdained forecasting and don’t look at it as a reliable way of managing your supply chain.
Amir: Yes that’s right actually it is a known fact for people who has been for some mathematical education. My undergraduate studies are in computer science, but it is very much a pull, an emphasis on mathematics. Any mathematician can tell you that forecast is something which is an entity which is sure to make a lot of mistakes on the way and I am mentioning most of them in my chapter.
Joe: What is the TOC way then, if it is not based on forecasting?
Amir: It’s based on reacting very fast to what actually happens. So, if I take something from a Lean Kanban system that it is more or less like a Kanban system just outside of the manufacturing floor. And we do change the size of the Kanban according to the supply and demand.
Joe: In Lean the first things we think about is pull and pull distribution. Would you explain pull distribution bit to us?
Amir: Of course what generally people do and there is good reason for it, they try to push the entire inventory to the selling points. That means to the shops that sell the product and therefore when; for example fashion when a new collection comes out they push everything to the stores. Which means if something is sold out they cannot really replenish it because they already finished producing it and until they do a new batch it will take a few months. What pull distribution means we keep a lot of it in a plant warehouse, we put on a little bit in the shops and when that’s sold it immediately gets pulled to the store. Which means if one store is selling and others is not then we get a much better sales ratio for that model.
Joe: When you are looking at that it takes a certain amount of time to develop this because you have to acquire history for your buffer stock correct?
Amir: Not really what we do is we start with what we call an initial guesstimate. Which means it is an estimate, but it is also a guess because you don’t have to be a particular in that. What we have later is a system which will adjust the inventories stock size according to actual supply and demand. So, what you really need to have is only like an initial guess somewhere in the area of it and then it will readjust itself in anyway within one month, two months depending on the replenishing time.
Joe: We talk a lot about Drum-Buffer-Rope, could you give an overview?
Amir: Definitely a buffer is a word in English I am not sure what it means exactly, but what we uses it for is protection. Eliyahu Goldratt put it in “The Goal” that’s at the time there was a time buffer which means an order due to be complete in two weeks, and actual touch time let’s say is two days. Because usually very small compared to the totally lead time, and the rest of it is generally protection time that we need in order for to have other orders processed then most of the time the order is on the floor as a queue time. So, that’s a time buffer and we got into manufacturing plants and the distribution chains that have stocks. And then we thought the best way to treat it is by creating an inventory buffer. That means that to cover consumption, we need to have a certain amount of units we call that buffer. When we discuss buffer management it means that we see how many of those units has been consumed we call it a buffer penetration.
And we divide into colors as well, so if less than one-third of the buffer has been consumed it’s green. Then between one third and two third its yellow two-thirds to a 99 percent would be red and if all the buffers consumed then we consider this black or a stock out. Generally this is a buffer for stock which usually is when I am talking about sometimes there is also buffers in time and in project management it also buffers in time.
Joe: In lean principles, you are looking at pull. If there is a buffer protecting you, aren’t you just having a lot of inventory anyway? The buffer determines how efficient you are for immediate replacement. I have this giant warehouse of parts to fill as my buffer. I always have inventory, right?
Amir: OK, so maybe the original meaning of protection is not very valid anymore because as I said the view change from time based buffers to stock-based buffers. Now in time-based buffers the idea was if you are if you are in green then you are very safe if you are in the yellow then start monitoring and know what you are going to do if you are in red. If you are in red then you are expediting to make sure it goes out in time before its black. If we are talking about stock buffers then it is a different ball game because what we do actually is we monitored the amount of stock at the site or at the warehouse or at the shop. We look at the current stock verses total amount of buffers that we put there. The total amount of buffer is what’s on stock, in transport, and in the production. And the pull means as soon as one unit is consumed from those totals then a new order should be triggered. What we expect to find at the site is usually being in the yellow zone meaning if we are in red then we are in a danger of going in a shortage. If we are in green then we are actually having two month’s stock, we are too safe and if we spend too much time in the green then we would recommend decreasing the buffer.
If it’s too much time in the red, then we would suggest increasing that buffer. That means that, actually, a buffer is no longer protection, but it’s more the correct amount to keep covering for supply and demand.
Joe: So it’s really managing, making the buffer dynamic and being your control point that you’re talking about?.
Amir: Yes. Yes, I am.
Joe: What differs from this than a typical Kanban system?
Amir: Basically, a Kanban system would do the pull replenishment thing. I mean, as soon as one unit is consumed, you’re supposed to issue a new order for the missing one, or two, whatever. That can be applied for a supply chain as well, and that’s part of the TOC solution for managing supply chain stocks. The other and I think more crucial, part comes when we discuss the dynamic buffer management, or changing the buffer, which is different. The Kanban system will usually be set up, and then from time to time, you’ll do an audit, and change those setups, but it will not be a daily procedure. In TOC what we do is a daily procedure to monitor those quantities, and changing them only on the basis of what needs to be changed.
I would also like to get back to your question about whether we’ll have too much inventory, and say that, just from my experience, and all of the experiences that I have heard about TOC implementations. That inventory went down by an average of 40 percent, from the initial values, while increasing availability.
Joe: The buffer is really what manages, what reduces, that inventory amount.
Amir: OK. So, that’s the process in which we’re looking at the zones in which we’re at, at each buffer. If we see we’re at the yellow, then we’re fine, or, half of the buffer sits at the site, and half of it is on order, somewhere, or in transit. When we talk about dynamic buffer management, it means if we go into the red, and we consume a lot of time there, and that depends on the specific algorithm that the software company, or a person who started it did. Then as soon as we accumulate too much red, it’s a signal to increase the buffer. If we accumulate too much green, being in the green too much, or the top part, then we suggest decreasing that buffer, or that queue in the interim.
Joe: If you’re software follows these principles, it’s a very visual system that we’re seeing…
Amir: It’s definitely very visual, very colorful, a very happy system.
Joe: When you’re sitting there looking at some, manage these inventories, now you’ve got slow movers and fast movers. So, are you increasing inventory, or do you just manage it more in the fast movers, and are you decreasing inventory, or just managing it less in the slow movers?
Amir: Well, the right amount of buffers should reflect the consumption, and, as I said, taking into account both demand, which is consumption, and supply, which is manufacturing, and shipping. So, a fast mover and slow mover would just dictate different sizes of buffers. The slow mover would sometimes have a buffer of one, meaning as soon as I sell one, then I replenish it. I don’t expect to sell two within that replenishment time. A fast mover would mean, I have more than one, possibly even hundreds of thousands of this unit, in order to cover for demand. So, the only thing the fast mover and slow mover would dictate is the size of the buffer.
The more interesting thing I think is dealing with uncertainty. Which is if we take again fashion business, because maybe it’s the most uncertain there is, that I know of, of managing stocks.
You put a new model, or a hundred different models, and you don’t know which ones are going to hit. You know from statistics maybe 10 percent would be fast movers, 50 percent would be slow movers, and another 40 would be somewhere in the middle. But you don’t know that when you make that initial order to the producer. The beauty of this system is, it enables very fast reaction to what happens in actual life.
The fast movers would be replenished very soon, very early in a short amount of time. Then you would accumulate too much red, increase the buffer, maybe do so a few other times, and then reach a high enough buffer. The slow movers would not consume. That means you’ll be too much in the green, and you’ll decrease those buffers. Overall you’ll have the right amount of buffer for each depending on its consumption rate.
Joe: Talk about something upsetting things. I think I read a reference where you talked about the death of Elvis Presley, or the death of Michael Jackson, and all the CDs were gone. Everything’s gone.
Amir: Maybe I should have picked a happier example. A good article in a newspaper is a happy example of, “Well they are saying the new soup that we got is excellent.” So suddenly sales go through the roof.
Joe: Maybe your product is shown on national television one day, or in a movie spot. I think that’s what happened with sunglasses with Tom Cruise in Top Gun. I think Oakleys was it, but they took off through the roof on that. When something like that happens is there an advantage of having this type of management system in place? Or it’s just like any other. You’re just guessing and by golly, you just react to it.
Amir: Well definitely there’s a plus, there’s actually two pluses. One, even without human intervention, the system will adjust itself very quickly to that event. So, you’ll still encounter some shortages or stock?outs, because there’s nothing that can supply so fast as demand. But very soon you’ll have the right amount of buffer. The other one is if sales go up very quickly, the system will signal you that it has, and that means you’ll not increase by the usual amount, but you’ll know that, “Well we’re facing something tremendous, it’s real.”
So, that’s something the user will have to do, that’s something the software cannot do. Which means the user will have to go and verify it’s real. He can increase the buffer by 200, 300 percent. That is signaled out, and he can see the signal as soon as he opens up the system. That definitely speeds things up, as relating to dealing with it.
Joe: So by having proper management techniques in place at the beginning, it allows the reaction and the probably the efficiencies of managing the reaction much better.
Amir: Definitely. I think what TLC brought to the table, is mainly the focus on planning just about right. Not trying to plan too much, but enabling a lot of tools for execution control, to change things while we go and adjust to the reality.
Joe: What’s the difference in managing the buffers, then bottlenecks in Theory of Constraints? Can you bridge that for someone that is, maybe, a Goal Disciple, and thinking of TOC just in bottlenecks, and bridge that gap for me, a little bit?
Amir: I’ll try to do it, Joe. I’ll try to. In The Goal, Eli Goldratt discussed a manufacturing plant, and at that time, that was 1985, a lot of the manufacturing facilities in the world faced much more demand than they could satisfy. That means that the constraint, what we call theory constraints, is not necessarily bottleneck, but the constraint is internal. If you increase the capacity of the constraint, then you increase the capacity of that manufacturing plant to sell more. Nowadays, it seems to be vice versa, meaning, demand is the bottleneck, is the constraint. Which means marketing is the real constraint of the company, usually, or the market.
So, what we do is, and we actually in today’s project we don’t first look for the constraint. We first try to make, to do the performance as best as possible. And then if something is coming in the way of it, then we treat that one as the constraint. That maybe is the bridge between what used to be a bottleneck within the production facility, and nowadays, in most of the environment what we see is, a market constraint, or the market is limiting them.
When we talk about distribution environment there was never a bottleneck in the first place. So, we can discuss focusing, which Eli Goldratt claims is the base of TOC, is focusing. So, it used to be finding the constraints, but nowadays it’s, maybe, let’s focus on the right things to do, and what not to do.
Joe: This kind of brings us into certain things. Now, Eli Goldratt had a recent book that really kind of talked about supply chain management. Did he not?
Amir: Yes, yes. It was called, in the true Eli Goldratt way, it was called, Isn’t it Obvious. Which means, after you read it, it’s quite obvious what the solution was. Of course, before that, you couldn’t guess. But, it describes a retail chain and management of it and how they faced the challenges of managing stocks, and it actually, yes it is exactly about what we’re doing in Inherent Simplicity.
Joe: You’re a software company, but there are other kinds of new knowledge you’re developing. What is that?
Amir: What happens is that, when we started with Eli Goldratt working on Viable Visions, and then we found some clients of our own that were TOC enthusiasts, and wanted to have TOC software, and they called us, and we faced different problems that had never been solved by anyone, not Eli Goldratt, or not any other TOC consultant. Since we do have the knowledge, and we encounter a lot of different organizations like that, then we actually had to develop it by ourselves. We try to spread the knowledge as much as possible as I put this chapter in the book. I try to talk to as many people as possible, but still some of it is embedded in the software, obviously. So, we became part of True Knowledge Developer in TOC.
Joe: Now, do you work with other consultants?
Amir: Yes, yes, we do. We have partners all over the world. As I said, one of them is Eli Goldratt, which uses our software exclusively for his projects. But, we use other TOC consultants around the world. We have consultants working for us with us, in Lithuania, and in Russia, and in Japan, and South Africa and Columbia and Mexico, and the U.S., and Italy, so, different regions of the world.
Joe: There is software for this, there’s software to do this, and everybody’s got a collection of software on their bookshelf behind them. What’s the key to using software? Why would I want to use software to manage my supply chain?
Amir: Well, in my view, you would only want to use software if it helps you… Well, there are two levels of it. One of them, if you’ve got software that will help your bottom line, and then definitely, as a logical person, you would want to use that. The other thing is, if that software makes your life easier, which I think, most of the software today is trying to go by that niche. We’re trying to target the better bottom line results. Obviously, there’s a whole bunch of software that doesn’t give anything, but it’s a question of how do you know that, prior to buying it.
Joe: I think the biggest problem, most people have in software, is that learning curve, and being able to pass it through the organization for everybody, to make it usable for everyone. You always have that one guy, in the one office, that’s a master at something. But do you really spread it through your organization for them to be able to use? Does your software provide, let’s say, a dashboard for everybody to just look at and provide input, maybe, or how does that work?
Amir: Well, we tried to create some kind of dashboard which is, obviously, different for everyone because the General Manager, or the CEO, of an organization, would like to see other things then the Regional Director, or the Plant Manager. We tried to create some kind of dashboard for each of them to see. Actually, what I’m putting a lot of emphasis on is, doing the software as simple as possible. So, Inherent Simplicity is not just a passphrase, it’s something that I try to actually do in the software. We’ve reached the stage in which, I think, even from the beginning, we never had an instruction session that went over two or three hours, on how to use the software, which I think is an achievement.
Joe: Now, do you do them virtually with people, now, with webinars and different things?
Amir: Not really. What we do have is a local implementation network. So, each region, in which we have clients, would have a local, either employee or partner. That is in charge of first of all, translating all the material, translating the user manual, the labels, the software itself, is in the local language, and do all the training, all the upgrades, et cetera. So, we provide local support, and of course, we have, in our headquarters, a second tier of support for those implementation partners.
Joe: The difficulty when you have a new product, or not necessarily, maybe, a new product, but a product that is not mainstream. Let’s say, we could use Excel Spreadsheets for managing our supply chain. But when you have a particular product, you have to manage the buy?in process: Do you have to get people sold on theory constraints before they want to use your software?
Amir: That’s a really good question. I mean, I’m often thinking about what would be a better angle. Is convincing them that TOC is the way, or just convincing them to try to use the software and then see the benefits. Part of it is, we’ve put the kind of simulation in the software, in which, we’re telling them, give us last year’s data of sales, and actual stock positions, and then do a simulation saying what would happen if we would use an art or symphony on your environment. And then we can compare the two and usually we find quite a difference between what really happened, what they could have achieved with TOC. And that I think is a good way to show how powerful TOC is, and how simple it is.
Joe: I always think that’s the best thing. It’s just if it is simple people understand it.
Amir: Yeah and that’s part of my problem with forecasting algorithms. I mean they could be very sophisticated. But in the end the person who would activate the forecast. He would see a number. Let’s say he punched in a thousand different variables and at the end they got fifty. What do you mean by fifty, I mean I don’t think it’s going to be fifty, I think it’s going to be eighty. So I’m going to plan according to eighty, and then there goes the forecast. I mean, it could be an excellent forecast but few of us understand how it works. So it’s not any good. That’s why I vote for simple solutions in which he can understand that. And that’s closer to where he would optimally be, then the other way around.
Joe: You mention product portfolios and that being applied with your software in the TOC methods. Can you kind of explain that a little more to me?
Amir: Definitely, let’s say you’ve got two different products. And I’m not getting into pricing too much. We’ll get there later. But let’s say one is sold for price X and the other for price Y. Now what I’m saying is, depending on the volatility of demand and supply. We would eventually reach, using TOC principles, a buffer for both of them. For product A and for product B. The buffer is actually the price we need to pay in order to have the ability of this product to the market and selling whatever we sell. That is if we take both of these escrow sales and divide it by the cost of the buffer. That means we have the return on investment that we did. Assuming of course, that it’s not a strategic product, taking it off the shelves will not harm other product sales, etc. So, if it’s just not linked to any other products. Then it will give us, what is a better product. That means if we keep it at a certain location, then in another one were not keeping it. Then we would like to try to sell it there. Because it’s making more sales compared to the shelf space it occupies, and the money investment it requires.
Joe: You talk about their frequency of replenishment and I’ve always thought that constantly, I mean, the best system ultimately is not having one thing on the floor: Selling it, replacing it immediately, all the way down the chain continuous. Which isn’t really possible, but ideally, that’s what you strive for? Correct?
Amir: Yes, it is, and actually in some industries it is. If you go to a lot of countries, and look at pharmacies, you’ll find out that most of their pills, they have one of each, and as soon as that is consumed, talking about daily replenishment, in some of the pharmacies, it’s six times a day replenishment. So, that system does exist sometimes, and yes, that would be the ideal. But, living in our reality, sometimes it’s not, and therefore we need to maintain a buffer. Obviously, we try to do it as frequently as possible. A lot of the environments are not doing it as frequently, as frequently as possible, because of some inertia in thinking, or whatever.
They’re saying, maybe, it’s the trucks, maybe it’s containers, but a lot of times, it’s just excuses for working a bit more, loading those trucks, putting some variability of excuse there, then really the lack of trucks.
Joe: I think a lot of times, people are scared that as they reduce their inventory levels, reduce their buffer levels, and what if there’s a hiccup. What if Murphy does take place and you’ve created a situation that you can’t replenish it?
Amir: That’s a good fear to have, and that’s some of the managing risks. Sometimes we do have a strategic product, and let’s say, we import it from China, and the Chinese Government is now contemplating about putting a tax on exports. Obviously, we need to take that into account, and create some kind of safety stock. But we need to understand that’s really a safety stock that means we’re paying capital to have that opportunity.
So, I classify that as strategic decisions by the management. It has nothing to do with any algorithm, but it’s part of risk management. They need to understand what would happen if they lacked that, and what would happen if they have too much of it.
Joe: Going back to the color codes like you said, it’s very simple to understand. If I’m sitting there always in the green would that be an indicator that I could possibly reduce it?
Amir: Not necessarily. That’s why I’m saying, staying in the green for too long. Because let’s say, you have a spike pattern of consumption, let’s say, one big client comes in every two weeks and consumes a lot of it. So it doesn’t make sense just because you’re green between those two weeks, doesn’t mean you can decrease the buffer. So, there’s some kind of perimeters that you would like to use in order to avoid that kind of situation.
Joe: It’s my observation and having the colors would definitely allow me to see trends very easily.
Amir: Yes, yes. Very much. You’ll see that, if you have a green for a long time, then that’s definitely a trend, but green for one day, doesn’t really tell you anything.
Joe: Again, we go back to that word that we used at the very beginning of this podcast is, dynamic. I want to have a dynamic buffer.
Amir: Definitely you’d like to have a dynamic buffer, because reality is dynamic.
Joe: So, if it’s staying in a certain area for too long, I’m really not managing it?
Amir: That’s right. That’s why, as soon as you’ve passed that threshold saying, you have too little or too much, then you need to correct it. We’re just saying what should be done logically, and if people would think about it, they will find out that that’s what they wanted to do. I think, this is why Eli Goldratt chose the title for his book, Isn’t It Obvious. I mean after you’ve solved that solution, you said, that’s exactly what I would do if I didn’t have all that nonsense they put in my head during my MBA studies.
Joe: What have you seen really change the supply chain management in recent years?
Amir: Well, definitely that idea of a plant warehouse, or a central warehouse, is something that’s really catching on. At least I can say in Israel, a lot of chains understood the potential of having a central warehouse. Before, they would have a hundred shops, and each one would order from Europe, or the U.S., by itself. Nowadays they’re thinking, OK, let’s have a central warehouse, it aggregates all of the demand, and then orders for us, one batch for all of us. That works better. It allows much higher frequency, so that’s a move in the right direction. Historically, there’s been a plant warehouse, only it went the other way around, and nowadays it’s coming, I think, back into a central warehouse.
That’s all of Eli always claimed, that you need to understand the logic of doing something, and then do that. I mean, sometimes you do the right stuff just because that’s the trend these days. In a way, I’m happy that this is the trend these days. I mean, I don’t think they know what they’re doing, but they’re doing some of the right things.
Joe: It’s really kind of the hub and spoke, FedEx, type thing, or maybe even the Wal-Mart type thing of distribution.
Amir: I’ll tell you one thing, I’m sure that Wal-Mart is using TOC as the basis of their replenishment system. By seeing how they operate, by seeing how they think, I’m sure they read some of Eli’s books.
Joe: I’m sure of that, too, that they have. The other thing that I look at is as you move to a central warehouse, do you reduce the variation in supply and demand?
Amir: You do, definitely, and that’s part of what I’m saying. Look, the forecasting model works. It does work; it just creates so many problems later, that it’s a problem. But, anyone that deals with forecasting will tell you that, the more aggregated the demand, the better the better the forecast is.
Joe: But am I just shifting problems up the chain, versus down the chain?
Amir: Definitely not. What you’re creating is a way to use your inventory in a better way, in a more responsive way to the market. Which is common logic, if you think of, OK, but the client comes to the store, he doesn’t come to the plant warehouse. Yes, but if he comes to the store, and I have enough to cover for immediate demand, I can replenish from the plant warehouse, and if I push everything, then I can’t replenish. Some stores would have stock out, some stores would have overstock, and cross-shipping, is a real pain, and most of the companies don’t do that.
Joe: Is there places that this type of management doesn’t work? That you would say, this industry, or you’re doing it this way, which really complicates problems, that I don’t see how buffer management and TOC supply chain would work?
Amir: Yes definitely I would not say that TOC doesn’t work I’m saying that the current solution would not work. It is part of why I added a few chapters or mini-chapters in my chapter there saying special environment like when we have short shelf life product we need to tweak some of the solution. What do I mean by twitch is just examining the logic seeing what assumptions were good for the general case and not a specific case. And then seeing what changes we need to do. But what we always should do is follow the logic follow the cause and effect logic, saying if that’s the case that’s what’s done then we need a different solution. And for example you mentioned some of the old business the old business’s not something that will work in this solution. Not as easy as it is because everything in the art is one so what’s the good in replenishment. And you discussed part of the solution which is well I’m not buffering specific item I’m buffering a group of items that represents some clients need, like lamps. If I order a new lamp it’s not going to be the same as the one that was sold right? This is an example of an environment in which you need to change the solution in order for it to make sense.
Joe: Is there something you’d like to add to this conversation maybe I didn’t ask?
Amir: There was one thing you asked back before and I had some more to add to it. You asked when we maintain buffer how do we make sure that some crisis happens and we don’t have enough. I wanted to add that some of the theories also just the fear of reducing inventory. When we say that you should reduce inventory, we were too much in the green and not really a strategic threat. We do have some SOPs or standard operating procedures and one of them is when we go to a new client or let’s say one of our clients have a new clients they’d like to replenish to. We tell them we tell ourselves people when you talk to the client don’t say we’re going to reduce the inventory. That moment we receive a signal to reduce the inventory. Let’s take all the things that we should reduce and put them in a contained space, we’re going to mark them. After a few months, we can show them OK now you see those ones you haven’t used them all. That means you don’t really need it then we can build confidence in the system. That’s something we use in the sales marketing side in order to enhance the confidence that the client has in the solution.
Lean Sales and Marketing: Learn about using CAP-Do
Special Marketing with Lean Book and Program offers on Facebook