How to Understand DDMRP as Yet Another Repackaging of JIT and Lean

Executive Summary

• The Demand-Driven Institute has created the DDMRP concept, which began with Orlicky’s 3rd edition MRP Book’s perversion.
• DDMRP is very simply a repackaged of JIT and Lean.

Introduction

DDMRP is an adjustment to supply planning proposed to improve the outcomes of supply planning systems. While it has MRP in the name, DDMRP is significantly closer to Lean approaches than MRP. In fact, in reading the literature on DDMRP, it is difficult not to be taken back to the claims made by Lean. DDMRP claims to be influenced by what seems like six different pre-existing philosophies. DDMRP makes big claims, and I will begin by analyzing a book that helped kick off the DDMRP craze. Then, I will analyze proposals made about something called DDMRP and review the likelihood that what DDMRP proponents say is true.

Our References for This Article

If you want to see our references for this article and related Brightwork articles, see this link.

Ridiculous Orlicky’s 3rd edition MRP Book

Joseph Orlicky was one of the originators of MRP. Orlicky wrote the first book on MRP in 1975, when MRP was becoming known. Orlicky died in 1986. In 1994 the 2nd edition of Orlickly’s MRP was written/adjusted by George Plossl. I am not a big fan of the 2nd edition of books written by authors other than the original author. This is an attempt by the publisher to extend a popular book to a new group of buyers. However, George Plossl is probably my favorite inventory management author. George worked with Orlicky, so the second edition was consistent with the first one.

Curiously, the 3rd edition of Orlicky’s Material Requirements Planning book was published in 2011. That is right; we are now releasing books 25 years after the author dies.

This book was not written by anyone who worked with Orlicky but by Carol Ptak and Chad Smith, who wrote DDMRP.

When I read the 3rd edition of Orlicky’s MRP book in 2012 (according to my Amazon account), it did not appear to have much correspondence to the 1st or 2nd editions. Furthermore, I don’t think Orlicky would have agreed with much of the material presented in the third edition of “his” book.

Therefore, the 3rd edition is only “Orlicky” in name. It is neither written by him nor inspired by him.

The Most Accurate Way to View The 3rd Edition of Orlicky’s MRP

In my view, Orlicky’s MRP 3rd edition is merely a way for JIT/Lean proponents to worm their way into MRP and try to get companies to change their MRP systems to conform with JIT and Lean principles.

To explain why this is so problematic, we need to revisit JIT and Lean’s history briefly.

The History of JIT and Lean

JIT was first introduced outside of Japan in the 1980s. JIT was a highly inaccurate presentation of parts of the Toyota Production System. At the time, Toyota (and other Japanese manufacturers) attained quality levels that no other automobile manufacturers from other countries could match (and continue to maintain). The TPS was developed during the postwar period and is based on deficient levels of waste. However, it was brought over to countries outside Japan by consultants, and consultants grossly oversimplified the TPS and JIT. Important things that the TPS/JIT consultants left out included the following:

1. Unionization and Empowerment of Factory Workers: Toyota plants were highly unionized. This means empowered workers that could stop the line to keep quality levels high. Consultants knew that US executives detested unions, so this feature of the TPS was entirely left out of the books on the TPS at the time.
2. Supplier Inventory Location: Much was made of keeping low inventories in the factory. However, JIT consultants left out the fact that Toyota suppliers were closely located to Toyota factories. Therefore, the inventory was usually “right around the corner,” although not on Toyota’s books. When the math did not add up, companies that moved to such low inventories based upon faith in JIT consultants experienced reduced production capabilities.
3. Stable Production Schedule: Quite the opposite of “Flexible Manufacturing,” Toyota, in the 1970s and 1980s at least, had a steady production schedule for one month out. This meant low material variability and a strong ability to coordinate supplier deliveries. Knowing that this would infuriate US executives who like a very unpredictable production schedule (outside of manufacturing), the consultants “left that little part out.”

JIT Definition or JIT Meaning

The JIT meaning or JIT definition is the reduction of inventory so that the new inventory replenishes the stocking level right before it is to be depleted. There is some debate as to when this happens technically. After safety stock has been partially depleted, there are many different JIT practitioners, and they have differing opinions.

The JIT definition or JIT meaning has several sub-areas. These JIT definitions or JIT meaning include just in-time inventory and only-in-time manufacturing, JIT manufacturing, or JIT production, which we will discuss further in the article.

While many people do not know the specifics of the JIT definition or JIT meaning, most do know that JIT results in lower inventory. But what is not at all well known is the method by which JIT proponents arrive at their stocking-level proposal is philosophical and based on the anecdotes of experience in inventory management from Japanese manufacturers.

Just in Time Inventory or JIT Inventory

In time, inventory or JIT inventory minimizes inventory based on the concept that a smaller stocking level can be maintained and an increase in delivery frequency performed. Quantification of the extra costs of the JIT inventory system is not part of the JIT method. The just-in-time inventory system or JIT inventory system is based upon philosophy, not based upon developing a body of evidence to support the move away from traditional inventory management.

JIT Delivery and Higher Ordering Costs, Delivery Costs, Receiving Costs and Put Away Costs and Delivery Frequency

Extra costs of the JIF inventory system include higher ordering costs, higher delivery costs, higher receiving costs and..

• Higher Ordering Costs
• Higher Delivery Costs
• Higher Receiving Costs
• Higher Put Away Costs

Put away is the process of moving and stocking the inventory at its stocking place. Put away follows the receipt of the goods.

One of the primary reasons why JIT proponents don’t calculate the ordering costs, delivery costs, receiving costs, or put away costs is that the overall costs would necessarily look higher if this were done. This is because the inventory carrying cost is far lower than all transactions that make up the stocking level.

Delivering in small quantities with high delivery frequency is called JIT delivery. Many shipping companies specialize in JIT delivery frequency to meet the market demand. JIT delivery can be driven by JIT or Lean thinking, or it can apply to factories in congested areas that lack sufficient stock space.

The EOQ formula produces an order quantity based on a trade-off between inventory holding and ordering costs. In such a formula, the ordering cost costs, delivery costs, receiving costs, and put away costs could all be placed into the ordering cost category. JIT inventory management does not support the concepts of such mathematical determinations.

Just in Time Manufacturing, JIT Manufacturing, or JIT Production

JIT primarily came from Japanese manufacturers, through US consultants, and to global companies. However, the greatest JIT craze was probably in the US. Just-in-time manufacturing, JIT manufacturing, or JIT production means JIT applied to manufacturing. Therefore, JIT is a manufacturing inventory concept that came from factories and was then applied to the overall supply chain.

Just in Time Supply Chain

In time, the supply chain applies just-in-time manufacturing, JIT manufacturing, or JIT production principles to the supply chain, which means inventory management outside the factory. In time, the supply chain supports seeing the overall supply chain as if every stocking position and every stocking level is a short lead time product location that is no different from a manufacturing facility that is lucky enough to be continuously able to be replenished under the Toyota inventory model.

JIT Inventory System

A JIT inventory system is simply a method that applies JIT. Any supply planning or MRP/ERP system can be made to operate under JIT principles. This typically results in the system being set to work on consumption-based planning using methods like reorder points. JIT is opposed to forecasting philosophically, considering it too unreliable. The problem is that while this may apply to a stocking level or stocking position, it is impossible to apply consumption-based planning in all circumstances. And the longer the lead time.

Misinformation on Inventory Conceptually Because of Lean or JIT

Lean is just rebranded JIT. Since at least the 1980s, a philosophy of keeping low inventories has gone by various names. At one time, it was JIT, and then it became Lean. JIT was based on low inventories that the Japanese were able to keep. But without understanding, that Japanese companies work more collaboratively than US companies. Second, many industrial areas in Japan have suppliers close to their customers. The US does not have the same supplier network setup as Japanese companies. Also, if a supplier maintains your inventories, then the overall system inventories are not lower. This distinction was left out of most of the explanations provided to US companies by JIT/Lean consultants.

Lean is primarily a philosophy based on taking a concept from production planning that works in specific circumstances. Lean makes sense when it uses an analytical approach to segment the product location database and converts some unforecastable product locations to reorder point planning.

JIT as Esoteric

With JIT consulting, the game’s name sounded leading edge and esoteric — not communicating the true nature of TPS and JIT as practiced in Japan.

JIT eventually developed such a bad reputation that JIT consultants and consultancies knew they had to make a change. Drunk on their own Kool-Aid, the answer would not be to include more accuracy in their consulting. That is to make it representative. Instead, they opted for a name change.

JIT became Lean — and the unsupported claims resumed. It resumed to this day.

Today you can receive all Lean certificates, primarily imposing unrealistic manufacturing and inventory management proposals. I have never seen any of these certificates, any Six Sigma plaque, Lean plaque, or other merit badges to have any relationship with better company inventory management outcomes. In operations, having them often makes the difference in getting or not getting jobs.

How JIT and Lean are Incorporated into Systems

JIT and Lean consultants like using Japanese words. They like saying Sensei and Kopai. They want to talk about Toyota….as often as humanly possible. Lean has its rituals and, in this way, is quite similar to the Crossfit cult.

However, JIT and Lean proponents don’t like to address that JIT and Lean capabilities have resided in software since supply and production software was first introduced before that in inventory formulations. I propose using some of these approaches, as well. They are outlined in my books Lean and Reorder Point Planning and Multi-Method Supply Planning in SAP APO (where “Lean” methods and forecast-based planning are mixed by product location in SAP and multiple forecast-based procedures are mixed in).

I cover how to assign product location combinations to reorder points, min-max, etc.. based on a concept called forecastability. I have a forecastable/non-forecastable formula at the following this How to Understand Forecastable Unforecastable Formula.

The matter is relatively simple.

• Some items can be reliably forecasted — and for those, using a forecast-based supply planning method makes sense. MRP is one of these available in the software.
• It makes sense to use consumption-based methods for items that cannot be reliably forecasted.

All of this can be set up in supply planning systems. It does not require donning a kimono learning Japanese, or an APICS certification. It requires no colored belts of any kind. It is the application of basic inventory management knowledge.

DDMRP proposes using MRP as a non-forecast-based planning approach — which makes little sense.

To explore why I have included quotations from several articles on DDMRP.

Articles on DDMRP

The article Demand Driven Material Requirements Planning (DDMRP) on Linkedin makes the following points.

“DDMRP is a revolutionary planning method that is designed to meet the needs of the modern day market. Compared to MRP, DDMRP generates orders based on actual sales orders, rather than forecast.”

That does not make any sense. This is because MRP is a forecast-based planning method. One can, of course, decide only to feed an MRP system sales orders. But as covered in the book Replenishment Triggers: Setting Systems for Make to Stock, Make to Order & Assemble to Order, most companies cannot move to order environments. The lead times just don’t work out. Therefore, right off the bat, this explanation of DDMRP primarily pitches fools gold to executives.

“This allows for much higher customer service levels, lower costs in expedite, and the right levels of inventory.”

Why is any of that true? Why does basing MRP on sales orders allow for higher customer service levels? If, as in most cases, the environment cannot be made to order, service levels will decline. The company will experience stock-outs and lost sales.

Basing MRP on Sales Orders

Secondly, how does basing MRP on sales orders reduce the cost of expediting? It would be the opposite. Also, it does not lead to the “right inventory levels.” This will only be the case if the environment is made to order.

“MRP hasn’t changed since its inception and this is where DDMRP was designed to tackle all the critical issues in order to maintain a healthy production environment.”

Well, the math may not have changed, but modern MRP systems are a lot better than the MRP systems that were first introduced. When MRP was first introduced, it was run off of computer tape. That is, MRP pre-dated disk storage.

“Today, there are more complex and planning scenarios than before. The past is no longer a predictor of the future.”

This all sounds quite sexy. But this has more to do with companies increasing their SKU count (with supermarkets in the US having roughly 40,000 to 50,000 SKUs. This lowers forecast accuracy. But even though products are becoming less forecastable, it does not mean that DDMRP is the answer. The problem is that it is not feasible to run planning off of sales orders.

• “Achievement of 98% customer service levels
• Revenue maximization
• Inventory reduction by 40%
• Expense minimization
• Cash flow”

This is where the author moves into exaggeration (Hasso Plattner style exaggeration, which is a higher level of exaggeration than Larry Ellison or Steve Jobs).

No Inventory Method is Designed for 98%

No inventory method or technique on Earth is designed to ” achieve 98% customer service levels.” The service level achieved depends on the input and the situation.

Does DDMRP maximize revenue? It’s hard to see how that would turn out to be true. Companies that only base MRP on sales orders will be in for a world of customer disappointment.

Why does 40% reduce inventory exactly? Why not 35% or 45%? This seems directed towards hooking executives by telling them what they want to hear.

“DDMRP is also a new way of planning and control, which shifts from a forecast driven model to a sales order driven model. In MRP, requirements are calculated based on the forecast, which eventually becomes irrelevant as time moves on.”

This is a highly uninformed statement. Forecasts will, in almost all circumstances, have an error.

• Low errors are good.
• High errors are bad.

However, this does not mean that forecasting is invalid.

Actual Percentage of Make-to-Order Products?

And once again, as customers demand products more quickly than they can be produced, make-to-stock environments are the most common environment to be found. Only a tiny percentage of business is true to make to order.

Make-to-order means no procurement orders are created until the sales orders are received. It does not mean that stock is maintained until an order is received, and then manufacturing begins. That is called assemble to order and is a different thing altogether.

Probably less than 5% of businesses can work this way. Defense contractors being a perfect example of this. Construction projects are another.

The Article “Why DDMRP”

The article Why DDMRP Is A Necessary Condition For Industry 4.0 To Deliver On The Promise makes more bizarre contentions about DDMRP.

“This vital element is the use of the Demand Driven Operating Model and the related planning methodology Demand Driven MRP (DDMRP). This is currently the only approach that allows to effectively synchronize supply and demand across complex and volatile supply networks.”

Let’s not hyperventilate too heavily DDMRP proponents! I know there are projects to be sold, but let us keep it within the realm of sanity.

So, according to this quotation, only DDMRP can synchronize supply and demand over volatile supply networks. This is quite interesting because MRP is not the most sophisticated matching supply and demand method. Inventory optimization and multi-echelon planning methods are far more advanced than MRP.

MRP Versus More Advanced Supply Planning Methods

Unlike MRP, it has the math to compare stocking locations across the network and can set stocking positions while cognizant of the stocking locations around the stocking location. Overall, it is entirely inaccurate to say that DDMRP is “the only way” to connect supply and demand.

“For instance, one of our clients recently reported to us that from the moment they have changed their distribution planning using DDMRP they completely eliminated shipments between distribution centers. This used to be a major supply chain expense before, due to inventory being in the wrong place. During the same period inventories went down by 20% and service levels improved. Meanwhile, order stability achieved perfection: not a single supply order has been changed once placed to the sourcing plant.”

This anecdote could only be true if this company is a make-to-order environment.

However, if this company is make-to-order, why was it performing redeployment in the first place?

Why is Inventory Carried…..Again?

The authors in DDMRP seem continually confused as to why inventory is carried in the first place. No one wants to carry inventory. But that is what the lead times companies face are required to do.

This misunderstanding extends to comments made in the article.

“Thanks Patrick for This excellent article. … 85% of forecast accuracy means that at least 15% of mistakes are propagated throught all the supply chain ! Is it good enough to reach more than 98% of service rate ? … probably no.”

Once again, unless you can hit 99% to 100% service levels, forecasting wastes time because of “propagation,” according to nascent DDMRP experts!

In another article titled When will SAP include a DDMRP solution in the existing supply chain solution? The proposal is that SAP must offer DDMRP. This is an attempt to move MRP into the software before it is proven as an approach.

How To Understand Trendiness in Supply Chain Management

With terms like JIT, TQM, Lean, B2B marketplaces, Kanban, and optimization, supply chain management is filled with trendy concepts that influence decision-makers (a strangely high percentage of Japanese origin for some reason).

In fact, supply chain management has been remarkably trendy for an area of study that is supposed to be more of a science than an art.

I have previously described that approaches applied to supply chain software frequently do not have to pass any logical test. As I stated in response to a comment on-demand sensing being a method to primarily fake forecast accuracy:

“One consultant I was working with stated that company XYZ was reported to have success with the approach. I had just come from that exact company, and my experience with them was they neither their executives nor their IT group knew anything about forecasting, and this multi-billion dollar company could not do the most elementary forecasting functions. Actually, very few companies can be used as models for forecasting excellence. Most companies do a horrible job of taking advantage of systems to improve their forecast.

However, if a big consulting company does something, or a big client does something, that seems to be sufficient evidence that other people should do it as well. I think the first question needs to be “does it make sense?” and secondly, “have we tested it?” The fact that a consulting company or a client did this or that really means nothing. Very few executives call in journalists into their office to report that they completely bombed on their IT implementation because they were ripped off by Accenture who lied to them about what software could do for them and this caused them to miss their quarter. This is called reporting bias, and obviously must be adjusted for.”

Illogical Supply Chain Management Trends

Observing the illogical nature of many supply chain management trends was noticed and written about decades ago by George Plossl. George Plossl was very focused on practical and often mathematical approaches to managing the supply chain. Therefore, many of the trends in a supply chain, most of which have failed to pay dividends, must have struck him as strange as they strike me.

“Probably the greatest misconception is that the job of effective planning and control is primarily technical. The literature of the technical societies and the words of a few consultants have led many managers to believe that all they need for control are the right techniques in a system. Overselling sound and necessary techniques like MRP has certainly been a great disservice to hard-pressed managers. Interest in new techniques flares up like fads in clothing and sports. Too many managers seem to believe that they can buy their way out of trouble quickly by adopting the Japanese “Kanban” technique or the Israeli super mathematical “Optimal Production Technology.” Over-simplified solutions to complex problems, like jogging for better health and fad diets, continue to beguile many people unwilling to adopt the necessary changes in life-style so needed for achieving their real goals. Sound planning, effective execution of the plan and adequate control requires more than techniques and computer programs however elegant and expensive these may be.” – George Plossl

In this quote, George Plossl explains the penchant for trends in his consulting work.

Conclusion

The false statements regarding DDMRP come hot and heavy from DDMRP proponents.

1. A Misleading Book: Orlicky’s 3rd edition has very little to do with Orlicky, and I believe Orlicky would disagree with much of it. If the authors wanted to promote DDMRP, they could have done this through their book, which has that title. Why infiltrate a pre-existing book that has little to do with what you are promoting?
2. Accuracy Issues: Statements made by DDMRP proponents are highly inaccurate in the articles that I have analyzed — many of the statements presenting a lack of knowledge about how supply planning systems function.
3. Plain Old MRP Was not Demand-Driven?: The term DDMRP or demand-driven MRP used to differentiate it from plain old MRP is nonsensical. MRP is a forecast-based supply and production planning method. As such, it is already demand-driven. The opposite of demand-driven would be supply-driven. This is where the supply side is the focus of production and distribution. Many environments should be supply-driven. This is covered in the article The Forgotten Supply Driven Supply Chains. However, plain old MRP is not “supply-driven.” It can’t be as MRP, not a constrained method of planning. MRP always runs as if supply is unconstrained (something addressed in a later planning run called capacity leveling). MRP can only incorporate supply limitations through the use of min lot sizing. Therefore, it is illogical to cast “plain old MRP” as something on the supply side. This is what “demand-driven” MRP seems to be doing. A more accurate name would have been sales order-based MRP.
4. But Wait, The Ginsu Knife Also Comes With…: DDMRP has many other areas, including sizing inventory buffers. Upon review, it isn’t easy to see how this adds value to the traditional stock-level setting methods. It is different and, therefore, more difficult to validate, but is it better? Overall, DDMRP has nothing that strikes me as having contributed to inventory management — so why would this area be anything but more unrealized promises?
5. Following the JIT/Lean Playbook of Exaggerated Claims: Much of the exaggerated claims, as well as the idea that forecasts cannot be trusted, is right out of the JIT/Lean playbook. However, after some decades now, JIT/Lean has produced minimal improvement in inventory management. The reason is simple. JIT/Lean proponents are more concerned with making an impression and a “splash” than presenting what is true. DDMRP will end up being simply more of the same.

The problem with DDMRP is it puts forward an illusory magic solution for supply planning. DDMRP requires plenty of adjustments for companies implementing DDMRP, extensive new training, etc..

One of the primary claims of DDMRP is that standard MRP does not work. However, the more significant or central issue is that MRP systems are poorly maintained. We find various irregularities whenever we extract parameters from an MRP or other supply planning system. Since the 1990s, multiple vendors have proposed that MRP was irrelevant and that companies must move to cost optimization, allocation, heuristics, or inventory optimization.

The net result?

The MRP procedure plans the vast majority of products planned in the US and Europe. DDMRP is, in our view, just another in a list of supposedly better supply planning methods that reduces the incentive to focus on the key areas necessary for improving MRP.