Why the Demand Driven MRP Opportunity Exists

The Problems that Demand Driven MRP solves are not new

The problems that DDMRP solves are not new – they’ve been causing serious damage to manufacturers’ performance for many years without even a halfway-effective resolution. And many (but not all!) of the elements of the solution have been known for years, … in theory even if not in a practical form.

So how come the problem hasn’t been solved years ago?

How come Demand Driven MRP (DDMRP) has just emerged (first gaining attention in 2007/2008) as the solution?

In what way is it breaking new ground? How is it able to generate such a degree of leverage on performance – typically, massive improvement – in companies already familiar with advanced manufacturing technique?

A combination of realities make the problems INEVITABLE

The need for DDMRP (i.e. the existence of the performance problems) exists because of a combination of realities that INEVITABLY create the problems.

This combination has ALWAYS  created them.

But changes in the manufacturing environment and in markets mean that the combination is more common today than ever; that extreme examples of the combination are more common than ever; and the problems are even more damaging today than ever before.

The problems – especially the problem of chronic shortages – have mostly gone unsolved, to this day. There has simply not been a standardized solution, and some companies – especially those in complex and challenging environments – are fighting the same battles they’ve been fighting for decades. (As consultants, with some clients we used to check back on the minutes of past Production meetings or S&OP sessions and it’s like reading a circular).

Of course, managers have tried to solve them.

They’ve tried, and are still trying, to have better forecasts, for example. It’s mostly hopeless as a solution, but they don’t know what else to do, and they feel they can’t do nothing, so … the drive continues for the fabled “better forecast.”

They’ve tried flattening their Bills of Material, which is often quoted as a “Best Practice.”  The only problem is, Demand Driven MRP sometimes improves performance  by ADDING NEW LEVELS to the BOM!  Correctly done this can compress lead time, reduce inventory and contribute to reducing shortages. It’s counter intuitive, and it highlights the lack of understanding at a cause-effect level of some of the “Best Practices” advocates.

Many companies have tried to go strictly make to order, and some have succeeded. But when the client tolerance time is 7 days, competitors can meet demands in 7 days from stock,  and even after tremendous improvement efforts your manufacturing lead time is 3 weeks, you’re going to lose sales – simple as that.

Personal, custom written spreadsheets and Access databases and reports are examples of common workaround; whiteboards and ink-and-paper notes clipped to computer screens are even more common! And they CAN help  – indeed, they can save a company’s rear-end.  But they should be completely unnecessary if the Planning system was adequate, and there can be a host of negatives associated with their use. Plus, it’s kind of embarrassing to see these proliferate immediately after a company’s multi-million-dollar ERP system gets implemented or upgraded.

The Solution Exists Today because …

a)  Constraints Management Group, LLC., deeply familiar with the basic, proven, and powerful Theory of Constraints Distribution and Supply Chain Solution, recognized that it was not sufficient to solve problems blocking some clients from performance improvement.

b) Out of that challenge grew a far more extensive application, with many innovative elements unavailable in any other technology.

c) The needs of their clients drove CMG to create and implement the solution in a range of environments – with education, training, consulting, and software – and extend the solution even further. Some of these environments were extremely complex and challenging – for example, with 165,000 part numbers, 600,000 BOM records, and 60,000 manufacturing orders per year dealing with shipments to remote international destinations and a maze of international regulatory requirements.

d) CMG hired and committed an in-house software development team and a hefty budget  (decisions that were not made without a lot of angst at the time!) to make the solution viable as a commercial software package for companies beyond their immediate client base, along with the accompanying “Thoughtware” for any manufacturer with ERP.

e) At a crucial stage, CMG found themselves in contact with Carol Ptak. Carol, a high profile expert in TOC and ERP with an MRP background almost unique today,  helped coin the term “Demand Driven Manufacturing” many years ago and recognized that the CMG solution was a practical application of the concept she’d envisaged years earlier. With an in-depth knowledge of MRP logic and the history of MRP, as well as expertise in a range of manufacturing technologies, Carol was able to provide a scale and context to CMG’s solution that greatly expanded its potential.

While the software is vital for anything but a small implementation (it’s also inexpensive), it’s important to recognize the role of the Thoughtware. The Demand Driven Institute has been created explicitly to provide education and training on this Thoughtware. A separate web site deals with the software solution.

The combination of realities that make the performance problems inevitable

The combination of realities that creates the need for DDMRP in some challenging ERP environments is:

  1. The MRP logic embedded inside ERP software was developed 50+ years ago, and reflects operating assumptions that were valid 50 years ago. The logic hasn’t changed since.
  2. However, the realities of manufacturing for many companies have changed dramatically in those 50 years; so while the logic itself (the Bill of Materials explosions and related MRP calculations) is still solid, the operating assumptions for MRP are no longer valid for many organizations.
  3. In challenging environments, this creates a conflict … today’s manufacturing realities make it more important than ever to have the (re)-calculating power of MRP … while those same realities render the flawed operating assumptions in MRP’s 50-year-old approach, devastating in terms of the performance damage they trigger.

What are the realities that have changed?

Customers demand shorter lead times,  faster responses, better customer service while outsourcing means longer lead times than ever before and more variability in supply lines than ever before; inventories have been leaned in many companies and in many supply chains; product life-cycles are shorter than ever, new materials are being introduced, new customers emerge quickly and die quickly, new markets emerge quickly and can fade quickly; the economy is like a yo-yo but now much more tied into global issues than ever before; and, etc.

You can summarize it as, new “highs” in variability, volatility and velocity.

The elements that make the Demand Driven MRP solution viable TODAY

The key elements that combined to generate a viable solution through Demand Driven MRP are:

  1. The maturity and breadth of the DDMRP Solution.

    The TOC Replenishment solution has existed for more than 15 years, and some software has supported it in recent years. But the DDMRP solution expands on, and adds substantially to the “basic” Replenishment solution; THEN adds whole new areas of functionality outside the TOC Replenishment solution. To a large degree it fuses TOC, MRP, Lean and some completely innovative elements to form the whole.

    DDMRP applies the Replenishment concepts internally inside manufacturing, not just in Finished Goods and Materials; which is a major advance. It codifies the logic behind sizing the Dynamic Buffers and the Static Buffers. It adds en entire “Strategic Inventory Positioning” strategy and associated tactics which simply doesn’t exist in any other technology, including TOC, and which is the key to lead time compression and to dampening the nervousness of an environment encountering a lot of variability and volatility.

    It adds a complete rewrite to the basic MRP logic to accommodate Replenish and Replenish Override part planning types, to include a dynamic Min/Max for “classical” min/max type parts, to incorporate the decoupling logic that prevents variability cascading throughout the system.

    It adds an entire Execution-focused application suite, with highly visible vehicles for priority management, a basis for precise priority management that underscores just how poor a practice it is to use demand dates as a priority planning basis, and a format for sharing this information with vendors, not just internally. It includes alerts to highlight On-hand issues needing attention and projected on-hand issues needing attention; alerts to warn of material synchronization issues; and alerts to ensure follow-ups on critical lead0-time managed purchased parts.


  2. The Thoughtware associated with DDMRP – education, training, systems design considerations and implementation planning  – is fully defined.

  4. Robust, commercially available software, from the Software initially developed for clients and proven successful in a variety of environments and with full support.

For those interested in going deeper …

1. Expanding on the MRP issue

  • Inside every ERP package there’s a module or function called “Material Requirements Planning,” or MRP.

    Its purpose is simple … explode the demand for finished products (originating from forecasts or sales orders or master scheduled orders or some combination) through the Bill of Materials to determine “gross” requirements for components at every level of the Bill of Materials, and requirements of materials and purchased parts; then “net” those requirements against on-hand stock and open or planned “supply” orders … work orders for manufactured parts, purchase orders for purchased parts.

    The outcome is a calculation of net requirements, and also a host of suggested activities (expedite, slide, launch a new order, cancel an order) necessary to re-sync existing supply orders against changed demands.

    What most managers don’t realize is that even in the most modern, multi-million ERP software package … the detailed MRP logic “embedded” in the ERP package is actually more than 50 years old! It hasn’t changed whatsoever in that time.

  • However, the world of manufacturing HAS changed substantially in that time.

    The markets place more pressure on lead times than ever before.

    Manpower is “leaned” more than ever before … in the shop and in the front office.

    Inventories are being “leaned.”

    Forecasts are as inaccurate as ever … or even more inaccurate.

    Offshore sourcing has substantially increased the lead time for many materials and parts.

    Variability is very high in the supply lines … even if a company is attacking variability internally with Lean or Six Sigma, the variability in long supply lines can be substantial.

    Then there are all the dynamic elements … new technologies, new materials, new products, new markets are always emerging. Suppliers are changing. Product life cycles are shrinking. There’s more product variety. Volatility is high.

  • The combination – especially in challenging environments such as those manufacturers with multiple-layered Bills of Material that cannot and SHOULD NOT be flattened in reality despite what “Best Practices” gurus suggest – can be devastating.

    For example, a change in top level demand – perhaps as simple and routine as sales orders arriving differently than the forecast suggested – can cause a “cascade” of changes in net requirements through layer after layer of Bills of Materials, across hundreds or even thousands of component records, causing MRP to spew out so many “reschedule” messages that no-one will ever get to them before they change again.

  • So, just when the need for MRP is at it’s highest – to help complex Bill of Material environments plan and re-plan materials and component needs to reflect changes in demand and supply – so its value is diminished by its poor “fit” to today’s realities.

    The system easily becomes unstable, the output “to be executed” plans are not really viable, and the execution of the MRP output plan simply becomes impractical.

2. The problems that DDMRP addresses

When changed manufacturing and marketing realities, 40-year-old MRP logic, and a challenging manufacturing environment come together … there are many negative consequences. These of course are the problems that DDMRP can solve or relieve, including:

  1. When you realize that all it takes is the lack of a single component or material to hold up assembly or fabrication of a product that might have thousands of components and sell for hundreds of thousands of dollars … it’s easy to see how the inadequate synchronization from “standard” MRP can easily lead to frequent shortages.

    When you add demand variability (markets expecting aggressively short lead times and extreme levels of responsiveness) and supply variability (lots can happen during long lead times from Asia) and complex Bills of Materials (thousands of parts, multiple layers in the BOM) … the shortage problem becomes chronic.

  2. If a company with these problems is attempting to implement a Lean-style “Pull” execution system in the plant, there is the potential here for a nightmare.

    While trying to duplicate what Toyota and similar companies do today, it’s easy to forget that in Japan, home of Lean, work on reducing variability had been proceeding for decades before the low-inventory kanban “Pull” mechanism was introduced. Whereas what we’re seeing in the West is often a worsening of whole-system (not necessarily process) variability as lead times extend to Asian supply sources.

    Companies are spending serious dollars (and energy) on Lean initiatives, we see Kaizen events by the score, we see the variability sometimes reduced in individual processes with lean or Six Sigma … but unless there is excellent synchronization of materials, parts and component availability with the Pull systems, the outcome is often frustration.

3. The most common attempts at a solution – but which fail to address the Synchronization problem

Commonly we see four different attempts at solving the poor synchronization issue:

  • Workarounds as described above i.e. personal Excel spreadsheets, personal Access databases, etc.

  • Attempts to flatten the Bills of Material.

This is straight out of the “Best Practices” textbook and there are definitely some environments where flattening the Bills can be a good idea.

However, there are some environments – for example, we have a client with 28 levels in their Bill of Materials – where the BOM simply cannot be flattened; in fact, we’d argue it SHOULD not be flattened.

While this is may sound counter-intuitive … we can sometimes ADD a layer to a Bill of Material and by so doing we can generate serious performance improvements in terms of product availability and reduced lead time that would NOT be possible without that extra layer.

  • Attempts to move to pure “make to order.”

Again, this is straight out of the “best Practices” text book and again, it can be 180 degrees away from what makes the most sense. It is often simply not feasible due to the cumulative lead times in the whole product build.

  • Attempts to develop more accurate forecasts.

This is quite possibly the largest red herring of all and can consume an astonishing amount of time and energy with little or no benefit.

Be Sociable, Share!