Build-to-Order & Mass Customization

15 Articles Seminars Consulting Results books Clients Credentials Inventory Reduction

by Dr. David M. Anderson, P.E., fASME, CMC 805-924-0100
Copyright © 2017 by David M. Anderson

The Results of Eliminating Inventory:

Save a lot of Money. For every $4 million of inventory eliminated, companies will save $1 million dollars a year!

Avoid Inventory Problems. Avoid inventory obsolescence, inventory deterioration, warehousing expenses, expediting, and lost sales from the inevitable out-or-stocks.

Quick Response. Instead of forecasting sales, spending money to build those products, paying inventory carrying costs until they are sold, and hoping you built the right ones in inventory, you could build all products on-demand without any inventory.


The following sections will show how to reduce and even eliminate Raw Materials inventory, Parts inventory, Work-in-Process inventory, and Finished-Goods inventory.

Reducing Raw Materials and Parts Inventory

The Problems with Raw Materials Inventory

If the rare case when delivery time doesn’t matter, raw materials and parts could be purchased after the manufacturer receives an order and then production would begin after they all arrive. But the vast majority of manufacturers must be much faster than that, so they may try to stock all raw materials and parts in inventory. However, it would cost way too much money to stock all of those needed for all products that might sell. So the manufacture will try to use sales forecasts to predict which product variations are most likely to sell and then tell Procurement which to order and stock. Of course, this scenario would only work if the forecasts were perfect and all the parts and materials were available, in the right quantity, to start production. If any of this is off, then expensive expediting will have to try to quickly procure whatever is missing.

The Solutions to Raw Material Inventory Problems

1. Standardize all raw material
s, even if most applications will get a “better” material. The money saved in overhead costs would be far greater than the going for the minimum spec for each of a proliferation of materials. If standardized enough, these materials can be ordered in a steady flow with a slight leveling buffer stock with confidence they will be used one way or another.

Results. All that redundant inventory can be eliminated. Out-of-stock delays will be eliminating since fully stocking a plethora of raw material part number is impossible, but expensive to try.

2. Standardize on the longest version and cut-to-length as needed. This applies very well to coiled linear material, such as tubing, hose, rope, chain, wire, strips, and so forth. Even straight material can be cut-to-length, for instance, hinges, thick tubing, pipe angles, channels, and I-beams. Both of these categories were cut-to-length at one of the author’s clients, Pentair’s Hoffman enclosures. The following link is a summary of a journal article they wrote:      
   Another client ordered heat-exchanger tubing in long straight lengths shipped in 50 foot boxes that easily fit on a truck trailer or shipping container.
    Any waste (which could be recycled) would be more then paid for by reductions in supply chain management and inventory costs.
The means to do this can range from manual cut-off, to manual cut-off up against a programmable stop, to uni-dimensional programmable cut-off machines.

3. Very long raw materials, like sheetmetal and fabric, can be ordered in coils or reels directly from the mill. Material “blanks” can be cut on-demand for all users in the factory.

Results of Dispensing Sheetmetal from Coils:

• The extra cost and delays of intermediaries can be avoided. Keep in mind that all sheets of metal are cut from coils and all bolts of cloth are cut from reels by these intermediaries, who are usually less efficient than the methodologies recommended herein because they have to do additional non-value-added steps to repackage materials into cut pieces.

• Nesting waste will be minimized because nesting is much better when the length is essentially infinite. And waste will be greatly reduced when large pieces must be cut from pre-cut sheets or bolts.

• Remnant management will be eliminated to store, identify, and retrieve remnants, which can be significant when large pieces are cut from not-much-larger sheets. These large remnants are hard to find uses for and they usually get damaged during storage and retrieval. Further, color-coded markings may have been cut off, thus opening possibilities for mistakes.

• Cut blank inventory will be eliminated when all blanks are cut in-demand. And inventory carrying costs are higher for heavy materials that must be stored on pallets, which need fork-lift aisles to access the pallets.

Reducing Inventory for Parts

In order to reduce parts inventory and avoid the problems mentioned for raw materials, parts must either be available on-demand or build to-order by the following methodologies:

Solutions to Reducing Parts Inventory

1. Standardizing parts can greatly reduce part inventory to only the standard parts, which can then be “ordered” as a steady flow, with the confidence that they will be used on way or another. With enough standardization, the standard parts can even have a small stock on hand to even out product demand

Results. Standard parts would be available for on-demand assembly with little inventory.

2. Kanban parts. Done right, the kanban system (see kanban article) can actually be build parts in batches and still always be available for on-demand assembly. When the assembler finishes a bin’s worth of parts, that is the pull signal to send the empty bin to its “source” to be refilled and returned before the second bin is depleted (as shown in the illustration in the kanban article).

Results. Kanban parts can be made in batches and, yet, always be available at all points of use for on-demand assembly without any overhead for MRP-based purchase orders.

3. Parts build on-demand. There are two ways to accomplish this:

a.  Parts may need to be designed for build-t-order .

b. The manufacturer builds parts on-demand from spontaneously available materials and parts, as described above and at the article on spontaneous resupply.

c. Your vendor/partner builds parts on-demand and delivers them quickly to your “pull signal.” Note that batching parts for shipping across oceans is not “quickly.” Trying to “pull” parts from a supplier’s inventory is not really build-to-order because it suffers from all the shortcomings of inventory, mentioned above and in the Mass Production article. One exception may apply if the manufacturer is big and powerful, like Dell Computer, and the manufacturer insist that suppliers build warehouses next to the assembly plant, stock them with their parts based on the manufacturer’s forecasts, and pay the inventory carrying charges until the manufacturer “pulls” a part into assembly. Such manufacturers have been know to have negative workings capital because the collect money from customers right away and pay the suppliers net 30 days. This arrangement is hard to pull off and can only build products from this inventory about 3/4 of the time.  See Dr. Anderson's article in Fabricating & Metalworking magazine, titled: Tearing Down the Walls; Design for Manufacturability and Concurrent Engineering require vendor/partnerships to reap the lowest total cost and the fastest time-to-market at

Reducing Work-in-Process Inventory

In any manufacturing process Work-in-Process (WIP) inventory carrying costs will be accruing while every assembly processing step is being done.

1. Slow Processing Steps Accrue Inventory Carrying Costs

Slow processing steps will incur more WIP inventory carrying cost. Slow process usually occur because products are not designed for manufacturability, are not built on concurrently engineered tooling and equipment, and need too much manual processing that require too much skill and judgement and then needs inspections and rework to correct the manual work. Further, inaccurate, incomplete, or ambiguous documentation can slow down part fabrication and product assembly, especially when outsourcing any of that production.

All of these delays will incur a significantly more work-in-processing inventory carrying costs proportional to the value of the WIP inventory at that point in the process, which increases at every step.

Consider the common practice when the last step is wiring an expensive machine, which usually involves a lot of manual wire routing (sometimes left to the judgement of the assemblers!) and hundreds of individual connections, when machines are “wired like a house.” All of this is labor-intensive and prone to errors, which take even more time to troubleshoot and repair.

The WIP inventory carrying cost is the highest at this point because all of the expensive parts and all the previous value-added processing is chalking up inventory carrying costs during these lengthy steps. So the WIP cost would be the inventory carrying cost multiplied times almost the entire value of the product!

Solutions to Slow Steps

Design for manufacturability  and minimal skill & judgement, concurrently engineered tooling.

2. Batches between Steps Accrue Even More Inventory Carrying Costs

Work-in-Process inventory carrying cost can be several times more if there are batches of parts or product waiting between each processing steps.

Solutions to WIP Carrying Costs of Batches Between Steps

The solution to the very expensive problem is flow manufacturing, sometimes called one-piece flow which has no inventory between steps.

3. Adding Expensive Parts First Greatly Raises WIP Inventory Cost

It may seem logical to start assembly with the most convention “foundation” part and then add the rest of the parts on top of that foundation. For instance, large trucks and other commercial vehicles usually start with a frame, which is not very expensive. But then all the expensive parts are added next, such as the engine, transmission, and axles, which bolt on quickly. The rest of the assembly adds low-cost parts that are very labor-intensive steps like doing all the wiring and plumbing
The problem with this “logical” approach is most of the procurement budget is spent up-front and, for the rest of the assembly process, the high cost parts keep chalking up large WIP inventory carrying costs.
And if steps are slow (problem # 1 above) and there is inventory between steps (problem # 2 above), then enormous WIP inventory carrying cost will be incurred. At one truck manufacturer, the WIP inventory carrying cost exceeded profits! So eliminating them would double profits.

Solutions to High Accumulations of WIP Inventory Costs

Optimize the order of assembly so the lowest cost parts go in first and the and the slowest, most labor-intensive steps happen first.
   Then bolt on the most expensive parts as late as possible, ideally just before shipping. If that is hard to do with the current design, then the next-generation architecture would need to be designed for lean production and build-to-order to enable that. For the truck example, the engine, transmission, and axles could “prepped” off-line and then quickly bolted on right before shipping. This may require “dummy” fixtures for the engine/transmission and some reusable wheeled dollies to fill in for the axles.

One excellent example : The most expensive part in a wind turbine is the generator. One company minimized WIP inventory carrying costs by bolting it on the day before shipping!

Finished Goods Inventory Elimination

The most expensive inventory is Finished Goods Inventory (FGI), which must be paid when Manufacturing builds products “to-forecast,” places then in FGI inventory, and waits for them to be sold, hoping that the forecast-built product variations in inventory are what is most likely to sell. And FGI inventory carrying costs will be ultimately paid by the custom no matter where the inventory is: at the company, at distributers, or at dealers.

Consider a car dealer with 200 cars in inventory with an average value of $20,000 each, which would represent a total finished goods inventory value of $4,000,000. The FGI inventory carrying cost would be 25% of that per year and that would actually cost the dealer $1,000,000 per year! So the dealer would pay a million dollars a year, just for the chance to make that impulse sale to a customer who could make that impulse buy and drive it home that day. However, the options may not be what the customer really wanted, but that would always be the case for BTO.

The numbers are much more shocking for the manufacturer. The book, “Build-to-Order; The Road to the 5-Day Car” said “A leading manufacturer estimated its current inventory located at distribution points in Europe to be worth €10 billion, reflecting the picture in the US.” That amount of inventory will cost €2.5 billion per year!

The Solution to Eliminate Finished Goods Inventory Costs

Using all the principles above and on this site to build-to-order products which would not only save all those inventory carrying cost but also give the customer exactly what the customer wants.


To learn more, read all the build-to-order articles on this site or for more thorough understanding, read the 512-page book “Build-to-Order Consulting & Mass Customization,” which can be ordered overnight at Amazon. com.  This book's order page is at:

For a free signed BTO book, signed by the author, delivered in the US, Call the author, Dr. Anderson, at 805-924-0100 (Pacific time zone) for a free assessment of how these principles can help your company.

To learn more fill out the form on the home page:

Dr. David M. Anderson, P.E.,  fASME, CMC
Management Consultant
Build-to-Order Consulting
phone: 1-805-924-0100
fax: 1-805-924-0200

[Home]  [15 Articles]    [In-house Seminars]   [Consulting]    [Books]    [Bio-Sketch]    [Clients]


Copyright © 2017 by David M. Anderson