The Reality vs Illusion of Make to Order and Forecasting

Executive Summary

• Companies, promoted by consultants, continually propose that they would like to move to make to order.
• What goes unmentioned is that only a tiny percentage of companies can go to make to order.
• Why assemble to order is confused with make to order.

Introduction

At Brightwork Research & Analysis, we receive questions on forecasting from managers and directors of demand planning. Something which has come up frequently is whether their company can move to make to order. This question is a type of fool’s gold based on an oversimplification of what make to order relies upon. You will learn about the definition of make to order, how this compares to other manufacturing environments, and the relationship between make to order and the replenishment trigger.

Our References for This Article

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

Make-to-Order Definition or MTO Meaning

Make-to-order, sometimes called (MTO) or build-to-order or built-to-order (BTO), is one of the major manufacturing environments. The others are:

• Assemble to order
• Engineered to order
• Make to stock

Companies will use this environment to have some sub-assemblies already built, stocked, and ready to assemble based on the customers’ requirements. This approach is also called build to order or built to order (BTO) and focuses primarily on production strategy.

Some of the required components are stocked, whereas more expensive or highly customized parts may need to be produced. Production only starts after an order is received.

MTO Meaning Versus Build to Order or Built to Order

Make-to-Order (MTO, a.k.a. Build to Order): This approach is also referred to as build to order or built to order (BTO) and focuses primarily on production strategy. Some of the required components are stocked, whereas more expensive or highly customized parts may need to be produced. Production only starts after an order is received.

MTO and the Job Shop or JobShop

The MTO meaning can be closely connected to the job or job shop regarding the production type. The job shop is a manufacturing process that produces small batches. The job shop or job shop would create custom items.

The naming is that the job shop (plant) moves to a new job, a custom job. Hence the term “job shop.”

Job shops were the earliest form of manufacturing. Because job shops are inefficient, the rise of mass production significantly reduced the price and quality of available products.

How Common is MTO or Build to Order?

While often claimed, MTO or BTO is quite rare in practice. Only a few industries, like aerospace and construction, follow a pure build-to-order model.

The Motivations of Moving to Make to Order

There are several motivations for companies that desire to move to make to order environment:

• A desire to seem leading edge
• A desire to remove the company from the complexities of forecasting or eliminate bad forecasting.
• The inability to forecast at a sufficient accuracy drives people to think about make to order and provides fuel to proponents of Lean. Lean is now a significant trend that influences corporate decision making. There are circumstances where Lean applies; Lean can also be used to justify things that are not feasible. There is no denying the influence of Lean philosophies.

Some of the motivation is to, in essence, find a magic bullet. Any company that could move to the stock environment would never have to forecast, never have to keep inventory, and would not have waste in the form of obsolete products before they can be sold. It would also eliminate the need to reposition stock (called redeployment). Sounds great. Whether a company can succeed in making it to the stock environment primarily depends on the market they are in and the type of product they sell.

Make to Stock Definition

Make-to-Stock (MTS, a.k.a. Build-to-Stock or Build-to-Forecast): Here, the replenishment is triggered by a forecast. Probably misnamed, MTS should probably have been called to forecast, as the forecast is the trigger for replenishment – which, of course, results in stock until the actual sales order arrives.

Make to stock, or sometimes called (MTS) is one of the significant manufacturing environments. The others are

• Assemble to order
• Engineered to order
• Make to order.

Make to stock is the most common in manufacturing environments. While little discussed, make to stock allows for economies of scale in procurement, manufacturing, and distribution, lowering unit costs.

The Relationship Between the Sales Order and the Replenishment Trigger

All procurement and production are performed before the sales order is received. By contrast, in both Engineering-to-Order and Make-to-Order, all procurement and production are performed after a sales order is received.

With the products in the BOM for Engineering-to-Order being procured and produced the latest after the sales order is received as, at the time of the receipt of the sales order, it is not known precisely what is to be built.

In MTS, the replenishment is triggered based on a forecast.

Where Make to Stock Environments Apply

In markets or submarkets where customization is more important than volume or cost to the consumer, production can be postponed until after the sales order.

In markets or submarkets where costs are more critical, and there is little in the way of benefits of customization for the product make to order is not an option.  A good example being light bulbs, for instance. Here, production should be performed before the receipt of the sales order. It should also be observed that multiple manufacturing environments are employed for the same category of products. For instance, one can either buy a dress shirt from Brooks Brothers in a store, which was produced in a make-to-stock manufacturing environment, or can provide one’s measurements to a tailor and have the shirt custom-made in a make-to-order manufacturing environment.

What Goes Unmentioned

It is often unmentioned whether the company exists in an environment where customers are willing to wait for their items to be produced or if they expect them to be available when they want them. Also, what is often left out of discussions related to make-to-order is the increase in cost and the decrease in quality that necessarily follows. Companies need to be cost-competitive, so the extra costs associated with making to order are not typically acceptable.

The Extra Costs of Make to Order for Make to Stock Environments

• Increased transportation costs (to get suppliers to the company faster)
• Increased changeovers on machinery
• Higher supplies costs, both in higher ordering costs per item and higher transportation costs as smaller orders lead to higher per-unit shipping costs.
• Decreased utilization of factory equipment (as equipment must wait for input material, which may be late or on route when the machine is ready to produce)
• Rush jobs can reduce the quality of the accepted items and produce more finished goods that must be scrapped. This means that more raw material must be purchased for every sellable item.

Considering the Total Costs and Quality Implications

In totality, while moving a make-to-stock environment to order, the company reduces its planning, increases its costs, and reduces its product quality. This is evidence that planning adds value to the overall supply chain process.

Companies may be frustrated with their planning results. Still, when they describe moving to make to order environment, they do not add up all the extra costs and problems produced in this way. The quote below describes why to make to order is an unrealistic goal for the vast majority of companies.

Example from Our Book

The following is an excerpt from the book “Demand Planning Methods in Software.”

“However, it is not possible for the vast majority of companies to move to a build-to-order environment. Except for extremely special-ized manufacturing (such as print-on-demand publishing), it’s difficult to come up with examples of products that cost the same amount to produce whether making one or a hundred or a thousand. Similar limitations apply to procurement, as procuring in larger batches is less expensive than procuring in smaller batches. Many people who talk about moving to a make to order environment are essentially fooling themselves.”

What Are They Describing When They Say “Make to Stock?

In fact, in haste to move to make to order environments, companies often use examples of “make to stock” environments that are not made to stock. Forecasting is still necessary. Two examples of environments that are called make to stock are listed below.

Assemble to Order Example 1: Computers

For instance, they may bring up Dell Computer, which is well-known for their online business, not building the entire computer but instead holding components and building on-demand once an order is received. A computer can be assembled quickly as long as all the components are in stock.

This is how Dell has avoided creating final products, meaning they would have to guess or forecast how many of each configuration would be required. This is not made to stock. I quote again from “Demand Planning Methods in Software.”

“Most environments that people call “make to order” are actually “assemble to order.” Assemble to order still requires a forecast at the subcomponent level. This is why companies must forecast.”

Dell has a type of product where the complexity resides in the components, and the assembly of the components is easy. Therefore, they defer the natural part until they receive a firm order.

Assemble to Order Example 2: Books

Suppose we take the example of book production. Some companies are at the point where they can produce one book at a time. This is impressive and means that authors and publishers do not have to engage in long production runs, particularly during the early stages of a book’s life when its future demand is tough to predict. This print-on-demand is only possible at a higher unit price and a lower quality. It also has limitations in that, at this time, color books, or books with colored graphics, do not come out very well. Therefore, even with a straightforward item to produce (print pages, bind pages, add cover), building to order has cost and quality disadvantages.

Secondly, these printing companies do not wait for a book to be purchased and then order paper and ink. They hold vast stocks of the components and can produce any book that is ordered. Again, this is assembled to order, not made to order.

How Online Shopping Changes Forecasting

Many companies would like to move to an order environment. Indeed, it simplifies supply chain planning and reduces the company’s risks in inventories. Few companies can move to this environment. Producing in batches is both cost-effective and easier to maintain quality levels. It also reduces costs in procurement. Finally, in most cases, customers are unwilling to wait for the lead time necessary to produce items. Online shopping has introduced a typical lag of shipping. Buying online provides more opportunities for assemble to order as consumers are becoming accustomed to waiting a few days to arrive by courier, even if the item is stocked. This is a change from when people went to stores, expecting to walk out with the item. This may allow companies that sell online to extend the time between ordering and when the product is received, but taking a day or two to assemble the product. Once a consumer is prepared to wait two days for their order, they may be willing to wait a few days longer.

Can this be leveraged to move some companies to an assemble-to-order, if not make-to-order, environment? The rise in online shopping is something that companies should include in their supply chain strategy. It has a strong effect on how forecasting is performed. However, it is not apparent that the move towards online ordering radically changes the available strategies to companies. Instead, it provides some extra leeway, which can reduce supply chain costs and postpone the final assembly depending on the lead assembly time.

Background on Online Shopping and Their Supply Chain Implications

If your customers are not willing to wait for your average lead times is either completely impossible or significantly increases the costs to provide the item and reduce the quality level of the item. However, moving from assembly to order is possible if customers can accept a lead time.

Online Shopping and the Changes to the Retail Supply Chain

We are in the middle of significant changes in retailing. Online retailing has tremendous advantages over buying in stores. These include the following:

1. Far greater selection
2. More and less biased information. This is, in fact, one of Amazon’s significant assets: its review database. There are no product descriptions or objective reviews when people shop in stores. Stores have poor information quality to them. The information about products is low, and the only sources of information tend to be the biased salespeople in the store.
3. Convenience (no driving to a store, finding parking, walking around the store)
4. Speed (of the shopping experience)
5. The ability to be shown related products (Amazon does this exceptionally well)
6. The capacity to try out some products (such as reading parts of books and listening to music), although the ability to try or test other products (such as trying on clothes or shoes or driving a car) is a disadvantage.

The Effect of Amazon on Online Shopping

However, for many years, poor e-commerce websites, combined with trust issues and the resistance that many people had to change their shopping habits, restricted online shopping from overall retail sales. However, that is rapidly changing. Lots of credit must go to Amazon. Amazon set the standard for web design and customer service that the rest of the online sources were required to emulate. Amazon has been so successful that it has been considered a culprit in Borders’ closure and some Best Buy Stores.

The Tipping Point

I believe that it is soon to reach a tipping point and see many physical stores close completely or shrink the number of stores. This brings up an essential aspect of the opportunities this presents for product postponement to companies.

How Customers are Being Habituated to Wait for Their Products Through Online Shopping

Buying online provides more opportunities for assemble to order as consumers are becoming accustomed to waiting a few days or accepting a lead time for their order to arrive by courier. Even if the item is stocked, this is a change from when people went to stores, expecting to walk out with the item in hand. This may allow companies with online shopping to extend the lead time between ordering and when the product is received, but taking a day or two to assemble the product.

Once a consumer is prepared to wait two days for their order, they may be willing to wait a few days longer. However, these are assembled to order opportunities, not make to order opportunities. If some customization is performed on the item (such as monogramming or configuration options), this does not change the fact that this is assembled to order.

Online sales do not increase the market for make to order items much more than they have always been. However, assemble to order is a real possibility, and this is a move that has substantial impacts on forecasting.

Online Shopping Strategic Assessment

The rise in online shopping is something that companies should include in their supply chain strategy. It is not apparent that the move towards online ordering radically changes the available strategies to companies. Instead, it provides some extra leeway, which can reduce supply chain costs and postpone the final assembly depending on the lead assembly time. Companies should analyze their assembly period and calculate the costs of expedited shipping.

Once a customer is willing to accept at least some period to receive a third order, they may be ready to take a little more lead time. It may make sense to cover expedited shipping costs to get that extra day to perform the assembly. It is a new sales channel with different requirements than the physical store model that needs to be evaluated on a company-to-company basis.

Questions to Ask

• What is our assembly time if all the components are in stock?
• How long can we expect customers to wait?
• What is the cost of us paying the costs of expedited shipping to make up for the time we spend assembling the items?
• How does our forecasting need to change?  (it should be simplified)
• How much can forecasting effort be reduced?
• What are the impacts on our inventory? (assemble to order reduces the inventory costs)

The result of the supply chain strategy analysis should be the total cost of moving to an assemble-to-order environment. Everything is a series of trade-offs that must be calculated. Furthermore, a company may be able to use assembly to order for part of their business and make it in stock for another portion.

Conclusion

Make-to-order is a bit of a canard. Companies making to order environments naturally have a much easier supply chain planning challenge. That is a luxury that is mostly out of the control of the company.

While some companies may think this creates opportunities to move to make to order, it is hard to see how it does. Make-to-order means that procurement orders are sent out when the order is received. However, this produces too much time for an overall lead time unless the product is customized. However, the opportunities for assembling to order are increasing, and this will mean more forecasting at the assembly or component level than at the finished product. The forecasting needs will be as high, just at a different level in the bill of material.

The Problem: Where to Perform the Forecast for Assemble to Order Environments?

Under assemble to order, one can forecast either at the finished good or the assembly. In scenarios with comparatively few input products versus finished goods, the law of large numbers dictates that the forecast accuracy can be higher at the assembly. A test can be performed at both the finished good and the assembly levels to test the optimal forecast location. We have performed this analysis several times ourselves.