Although It Is Generally Agreed That Companies Are Better Off When Their Manufacturing

Rethinking Lead Time Reduction Investment:

A Real Options Perspective

Abstract 003-0316

Suzanne de Treville1

University of Lausanne

Lenos Trigeorgis

University of Cyprus

Alessandro Crego

General Motors Brazil

1Corresponding author:
Suzanne de Treville

Ecole des Hautes Etudes Commerciales

University of Lausanne

BFSH-1

1015 Lausanne, Switzerland

Telephone: +41 21 692 3448

E-mail:

Abstract

Although it is generally agreed that companies are better off with shorter manufacturing lead times, investment in lead time reduction is often difficult to justify using traditional project valuation techniques such as net present value (NPV). In this article, we suggest that evaluating investment in lead time reduction from a real options perspective facilitates quantification of the value of manufacturing flexibility brought about by lead time reduction, particularly the value of the option to time production commitment based on better demand information. This flexibility is significant when demand is volatile. We also present examples to demonstrate how options inherent in lead time reduction can have synergistic effects with related investments, such that a combination of such investments may have positive value even when the NPV of the individual investments is negative.

Company A, a producer of telecommunications equipment, has been losing an average of 25% of sales because of stockouts and markdowns resulting from supply chain mismatches. These mismatches occur because Company A’s several-week lead times require that production be committed before full information is available concerning demand. The long lead times are due to Company A’s policy of maintaining a utilization level of 90% to 95% for its extremely expensive equipment. Operating at a lower capacity utilization would increase per-unit production costs; hence, management has opted to maintain a high utilization level and accept the inevitable supply chain imbalances.

Company B produces equipment for winter sports in a factory located in China. A decade ago, Company B operated plants in its domestic (European) market, but it closed these plants and shifted production to China because of the substantial difference in manufacturing costs, due primarily to low labor costs in China. A manager from Company B commented:

The Chinese operations save us a fortune in manufacturing. The quality problems that we expected to encounter never materialized—the quality coming from China is very high. The only serious problem is that we have to commit production about a year in advance. If we get this wrong, then it costs us a fortune.

This company had considered building a small domestic plant for the production of products with the most volatile demand, but the NPV of this domestic plant project was negative, so the project was turned down.

Company C, a pulp and paper manufacturer, was considering investment in an Enterprise Resource Planning (ERP) system that was expected to improve the transfer of demand information from customer to supplier in the supply chain. Some managers, however, questioned the value of such information, as it would arrive too late to allow the company to respond effectively because of the long lead times. Standard discounted cash flow (DCF) analysis indicated that the value of the project was negative. These managers felt they should first invest in reducing lead times, but again the net present value of the lead time reduction project was negative. For Company C, the valuation challenge went beyond valuing stand-alone flexibility: It required evaluating possible interaction effects between two projects, each of which had a negative stand-alone NPV.

Problems in valuing lead time reduction investment

In each of these prototypical cases, the negative NPV of the lead time reduction investments was in direct conflict with the intuition of many company managers. A typical comment was the following:

It is clear to us that shorter lead times will make us much more responsive to the needs of our customers, which ought to show up in the financials eventually. It isn’t just a question of taking business away from competitors because we are faster—we saw a long time ago that competitiveness is a lot more than just being faster than the competition. But it has been very difficult for us to quantify the value of speed using traditional valuation techniques.

Why is it so difficult to capture the value of lead time reduction? Part of the problem may stem from a long-standing controversy concerning the impact of lead time reduction on product cost. The lead time reduction literature has argued that traditional accounting practices do not fully reflect the cost reduction inherent in lead time reduction,[1] suggesting that NPV leads to a systematic undervaluation of lead time reduction investment. In the two to three decades since lead time reduction became a primary issue in operations management, however, little progress has been made in resolving this controversy. Few companies have been observed to change their management accounting systems to facilitate lead time reduction. Upton[2] observed that “while flexibility can provide a powerful means for increasing revenues, most managers still focus on comparing cost figures—if they make any comparisons at all” (p. 79).

Differences concerning the impact of lead time reduction on production cost represent only a small part of the problem, however. More generally, it has been suggested in the finance literature[3] that traditional project valuation methods such as DCF do not properly recognize the value of flexibility and the ability to adjust future decisions based on better information. Even if managers were open to balancing cost with flexibility, standard project valuation methods would lead them in the wrong direction. Traditional project valuation techniques implicitly assume that managers run projects passively (under a static set of prespecified operating policies), which is not the case with flexible investments. The greater the uncertainty and the strategic value of the flexibility inherent in a given investment, the more traditional project valuation techniques will tend to undervalue such investment.[4] In such cases, option valuation is a more appropriate methodology. Given that investments in lead time reduction tend to be driven primarily by strategic flexibility rather than by cost reduction, the undervaluation inherent in DCF techniques may be particularly relevant to such projects. Our objective in this article is to move beyond the long-standing controversy concerning whether or not lead time reduction investments reduce production costs or not. We suggest that a more interesting question is the quantification of the value of strategic flexibility inherent in lead time reduction investment. Such investments should be valued using option valuation rather than DCF analysis.

An overview of lead time reduction

Let us revisit Company A’s situation from above. This company suffers from losses of sales due to stockouts and is considering investment in lead time reduction as an option to postpone production until better demand information is available. What will that firm need to do to “remove time” from the process? To fully comprehend lead time reduction as an option, it is important to integrate the basic principles that drive lead time with options theory.

The first key principle of lead time reduction is illustrated in Figure 1. As utilization increases, average waiting time increases at an increasing rate. Utilization is determined by the process bottleneck. All resources in the manufacturing operation can have extra capacity, except one: the bottleneck resource that defines the capacity of the entire operation. It is the utilization of this bottleneck resource that determines the amount of time that units produced have to wait to be processed. Thus, part of investing in lead time reduction consists of acquiring a capacity buffer, perhaps through adding equipment, labor, or both.

[insert Figure 1 about here]

The second key principle concerns the relationship between lot sizes and lead times, as illustrated in Figure 2. In situations where a setup is required for each new lot, the increase in capacity utilization from the incremental setups may cause a net increase in lead time—or even a lack of capacity—if lot sizes are reduced too much without reducing setup times. Once there is sufficient capacity to perform the extra setups, however, the relationship between lot sizes and lead time is approximately linear. This means that, assuming sufficient capacity, a 50% reduction in lot size amounts to a 50% reduction in lead times. Therefore, investing in lead time reduction is likely to require reductions in lot sizes, implying increased resources dedicated to setups or an investment in setup time reduction.

[insert Figure 2 about here]

Suppose there is enough capacity on the relevant machines to permit lot size reduction, but the machines are organized in a traditional job shop formation such that a piece must be transported some distance from one operation to another. In this case, extensive lot size reduction will not be feasible without a reorganization of the layout. Such layout change is often accomplished through a reorganization of the shop into cells. A cell can be thought of as a mini-factory that produces a complete part or subassembly and that has all of the equipment and workers necessary to perform that production.

It is well-established that a cellular layout facilitates lead time reduction.[5] Transforming the layout, however, requires investment. Some machines may need to be duplicated. Workers must be dedicated to the cell; hence, they may be idle while parts wait elsewhere in the factory. The task of moving machines into a cell configuration is not a trivial one and may imply substantial costs, such as for changing electrical and plumbing systems.

Improving the flow of the product can be done not only in the manufacturing facility but also in the supply chain; for example, through choosing suppliers located close to the purchasing factory or through locating manufacturing closer to the market. A firm might even choose to produce domestically rather than offshore.

Finally, lead time reduction is facilitated through reduction in variability (whether in arrivals or service rates) of the operation. Figure 3 shows the impact on inventory buildup in a system as the coefficient of variation (i.e., the ratio between the standard deviation and mean) of either service or interarrival times increases. As the variability of a process increases relative to the mean service and arrival rates, the average waiting time (and parts awaiting processing) will increase for a given utilization level. Some variability can be a source of strategic flexibility, allowing a firm to compete through its ability to adjust production quantities or items to reflect changes in demand. Other variability, however—such as machine downtime, quality problems, or worker absenteeism—has no strategic value and serves only to reduce system performance. Optimal lead time reduction, therefore, calls for the reduction of such nonstrategic variability.

[insert Figure 3 about here]

If Company A decides to invest in reducing lead times, the investment is likely to require (a) addition of a capacity buffer to ensure that parts do not spend too much time waiting in front of bottleneck machines; (b) reduction of lot sizes, which often requires setup time reduction and a change in the layout to permit smaller lots to be transferred immediately to the next workstation; and (c) a reduction in variability. The supply chain may need to be redesigned to avoid long logistics-related lead times. All these investments appear to increase the per-unit production cost. In the following section, we demonstrate how real options theory can be used to assign a value to the increased flexibility arising from investment in lead time reduction. The value of this flexibility may well justify the investment required to reduce lead times.

A real options approach

Consider two projects. In the first project, managers make a precommitted decision and then passively observe what the results of that decision will be. In the second, managers make the first of a series of decisions that can be revised and adapted later based on market developments and better information. Obviously, the managerial flexibility inherent in the second project will make it more valuable. This extra value comes from the fact that the ability to adjust the initial decision limits the downside risk and increases the upside potential of the investment. Properly valuing the second project presupposes incorporating the value of the flexibility into the NPV of the project.

To demonstrate the value of this flexibility, consider a simple numerical example that resembles the situation faced by Company A above. A firm is considering producing a batch of an item with a per-unit production cost of $60. The selling price of the item is $100. The lead time for producing the item is 2 days. At the time production begins, there is a 50% chance that there will be demand for the item and a 50% chance that demand will be zero, in which case the lot of product will be scrapped. Therefore, the expected value of the cash inflows is .5 × $100 + .5 × $0 = $50, which is $10 less than the production cost. That is, the expected value of production is –$10. How much should the firm be willing to pay to reduce lead times such that it has the option of postponing production commitment until demand is known? To address this, we must consider how the firm’s behavior will be different once demand information is available. If there is demand, the firm will spend $60 and receive $100, for a profit of $40. If there is no demand, then the firm will choose not to produce, for a net cash flow of $0. The expected value of the cash flows considering this managerial flexibility thus rises to .5 × $40 + .5 × $0 = $20. As a result of this flexibility, the firm is protected against the bad scenario in which the lot is produced but cannot be sold due to lack of demand. Therefore, once the option of postponing the commitment of production until demand is known is incorporated, the expected value rises from –$10 to +$20. The difference is due to the option value of timing production so as to make use of better demand information. If the cost of reducing lead times enough to permit postponing the production decision until demand information is available is less than $30, then the investment in lead time reduction is profitable. In this case, we can think of investment in lead time reduction as purchasing a postponement option on when to commit to production.

Several common types of real options have been identified, including the option to postpone, abandon, adjust the volume (i.e., expand or contract), switch (e.g., to another site or technology), and add a follow-on opportunity.[6] Any of these option types can result from lead time reduction investment. The most obvious option from reduced lead times is that of postponing production until better information is available about demand. Once demand is better known, lead time reduction enables management to adjust the production quantity of a given item or to cancel the order completely (i.e., expand, contract, or abandon a production order). Management may also switch colors, sizes, configurations, production equipment, or even manufacturing sites. Finally, lead time reduction may open the door for some intriguing follow-on investments—such as developing the relationship with the customer—that we discuss later.

The above options are not mutually exclusive. For example, shorter lead times may give management the flexibility to reduce the production quantity for one item, increase it for another, and switch the color of a third item, perhaps in combination with a follow-on investment. Neither are these options independent: As we discuss later, combinations of options may involve substantial synergies, working together as a configuration.

Overseas vs. domestic production: A postponement option example

Let us revisit Company B, which had made the decision to move its production overseas to reduce production costs. The transfer of production offshore resulted in a reduction of direct manufacturing costs from $70 (domestic) to $60 (offshore) per unit. As a result of this transfer, the company had to commit to production quantities and begin production 1 year before the items were to be sold—although the actual time required to manufacture the part was only a few hours. Reasons for the long lead times included the high demand for manufacturing capacity in that region, the high utilizations of such capacity, the large minimum lot sizes required for production, the long transportation times for some critical raw materials, and the shipping times to market. Although the firm was aware that an increase in lead times might severely hinder its ability to match production to demand due to the high demand volatility, it appeared to Company B’s managers that the substantial reduction in production cost represented a sufficient competitive advantage to warrant the shift. Some managers objected that the lead time problem might be sufficiently grave to justify establishment of a domestic manufacturing facility to produce items having high demand volatility, because the domestic facility’s shorter lead times would permit production to be postponed until demand was better known. Others pointed out, however, that the same items would be produced at a substantially increased production cost per unit and that the NPV of the investment in a domestic manufacturing facility was negative.