Timing of Capacity Increments

At its heart a capacity strategy suggests how the amount and timing of capacity changes should relate to long term changes in demand. Three options illustrate the range of possibilities available when demand is expected to grow steadily.

Alternative Types of capacity Cushion

One obvious way that companies can meet demand that is temporarily greater than their operating rate is by carrying inventories, either of finished goods or of parts and components that can be converted quickly into finished goods. As with the airline seat referred to earlier, most services cannot be inventoried unfortunately although some back room services (usually involving the processing of documents or information rather than direct contact with customers) can build WIP inventories. If inventories are possible they enable the quickest response to a demand surge, but building them for this purpose can be quite risky because it requires that you know  exactly which products will be demanded and in what quantities . If your forecasts are wrong, then you not only may not have enough of some products, you also may find yourself with a large inventory of unwanted and eventually obsolete products.

The more common view of a capacity cushion is that associated with the various resources described earlier: floor space, equipment and people. This kind of capacity cannot provide the same speed of response as can inventories but it is more flexible, in that the specific mix and volumes of products demanded by customers can be produced within the company’s normal lead time. Instead of incurring the risk of obsolete inventory, one now incurs the cost of unused capital resources that have a smaller risk of obsolescence.

It is possible, of course, to provide only a partial capacity cushion – of plant and equipment say but not of people. Then, if demand increases, overtime will be required and /or additional people hired and trained. This becomes increasingly difficult as the skills required become more and more specialized. For example, one study of the biotech industry at the turn of the century concluded that even if biotech companies were able to obtain sufficient funds for expansion they faced a looming shortage of the highly trained people needed to design, build and operate facilities. The trade-off is between the speed of response and the investment required – just as when deciding whether to hold inventories or provide additional production capacity. Alternatively one might provide capacity cushions for certain critical products or services and not for others.

If however, an operation is organized as a collection of work areas, each of which is dedicated to the creation of a small group of products or services, capacity cushions will have to be created for each one.  As suggested by our simple bank teller example, the total of these specialized cushions generally will have to be greater than the cushion required for a single facility capable of producing all products. On the other hand, both the investment and operating costs of this kind of flexible facility are usually greater than those of several simpler, more specialized facilities.

Similarly, if a product or service is delivered through several regional facilities each of which supplies a specific geographic region, more inventory is required to maintain an acceptably low profitability of stock-out for the system as a whole than would be needed to maintain the same level of service at a single facility that served of all regions. In the simplest case, where the probability distributions of demand in each of N different regions are independent of one another and have the same standard deviation it can be shown that the total cushion (i.e. safety stock or excess capacity) required to keep the probability of a stock out throughout the system below some value (95 per cent, say) is approximately equal to √ N times the cushion required if everything were held in the same location (or produced by the same facility). That is, four storage points need twice the capacity cushion that a single storage point requires to maintain the same probability of a stock out. This phenomenon affects all kinds of inventories – including those consisting of people, computing capacity, fire departments and urban services. One analysis for example estimated that Japan’s productive labour supply would be increased by more than 5 per cent if there were more mobility of people among firms since the traditional one firm for life practice in Japan essentially creates a separate inventory of people in each company.

As an example of the various kinds of capacity cushions that might be employed, one consumer appliance company considered a wide range of options. Aware of the uncertainties associated with planning capacity for a product that was still in the early stages of its life cycle, but anticipating substantial demand growth, this firm considered:

  • maintaining its existing production capacity for another six to twelve months (so that its capacity cushion   eroded as demand grew),
  • building inventory and working overtime in order to meet unexpected demand surges,
  • adding a second shift at an existing facility,
  • adding floor space and equipment to that facility and
  • building a new facility.

In deciding among these management considered (in addition to their respective costs) the likely growth rate of demand, the time required to get capital requests approved by corporate management,  and the likely reactions of competitors.

“What? Gaming in the workplace? No way!” This is something that we hear from Corporate
Closely tied to the question of how much capacity should be provided to meet forecasted
The notion of focus naturally, almost inevitably from the concept of fit. Just as a
However, as with most strategic decisions, the issue is more complex than it first appears.
A company’s operations infrastructure is composed of its policies and systems governing a number of