Background

The use of flexible manufacturing capacity offers the product-developing firm a portfolio of  ‘real options.’  Academically speaking, a real option is “the right, but not the obligation to take some action contingent on the realization of some uncertain event.”  Such options can have great value in R&D and new product development because they present opportunities to respond to new information. In such cases, higher levels of uncertainty increase the value of opportunities to postpone a decision until that uncertainty is resolved. In new product introductions, uncertainty abounds in many areas including product demand, introduction lead times, product and process development success, and eventual production costs. These uncertainties must be dealt with carefully in the pharmaceutical industries because of the huge amounts of capital involved.  A survey of the 16 largest pharmaceutical companies shows that drug R&D averaged 13% of sales in 1998. A 1994 government study says that firms average $360 million to develop and commercialize one product.  More recent estimates range from $500 million to $2 billion.  The research being done by Professors Chambers and Snir focuses on one major aspect of this problem.  Specifically, they look at how process flexibility generates a deferral option which allows firms to delay the expenditures for product-specific capacity until after most of the uncertainty has been resolved.  

The study of this problem is also important because the flexible capacity which creates these options is very expensive in traditional pharmaceutical manufacturing (bioengineered pharmaceuticals are another story), costing up to four times that of comparable dedicated (product-driven) capacity. By dedicated capacity, we mean a fixed facility for that particular chemical construct or drug. While the pharmaceutical industry has gotten more serious about using flexible capacity over the last decade or so, no one has sorted through the optimal allocation of this capacity. Chambers comments, “To our knowledge, no one has really studied this allocation decision. Pharmaceutical firms have traditionally been R&D- and marketing-oriented. The idea of getting serious about manufacturing strategy is fairly new.” So what is the optimal level of flexible capacity? And, why and when would its use be most beneficial?

The Allocation Decision

Given the emerging pressures, both from consumers and governments to control costs, the pharmaceutical industry is seeking new ways to retain its profitability.  Consequently, a fresh look at its manufacturing strategy seems in order.  In the problem under study by Chambers, Snir, and Ata, a company must decide how much flexible capacity to own and the allocation of that capacity to new products. In other words the decision maker needs to figure out how long a new product should be manufactured in a flexible facility awaiting the construction of dedicated capacity. Significantly, the authors’ analysis suggests that the optimal level of flexible capacity is actually significantly less than the amount which would allow all products to be produced in a flexible facility until all uncertainty is resolved. “We suggest a much lower level of flexible capacity than the big drug manufacturers appear to be currently allocating. That is, there may be over-capacity of flexible capacity in the industry at present,” Snir states.

Value of Flexibility

The research details the optimal time a product should be assigned to a flexible facility. Findings reveal that it is most profitable to delay the construction of dedicated capacity between two and four years for each product category under consideration.  However, in all cases there is substantial value in delaying for two years, with only incremental value generated for longer delays. Extending beyond four years always results in decreasing contributions because the added production costs associated with the flexible facility outweigh any additional benefits.  Using the profit maximizing decision for each product, the firm may increase its net contribution from new products by as much as 10% in comparison to the case without flexible capacity.

The deferral of construction serves to "buy time," which decision makers can put to profitable use. For example, gaining production experience in a flexible facility prior to the construction of dedicated capacity should result in more efficient process design and resource management.  This alters the projected stream of expenses, and enhances the value of flexibility. The authors cite efficiency gains if the flexible facility is used for at least two years. Initiating production in a flexible facility enables learning about process improvements, reducing the size and cost of a dedicated facility. A third advantage in using the flexible facility is that delaying construction of the dedicated facility lowers the net present value of its cost. For the most common size product this amounts to about $2 million per year of delay. Conversely, the disadvantage from using the flexible facility is that the cost of goods sold in the flexible facility is 50% higher than in the dedicated facility. When quantities are low in early production years, this penalty is relatively minor, but becomes critical as production scale increases. 

The authors’ model shows the optimal allocation of flexible capacity given different capacity constraint scenarios. It reveals how long certain categories of products would benefit from flexibility and when to move on to dedicated capacity. Given finite patent lives for new drugs, generally 17 years, a year in the life of a drug becomes significant in terms of real costs, opportunity costs, and real options. Firm-specific data can be easily incorporated into the model to evaluate the use of existing capacity and projections can be made to evaluate the attractiveness of additional investment in flexible capacity.

Looking Ahead

The larger value of this work is in its straightforward way of considering flexible capacity allocation in a number of manufacturing settings. The ideas in this research can apply to any firm using a combination of flexible manufacturing systems and dedicated production lines, or a firm holding a network of production facilities, some being job shops while others are dedicated to a specific product or product family. Chambers mentions that these same basic problems apply across all manufacturing settings and the problem is of such magnitude that the research warrants study even if the reader is not particularly interested in the pharmaceutical industry. This approach can also be applied to the task of assigning work to employees, where less experienced workers may only be proficient in one type of task while more senior employees may be viewed as more flexible resources.   

Snir concludes, “The cognitive thought by pharmaceutical firms of moving from flexible facilities and subsequently to build dedicated facilities occurred in the late ’80s and early ’90s when patent life became of primary importance. Additionally, drug R&D became more important in the last two decades, with the need to optimize that process more apparent in recent years. Drug companies had a long run with high margins. In today’s environment, firms are finding that they need to get smarter about manufacturing and how to allocate capacity under conditions of greater uncertainty and competition, higher construction costs, and environmental considerations.” Thus, a deeper analysis of manufacturing and allocation decisions is both needed and emerging.

“The Use of Flexible Manufacturing Capacity in Pharmaceutical Product Introductions” by ITOM Professors Chester Chambers and Eli Snir with SMU Ph.D. student Asad Ata is under review.

Summary by Jennifer Warren.