Learning-before-doing in the development of new process technology

The concept of learning-by-doing has featured prominently in the literature on manufacturing improvement and technological innovation. Much of the focus of this literature is on the 'doing' or 'using' that takes place in an actual commercial production or usage environment. But problem-solving can also occur long before a new product or process design is introduced into the factory through computer simulations, laboratory experiments, prototype testing, pilot production runs, and other experiments. These approaches to problem-solving are referred to in this paper as 'learning-before-doing.' Although there is an extensive literature on prototyping and technical problem-solving in R&D, there have been few attempts to determine the type of conditions under which different approaches to learning are most effective. This paper explores the impact of different learning strategies on development performance with detailed data on 23 process development projects from pharmaceuticals and biotechnology. The empirical analysis focuses specifically on how the timing of technology transfer to the factory influences development costs. The results of the analysis indicate learning-by-doing is essential for efficient development in an environment like biotechnology where underlying theoretical and practical knowledge is relatively thin. In contrast, the need for learning-by-doing is far lower in environments like chemical synthesis where underlying theoretical and practical knowledge is deep enough to enable the design of laboratory experiments that effectively model future production experience. The paper concludes with a discussion of the implication of these findings for the management of process development and learning.