Combining Innovation and Capacity Utilization in High Throughput Systems: Moving Beyond the Product Life Cycle Model by Introducing Second-Order Innovations

Abstract

The literature on the product life cycle and on high throughput systems has been preoccupied with studying an apparent lack of flexibility in capital-intensive production systems. Companies in capital-intensive industries need to maintain a high level of capacity utilization in order to stay economically viable, however, their efforts to uphold the throughput of their systems often have the unforeseen and unintended consequence of limiting their ability to introduce new products and services. Nevertheless, some companies have managed to resolve these tensions by introducing what we describe as “second-order innovations”, a type of innovation which acts on the innovation process itself and enables new products and services to be introduced without a steep decline in capacity utilization. By focusing on these cases and discussing their theoretical implications, we want to contribute to the existing literature on high throughput systems by identifying key mechanisms for introducing and maintaining such second-order innovations and describing the patterns of industrial evolution that they create.

Keywords:

• Architectural knowledge
• automobile industry
• capacity utilization
• higher order innovations
• interactive computing
• organizational capabilities
• product life cycle model
• second-order innovations

Acknowledgements

The authors would like to thank Paul S. Adler, Erik Bohlin, Martin Campell-Kelly, Fulvio Castellacci, Andrew Davies, Jarle Hildrum, Mark Lorenzen, Paul Nightingale, Dorian Rutter and Edward Steinmueller, as well as two anonymous reviewers, for helpful comments.

Notes

¹ In this paper, “strategy” refers to company decisions about where and how to compete, as they are discussed in conventional business strategy books, “organization” refers to internal routines and management, whereas “technological innovation” refers to any changes in production machinery or product specifications.

² Regarding interactivity, many of the early computers were obviously rather unreliable, difficult to program and to operate, whereas some (such as Whirlwind) allowed for elementary interaction (Datamation, 1966a, b).