A review on learning effects in prospective technology assessment

Abstract

G. Thomassen, S. Van Passel, J. Dewulf

The transition to a circular economy requires new and improved technologies. For example, the collection of additional post-consumer packaging streams in the P+MD system may require improved sorting methods and new recycling technologies. However, these new technologies experience a disadvantage when they are compared with conventional technologies. While the costs and resource use of conventional technologies has decreased over time, new technologies did not yet experience this ‘learning effect’. The inclusion of this learning effects can provide a better estimation of the future costs and environmental impact of such technologies. Ignoring these learning effects, on the other hand, will lead to a structural underestimation of the potential of new technologies and can therefore limit its innovative potential.

To assess this learning effect, an extensive literature research was performed, where 105 studies that incorporated or calculated these learning effects, were discussed. A main outcome of this review, was that until now, learning effects were mainly included in the cost assessment of renewable energy technologies. An example of a typical application would be the use of learning effects in an outlook study for photovoltaic energy. However, there seems to be no reason why learning effects cannot and should not be used in other research fields as well. To enable the application of learning effects in the circular economy field, the following best practices were recommended:

  • Learning effects should be discussed both on the overall functional level (for example, the overall cost of a recycling technology) as well as on the underlying parameter level (for example, the energy consumption of a new separation technology).
  • Learning effects should be discussed both on the impact level (for example the economic cost or environmental impact of a recycling technology) as on the technological level (for example, the recycling rate of a specific waste stream)
  • Learning effects should incorporate both historical evolutions as expert estimates
  • Learning effects should include both learning-by-doing (learning by repeating an action) and learning-by-searching (learning by using a better action)
  • Data availability and quality considerations should be taken into account when using learning effects, for example by the use of scenarios and uncertainty analyses

These recommendations can be used to include learning effects in a broad spectrum of research fields, which can enable a more fair comparison of innovative and conventional technologies. In addition, they can also be used as a checklist to evaluate predictions concerning the costs and environmental impacts of emerging technologies, which can typically be found in roadmaps, and to provide recommendations to further improve these predictions.

Read the full article

CE Center co-authors:

 

Manager/Researcher CE Center

 

Promoter CE Center

 

Promoter CE Center