This paper presents the outcomes from 2 runs of the Working Group on Learning from Deliberate Failure from 2 successive International CDIO Conferences: first in 2023 and another in 2024. This paper outlines a holistic approach based on the CDIO Framework to the development of failure-tolerant mindset among students. We believe this paper is the first of its' kind to offer a comprehensive set of design principles that provide guidance to faculty interested in engaging students in learning from Deliberate Failure. The purpose is to assist in the design of a series of learning tasks that had “failure elements” built-in (hence the term “deliberate failure”) so that students can become more receptive towards accepting failure as a way of learning. This paper first briefly summaries outcomes from earlier work where we highlighted that there exist 2 seemingly polar extremes of learning from failure: one that seeks to promote innovation (such as in product design) via frequent exposure to failure where frequent experimentations are encouraged so as to improve with every iteration; and another that strive to avoid failure at all costs, for example, in operating complex systems such as nuclear or chemical plants. For the latter, learning from failure is achieved through case studies of past failures or from simulation of known failure events. This is followed by a literature review on design principles. We then proposed a holistic approach that faculty can use to design learning tasks for learning from failure. The approach is holistic in 2 ways: First it comprehensively covers all aspects of students learning experiences, for a transformative change in their mindset on failure, from the initial appreciation of failure’s role in promoting learning to the eventual embrace of failure as an experience that is to be welcomed; to be achieved through a range of design principles appropriate for various stages of the transformative process. Secondly, the approach is holistic in that it also provides guidance on how faculty can facilitate the learning process, for example using autonomy supportive teaching, and role-modelling. This paper concludes with discussion on limitations of the approach, and possible future work notably in the areas of assessment of learning.
