Recent Findings About the Novelty Metric of Shah

The work presented in this paper aims at presenting the last findings about one of the most acknowledged and used novelty metrics in the field of engineering design. The a-posteriori novelty metric of Shah, Vargas-Hernandez and Smith is often considered in design research, but after two decades, there are still some critical issues to be solved. This paper highlights the issues identified in the last research works, by referring to the few published works that directly face the argument. It emerged that notwithstanding the presence of metric variants proposed by scholars, none of them is capable to overcome the problems. In particular, the assessment of set of ideas implementing heterogeneous numbers of attributes can lead to misleading results. In absence of a comprehensive solution, this paper provides a list of practical hints in order to allow the application of the metric (or the related variants) and ensuring the repeatability of the experiments.