The article featured today is from Quality and Reliability Engineering International with the full Open Access article now available to read here.

 

The recent increase in digitalization of industrial systems has resulted in a boost in data availability in the industrial environment. This has favoured the adoption of machine learning methodologies for the analysis of data, but not all contexts boast data abundance. When data is scarce or costly to collect, Design of Experiments can be used to provide an informative data set for analysis using machine learning techniques. This article aims to provide a systematic overview of the literature on the joint application of Design of Experiments and machine learning in product innovation settings. The results delineate the state of the art and identify the main trends in terms of experimental designs and machine learning algorithms selected for joint application on product innovation.The analysis shows that the adoption of a Design of Experiments + machine learning framework positively impacts the product innovation process in terms of time, cost, return on investment and robustness of the final decisions.However, this field of research is far from its saturation and further contributions are needed in order to advance the state of the art. For instance, scarce relevance is put on the issues of causal investigation and uncertainty quantification of the machine learning models. Another problem regards the impact of traditional data partitioning methods on the experimental data set.