The concept of Design for FDA (DfFDA) has a strong basis on the Food and Drug Administration’s (FDA) regulation for medical devices in the Unites States. In fact; an analysis of the factors that impact the time it takes the FDA’s to provide market approval for medical devices, the product design process model, and Design for X (DfX) methods with overlapping FDA objectives lead to the development of DfFDA as a means to increase awareness about regulatory compliance and promote designers to consider the regulations throughout the development process of medical devices. For doing so, the main objective of DfFDA is to provide regulation-focused guidelines to producers of medical devices. An important part of these guidelines and this paper’s major contribution is the development of a prediction model for the FDA’s decision time. Overall, we want this model to become a tool that allows medical device companies to come up with an accurate estimate of a product’s time-to-market after accounting for the FDA’s decision time. In this work, we provide a comparison and discussion on the adequacy of linear regression models and other non-linear models such as parallel and serial tree-based ensembles for prediction the FDA’s decision time.

This content is only available via PDF.
You do not currently have access to this content.