Open Access

Additive manufacturing (AM) is rapidly emerging from prototyping to industrial production [1]. Thus, providing AM parts with a tagging feature that allows unambiguous identification, can be crucial for logistics, certification, and anti-counterfeiting purposes. The digital object identifier (DOI) acts like a fingerprint for the product throughout its entire lifecycle. Several methods are already available, which range from encasing a detector [2] to leveraging the stochastic defects of AM parts [3], printing a quick response (QR) code or a set of voids partially filled with loose powder within the part [3]. A new method is using microstructural features of the AM part without altering their properties. This technology transfer project aims to implement this authentication methode as software solution to act as certificate in the Digital Product Passport (DPP) [5].

A method to uniquely identify samples without printed or handwritten labels is an advantage not just for additively manufactured parts. To kickstart industry use cases, it is also important to provide a ready-made implementation kit. Following an open-science and open-source software approach Germanys Federal Institute for Materials Research and Testing (BAM) seeks to promote digital solutions of ongoing research projects. With this software package a novel method based on microstructural features as identifiers – DOI4AM (digital object identifier for additively manufactured parts) – will be explained alongside its implementation as open-source Python software package. The digital object identifier (DOI) links product data clearly and forgery-proof with real components. Its implementation helps to identify and securely authenticate additively manufactured components during its product life cycle by using characteristic microstructure features - just like a fingerprint. To calculate the DOI fingerprint, a few preprocessing steps need to be performed to detect the uniquely distributed microstructure features that occur during the 3D printing process. A go-through guide shows the preprocessing steps that include CT image capturing, feature segmentation, and data distribution with CSV files. While all steps can be followed along in a Jupyter notebook, the software package includes an application for creating and checking of previously created fingerprints, as well, as a containerized API (application programming interface) service for implementation in existing software platforms or workflows. While data visualization is crucial to understanding the methodology and an essential tool to check for data correctness, an implementation in an industry use case needs to be slim and resource efficient. Therefor the software’s API can be used as an independent service. The project's industry partner proofs its first successful implementation in their digital product passport web solution PASS-X.