Information
Materials

WHAT BECAME OF.... “KADI4MAT”

In NEULAND Magazine 2020, we presented the digital research data platform Kadi4Mat, which was developed at KIT to record the entire research process in materials science in a structured way. What has happened since the project was presented? Time to take a look at the state of play.

Aktuell umfasst die Plattform fünf Softwaremodule: (1) KadiWeb – eine webbasierte Forschungsumgebung mit ELN-Funktionalitäten und Repositorien, (2) KadiStudio – eine Desktop-Version zur Formulierung und Ausführung von Arbeitsabläufen, (3) KadiFS – zur Integration externer Software, (4) KadiAPY – für die Einbindung in Python-Programme und Skripte sowie (5) KadiAI – für den Einsatz von maschinellem Lernen und künstlicher Intelligenz bei der Datenanalyse. (Bild: KIT)

Modular extensions for maximum flexibility

From data storage, analysis and visualization to long-term archiving, Kadi4Mat enables efficient and transparent management of scientific data. With a modular architecture, open interfaces and AI-supported evaluation methods, the platform helps to optimize research processes, make data usable in the long term and promote interdisciplinary collaboration. Since the project was presented in NEULAND Magazine 2020, the research team has been working hard over the past few years to further develop the platform. In addition to the tried-and-tested documentation and data management in the form of an electronic laboratory journal (ELN 2.0), completely new program modules have been introduced to document, control and automate the entire research process. Specialized modules, such as KadiStudio for workflow management and KadiAI for machine learning, enable flexible adaptation to different application requirements. These innovations lead to significant efficiency and automation gains: Processes such as the automatic recording of test parameters and the start of test series significantly reduce manual effort.

International community and security relevance

Kadi4Mat has been used successfully in various research projects and use cases since its inception. In 2024, the first stable version 1.0 of the virtual open source research environment was officially made available. In addition, the Kadi4Mat team is pursuing an intensified approach to internationalization with the aim of building a global development community and promoting interdisciplinary exchange. New security concepts that cover both cloud and local operation and include comprehensive rights and role management have also been established.

For more detailed information on the current project status, please refer to the recently published, detailed blog post on the RESEARCH TO BUSINESS platform.

Conceptual overview of Kadi4Mat with currently five software modules: (1) KadiWeb, a web-based virtual research environment with ELN functionalities and repositories, and (2) KadiStudio, a desktop-based software version that enables the formulation and execution of workflows. (3) KadiFS for connecting any external software. (4) KadiAPY for integration into Python programs and scripts. (5) KadiAI for using machine learning and artificial intelligence to evaluate data. (Image: KIT)

The image shows a hand pointing to a printed document with diagrams, while digital data visualizations are projected above it. It illustrates technology-supported research data management with a focus on data analysis and visualization using Kadi4Mat. (Image: Arnd Koeppe / KIT)

Efficient research data management: A data infrastructure for the entire research process such as Kadi4Mat supports the acquisition of new findings. The integration of digital analysis tools facilitates the evaluation and visualization of complex data sets. (Image: Arnd Koeppe / KIT)

Schematic sequence of a spectral data analysis as an example: In materials science, the characterization of complex materials often leads to large and complex data sets that are difficult to analyze using conventional methods. Machine learning (ML) has proven to be a powerful tool for the efficient handling and interpretation of such data. KadiAI lends itself to the processing of such tasks. (Image: KIT)

Example application for Kadi4Mat: In materials science, the characterization of complex materials often leads to large and complex data sets that are difficult to analyze using conventional methods. Machine learning (ML) has proven to be a powerful tool for the efficient handling and interpretation of such data. KadiAI lends itself to the processing of such tasks. (Image: KIT)

Provider:	kit-neuland.de
Cookiename:	PHPSESSID
Runtime:	Session

Provider:	TYPO3
Cookiename:	waconcookiemanagement
Runtime:	1 Tag