The project, funded by the German Federal Ministry of Education and Research (grant 01QE1957C) and carried out from 1 March 2020 to 28 February 2023, aimed to create IDEAL (Intelligent Design for Automated Laboratories), a software platform that fully integrates and automates high‑throughput robotic systems for biotechnological production. IDEAL was designed to control a wide range of bioreactors, streamline data acquisition, and apply advanced algorithms for process development and optimisation. A key feature is a modelling framework that can simulate and optimise the production process for different biological organisms, thereby enabling predictive control and improved product quality.
The platform supports interoperability with diverse analytical instruments and data standards, providing rapid and user‑friendly communication between devices and operators. It incorporates a platform‑agnostic control layer, “laboperator”, supplied by labforward GmbH, which manages all laboratory equipment, and an electronic lab book, “labfolder”, that digitises experimental records. DataHowLab, developed by DataHow AG, supplies process analytics and model‑based optimisation tools that feed into IDEAL’s decision‑making engine. Together, these components allow the system to operate autonomously, from sample handling to real‑time monitoring and adjustment of process parameters.
A central part of the work was the development of model‑predictive control (MPC) algorithms that adaptively regulate key process variables. These MPC methods were validated in a case study executed on the TU Berlin robotic platform and subsequently tested on commercial robotic systems. The case study demonstrated that IDEAL could reduce manual intervention, shorten cycle times, and enhance reproducibility, although specific quantitative performance figures were not reported in the final report. The software’s ability to integrate heterogeneous data sources and to model complex biological processes positions it as a versatile tool for future bioprocess development.
The project built on earlier initiatives, notably LEAN‑PROT and Bio‑RAPID, which had established foundational robotic and digital technologies for the “lab of the future”. By extending these concepts, IDEAL delivers a more comprehensive automation solution that reduces the need for human programming of robots and analytical instruments, thereby increasing overall laboratory throughput.
Collaboration was structured around three main partners. The Technical University of Berlin’s Faculty of Bioengineering (FG Bioverfahrenstechnik) provided the robotic platform, integrated the software frameworks, and conducted the case study. DataHow AG contributed the analytical software suite DataHowLab, while labforward GmbH supplied the device‑control software laboperator and the electronic lab book labfolder. The project’s deliverables included not only the software and algorithms but also marketing materials intended to promote IDEAL to potential industrial users.
Results from the project were disseminated through several peer‑reviewed publications, including articles in Bioengineering and Biotechnology & Bioengineering, and presented at international conferences such as ESCAPE, DECHEMA, AIChE, and ECCE/ECAB. The MPC methods developed are planned for use in subsequent research projects, including the KIWI‑biolab initiative, and will serve as a foundation for future third‑party funding proposals. The project’s outcomes lay the groundwork for commercial deployment of fully automated, data‑driven bioprocess laboratories.