Name: 5th Automotive CFD Prediction Workshop
Start: 2026-10-08
End: 2026-10-09
Location: DAMA Technopole, Bologna, Italy

Announcements

Aug 14, 2026	The AutoCFD5 results data submission deadline has been extended from 21 August to 4 September 2026, giving participants more time to complete and upload their spreadsheets. Sign in and submit through the AutoCFD5 results dashboard, then see all key dates.
Jul 17, 2026	The AutoCFD5 agenda has been updated. View the latest AutoCFD5 workshop schedule.
Jun 09, 2026	Abstract deadline now extended to June 15th.

View all AutoCFD5 announcements

Key dates and next steps

Current status: Abstract submissions and early-bird registration are now closed. The results data submission deadline has been extended to 4 September 2026, final registration closes on 1 October 2026, and the workshop takes place 8–9 October 2026.

The results dashboard requires sign-in and includes submission instructions and a Dropbox link for auxiliary files.

Aims and scope

AutoCFD5 will take place on 8–9 October 2026 at DAMA Technopole, Bologna, Italy, with Upstream CFD as the main event organiser, CINECA providing the venue and local organisational support, and EVCM as financial administrator. This follows the success of the previous four workshops: AutoCFD1 in Oxford (2019), AutoCFD2 in Berlin (2021), AutoCFD3 in Barcelona (2022) and AutoCFD4 in Belfast (2024).

The main objective of the 5th Automotive CFD Prediction Workshop is to continue the community effort to assess the predictive capability of CFD codes for road-car geometries. Using prescribed geometry, boundary conditions and computational grids, the workshop aims to provide practical modelling guidance for the automotive community, including best practice for turbulence modelling, meshing and numerical schemes. It also brings together automotive CFD researchers and practitioners from academia and industry to discuss future directions. The AutoCFD5 test cases build on the benchmark configurations used for AutoCFD4. For the AI/ML activity, the open-source DrivAerML dataset is the standard training dataset used to assess model architectures.

Computing the test cases is not required to attend the workshop, but participants are strongly encouraged to submit results.

Technology Focus Groups

Following positive feedback from AutoCFD4, the workshop continues to use the Technology Focus Group (TFG) model. Each participant is invited to join a TFG covering one of five core areas. The groups meet every four to five weeks to share findings and results collaboratively in the run-up to the workshop. This enables deeper technical discussion over a longer period than the two workshop days alone. Contact the relevant TFG leaders below to join the meetings, which began in mid-2025.

Meshing TFG - Vangelis Skaperdas (Cadence / BETA-CAE Systems)
Noise Factors TFG - Burkhard Hupertz (Ford)
AI/ML TFG - Neil Ashton (NVIDIA) and Astrid Walle (Pasteur Labs)
Scale Resolving Simulations (SRS) TFG - Charles Mockett and Marian Fuchs (Upstream CFD)
HPC TFG - Herbert Owen and Oriol Lehmkuhl (Barcelona Supercomputing Center)

Read the Technology Focus Group scope and goals.

Read the AutoCFD data policy.