Governance & Audit Trails
Commercial shipping operates under increasingly stringent regulatory requirements, ranging from IMO decarbonisation targets to EU emissions trading. Expanse automatically captures provenance data for every simulation, enabling full reproducibility and compliance with the regulatory frameworks that companies like Oldendorff, Maersk, and other major fleet operators must satisfy.
Regulatory Landscape for Shipping
Computational simulations for hull design, propeller optimisation, and operational efficiency must be traceable and auditable to satisfy:
IMO Regulations
| Regulation | Relevance to Simulation |
|---|---|
| EEDI / EEXI | Energy Efficiency Design/Existing Ship Index; CFD resistance predictions must be auditable for attained EEDI calculations |
| CII | Carbon Intensity Indicator; operational efficiency simulations feed into annual CII ratings |
| MEPC.1/Circ.850 | Guidelines for CFD verification and validation in EEDI calculations |
| IMO 2030/2050 | Decarbonisation targets require documented R&D for alternative fuels, wind-assist, air lubrication |
| MARPOL Annex VI | Air pollution limits; simulations supporting scrubber design, LNG/ammonia fuel systems |
EU Regulations
| Regulation | Relevance to Simulation |
|---|---|
| EU ETS | Emissions Trading System (from 2024); fuel consumption predictions must be defensible |
| EU MRV | Monitoring, Reporting, Verification; computational methods supporting reported data require audit trails |
| FuelEU Maritime | GHG intensity targets (from 2025); alternative fuel simulations must be reproducible |
Classification Society Requirements
| Standard | Description |
|---|---|
| IACS UR S11A | Unified requirements for computational methods in structural assessment |
| DNV-CG-0039 | Computational Fluid Dynamics guidelines for maritime applications |
| DNV-ST-F101 | Submarine pipeline systems; requires traceable computational verification |
| Lloyd's ShipRight | Structural design assessment procedures with CFD/FEA requirements |
| Bureau Veritas NR 467 | Rules for steel ships; computational method verification |
Why Provenance Matters for Fleet Operators
For companies like Oldendorff operating large fleets of bulk carriers:
- EEDI Compliance: Demonstrate that newbuild designs meet required EEDI thresholds with auditable CFD evidence
- EEXI Verification: Prove that retrofits (bulbous bow modifications, propeller upgrades) achieve claimed efficiency gains
- CII Improvement: Document simulations supporting operational measures (trim optimisation, hull coating selection)
- Charter Party Disputes: Traceable speed/consumption predictions for performance warranty claims
- Insurance & P&I: Computational evidence for hull stress analysis, fatigue assessment
- Due Diligence: Documented R&D for ESG reporting and green financing requirements
What's Captured
For every job submission, Expanse records:
| Data Point | Description |
|---|---|
| Source code hash | SHA-256 of all source files at submission time |
| Input artifacts | Hashes and paths of all input data (hull geometry, meshes, boundary conditions) |
| Solver configuration | Complete solver settings, turbulence models, discretisation schemes |
| Job parameters | Resource requests, environment variables, CLI arguments |
| User identity | Authenticated user who submitted the job |
| Timestamps | Submit time, start time, end time |
| Runtime metrics | Actual resource usage, exit codes, convergence data |
| Output artifacts | Hashes of all result files for tamper detection |
MEPC.1/Circ.850 Compliance
IMO's guidelines for CFD in EEDI calculations (MEPC.1/Circ.850) require:
- Verification: Documented mesh convergence studies, numerical uncertainty quantification
- Validation: Comparison against model test data with stated accuracy
- Documentation: Complete record of computational setup, boundary conditions, assumptions
Expanse captures all of this automatically:
# Generate MEPC.1/Circ.850 compliance report
expanse jobs report <job-id> --format mepc-850 --output eedi_cfd_report.pdf
Reproducibility Certificate
For each completed simulation, Expanse generates a reproducibility certificate suitable for regulatory submissions:
expanse jobs certificate <job-id> --output certificate.pdf
This certificate can be attached to:
- EEDI Technical File submissions to classification societies
- CII correction factor applications
- Flag state approval requests
- Charter party performance documentation
Viewing Audit Logs
# View recent job history
expanse jobs list
# Get full provenance for a specific job
expanse jobs show <job-id> --provenance
# Export audit trail for compliance
expanse jobs export --format json --output audit.json
# Export in classification-society formats
expanse jobs export --format dnv --output dnv_submission.xml
expanse jobs export --format lloyds --output lloyds_audit.json
expanse jobs export --format imo-mepc --output mepc_documentation.pdf
Example: EEDI Resistance Prediction Audit Trail
$ expanse jobs show job-2024-1234 --provenance
Job: job-2024-1234
Workflow: eedi_resistance_prediction
Status: COMPLETED
Duration: 6h 12m 41s
Compliance: MEPC.1/Circ.850 ✓
Provenance Chain:
─────────────────
1. hull_preparation (job-2024-1231)
Input: lines_plan.iges SHA256: 8f4a2b...
Input: appendages.step SHA256: 2d7c9a...
Output: watertight_hull.stl SHA256: 3c91e7...
2. mesh_generation (job-2024-1232)
Input: watertight_hull.stl SHA256: 3c91e7... ✓
Output: volume_mesh_12M.cgns SHA256: 7b2e4f...
Output: volume_mesh_24M.cgns SHA256: 9a1d8c...
Output: volume_mesh_48M.cgns SHA256: 5f3e2a...
3. mesh_convergence_study (job-2024-1233)
Input: volume_mesh_*.cgns [3 meshes]
Output: gci_uncertainty.json SHA256: 1c4d7e...
Result: GCI = 1.2% (fine mesh), Grid converged ✓
4. resistance_cfd (job-2024-1234)
Input: volume_mesh_48M.cgns SHA256: 5f3e2a... ✓
Input: boundary_conditions.json SHA256: a72f1d...
Input: solver_config.yaml SHA256: 91bc4e...
Solver: OpenFOAM 2312
Turbulence: k-omega SST (Menter 1994)
Wall treatment: y+ < 1, resolved boundary layer
Mesh cells: 48.2M
Froude number: 0.182
Output: resistance_forces.csv SHA256: 6d8f2a...
Output: wave_pattern.vtk SHA256: 4a2d9f...
Output: pressure_distribution.vtk SHA256: 8c1b3e...
Results:
Total resistance (Rt): 1,247 kN
Numerical uncertainty: ±1.2% (GCI)
User: j.smith@strath.ac.uk
Cluster: archer2
Submitted: 2024-12-15T08:14:22Z
Completed: 2024-12-15T14:27:03Z
Verification: All input hashes verified against upstream outputs ✓
MEPC.1/Circ.850 checklist: 14/14 requirements satisfied ✓
Long-Term Archival
Vessel design documentation must be retained for the ship's operational lifetime (typically 25-30 years). Configure long-term archival:
# project.yaml
governance:
retention: 30y # Vessel design life + margin
archive_outputs: true # Archive all result files
archive_storage: s3://archive-bucket/naome-projects/
compliance_standards:
- IMO-MEPC-850
- IACS-UR-S11A
- DNV-CG-0039
- EU-MRV
# Automatic compliance checking
require_mesh_convergence: true
require_validation_case: true
minimum_documentation_level: full
Expanse will automatically:
- Archive completed jobs with full provenance to long-term storage
- Generate compliance reports for configured standards
- Alert if jobs don't meet minimum documentation requirements
- Maintain chain-of-custody records for regulatory audits
For Industry Partners
These governance features ensure that research simulations can seamlessly transition to commercial applications. When Oldendorff or other fleet operators need to reference university research in regulatory submissions, the full provenance chain is already in place, so no retroactive documentation is required.