Regulatory authorities accept virtual testing as design verification evidence when the simulation model has been validated against physical test data confirming its accuracy for the intended use. EASA accepts FEA-based structural analysis as design verification evidence under CS-25 Appendix H when the analysis method is validated against test data and the validation is documented in the analysis report. IATF 16949 accepts simulation analysis as DVP&R evidence when the simulation tool is validated and the analysis is performed by a competent engineer. For ISO 26262 functional safety, simulation analysis is accepted as a verification method at the appropriate ASIL level when the tool is qualified under ISO 26262 Part 8 software tool qualification. EMUG’s VIRT Framework produces the model validation documentation that satisfies each of these acceptance requirements.
Model validation is the process of confirming that a computational simulation model predicts physical behavior accurately enough for its intended use as design verification evidence. Validation is performed by comparing simulation predictions against measured physical test data from representative specimens under defined loading and boundary conditions, and confirming that the difference between simulation and measurement is within an accepted tolerance. For structural analysis, EMUG applies a correlation tolerance of within 10 percent for stress predictions compared to strain gauge measurements, and MAC (Modal Assurance Criterion) above 0.9 for mode shape correlation in NVH models. Without model validation, simulation results are engineering estimates — useful for design guidance but not acceptable as verification evidence to regulatory authorities who require a documented basis for trusting the simulation’s accuracy.
The EMUG VIRT Framework is EMUG’s five-phase virtual testing delivery methodology, standing for: Validate model, Integrate analysis plan, Rigorous simulation execute, Test correlation confirm, and Report deliver. It ensures simulation evidence quality through three structural mechanisms: model validation is the mandatory first phase — no virtual testing proceeds without a documented validation study confirming model accuracy. Analysis plans are linked to specific DVP&R requirements before analysis execution — ensuring every simulation generates evidence that is traceable to a requirement. Reports are formatted as formal verification records with documented assumptions, limitations, and confidence basis — not engineering calculation notes, but documents that can be reviewed by a regulatory auditor and accepted as verification evidence.
Crash simulation for automotive occupant protection uses explicit finite element analysis — where the simulation solves the crash event in small time steps to capture the dynamic structural response during the crash pulse. EMUG uses Abaqus/Explicit, LS-DYNA, and PAM-CRASH for vehicle crash simulation, with simulation models including structural components, occupant restraint systems (seatbelt, airbag), and validated dummy models (Hybrid III, THOR) for occupant kinematics. Crash simulation models are validated against physical sled test and barrier crash test data confirming that the model correctly predicts structural deformation, occupant kinematics, and injury metrics (HIC, chest deflection, femur load). Validated crash simulation models are used for design development — exploring design variants and optimizing structural crashworthiness before final physical crash test confirmation for regulatory type approval.
FEA (Finite Element Analysis) is the computational method used for solid mechanics problems — structural stress, strain, and deformation; fatigue life; vibration and NVH; and thermal conduction through solid components. CFD (Computational Fluid Dynamics) is the computational method used for fluid flow problems — external aerodynamics (drag, lift, cooling airflow), internal flow (HVAC distribution, cooling circuit, lubrication), heat transfer between fluids and solid surfaces, and combustion. Most engineering products require both: a vehicle needs FEA for structural integrity and crash, and CFD for aerodynamic drag, cooling system performance, and cabin HVAC. EMUG delivers both FEA and CFD virtual testing within integrated programs — using ANSYS Mechanical for structural FEA and ANSYS Fluent for CFD, enabling the simulation team to share model geometry and boundary condition data between disciplines efficiently.
Fatigue virtual testing addresses the requirement to confirm that a component will survive a defined number of load cycles — typically the equivalent of the product’s design life converted to a laboratory test cycle count. EMUG performs fatigue virtual testing using the stress-life (S-N) approach for high-cycle fatigue (above 100,000 cycles), the strain-life (epsilon-N) approach for low-cycle fatigue, and multi-axial fatigue criteria (von Mises, Findley) for components under combined loading states. Loading input comes from customer-provided durability load spectra (measured road load data or standard customer endurance profiles) or from random vibration PSD inputs for components subject to vibration fatigue. Analysis tools include fe-safe (Dassault Systemes), nCode DesignLife (HBK), ANSYS Fatigue Module, and MSC Nastran fatigue. Fatigue virtual test evidence is accepted by automotive OEM customers for component DVP&R closure where model correlation data satisfies the OEM’s virtual testing protocol.
Yes. Parametric design studies are a core capability in EMUG’s virtual testing programs — systematically varying key design parameters (wall thickness, material grade, fillet radius, weld quality class) and analysing the effect on performance metrics (safety factor, fatigue life, natural frequency, heat rejection) to produce a design sensitivity map that shows which parameters most affect performance. For design optimisation, EMUG combines parametric studies with ANSYS optiSLang or direct Python-scripted parameter sweeps to identify the design configuration that best satisfies all performance requirements simultaneously. Parametric studies performed in virtual testing programs reduce the number of physical test iterations required — the physical test is performed on the optimised design configuration, not on intermediate design variants that would have been eliminated by earlier virtual screening.
EMUG delivers virtual testing to automotive OEMs and Tier 1 suppliers (IATF 16949, Euro NCAP, ECE, FMVSS), aerospace and defense organizations (EASA CS-25, FAA 14 CFR Part 25, DO-178C correlation evidence), industrial machinery manufacturers (EU Machinery Directive, ATEX, PED), energy and oil and gas companies (ASME VIII DBA, DNV, API), and engineering services firms. Delivery countries include Germany, France, UK, Netherlands, Sweden, Italy, Spain, Poland, Czech Republic, UAE, Saudi Arabia, Qatar, Kuwait, Bahrain, India, China, Japan, South Korea, Malaysia, Thailand, USA, Canada, Mexico, Brazil, South Africa, Nigeria, and Kenya.
