Defensible Random Vibration Analysis Workflow

What Is It?

A defensible random vibration analysis workflow is a structured, traceable process that takes an engineering question about random vibration response and produces a credible, documented answer. It covers every step from defining the vibration environment through to reporting the results, with verification and validation at each stage. The workflow ensures that the analysis is correct, the results are credible and the evidence is traceable.

Why It Matters

Random vibration analysis is used for critical engineering decisions — equipment qualification, structural substantiation, fatigue life prediction. A defensible workflow ensures that these decisions are based on credible analysis, not on unverified predictions. It provides the evidence chain that an independent reviewer can follow to verify the analysis and confirm the conclusions. Without a defensible workflow, the analysis is an assertion; with it, the analysis is evidence.

A defensible workflow transforms an analysis from an assertion into evidence. It provides the traceable chain that an independent reviewer can follow. Every step is documented, verified and connected to the engineering decision.

The Complete Workflow

The defensible random vibration analysis workflow follows a defined chain from environment to report. Each step builds on the previous one, and each produces documented output that feeds the next step.

Environment definition → PSD definition → Support/input definition → Modal analysis → Damping → Random response solution → RMS/peak interpretation → Stress/fatigue assessment → Qualification test planning → Correlation → Verification → Reporting

Environment Definition

The first step is defining the vibration environment. What is the source of the random excitation? What is the expected PSD? Is the environment stationary or non-stationary? What is the duration? The environment definition comes from measured data, standards, customer specifications or analytical predictions. The source and basis of the environment must be documented.

1. Identify the vibration source — acoustic, mechanical, road, launch, machinery
2. Obtain the PSD specification — from measurement, standard or prediction
3. Determine if the environment is stationary or non-stationary
4. Define the duration — how long the excitation persists
5. Document the source and basis of the environment definition

PSD Definition

The PSD is the quantitative definition of the random vibration environment. Verify the PSD breakpoints, slopes, frequency range and units. Compute the Grms and verify it is reasonable. Confirm the PSD is in the correct form (g²/Hz, one-sided) for the analysis tool. Document the PSD and its source.

• Verify PSD breakpoints, slopes, frequency range and units
• Compute Grms — verify it is reasonable for the environment
• Confirm the PSD form (g²/Hz, one-sided) matches the analysis tool
• Document the PSD and its source

Support/Input Definition

Define how the PSD is applied to the structure. Is it base excitation at a single support, multiple supports, or force PSD? Are the supports correlated? What are the boundary conditions? The support/input definition determines how the environment is transferred to the structure. Document the support conditions and their basis.

Modal Analysis

Run the modal analysis to extract natural frequencies, mode shapes and modal parameters. Check the cumulative effective mass in each excitation direction — target >90%. Check for local modes at critical locations. Verify the modal frequencies are plausible. The modal analysis is the foundation — if it is wrong, everything downstream is wrong.

• Extract modes covering the PSD frequency range plus margin (1.2-1.5×)
• Check cumulative effective mass >90% in each excitation direction
• Check for local modes at critical locations
• Verify modal frequencies are plausible (hand calculations, experience)
• Document the modal basis and adequacy checks

Damping

Specify the damping for each mode. Use test data from a similar structure if available. Without test data, use conservative values from standards or experience. Run a sensitivity study with at least two damping values. Document the damping source, values and sensitivity results. Damping is the most influential parameter — treat it with appropriate rigour.

Damping is the most influential parameter. Use the best available source (test data). Run sensitivity studies with at least two values. Document the source and values. Never use an arbitrary single value without justification.

Random Response Solution

Run the random response solution. Specify the output frequency range to match the PSD input range. Use sufficient frequency resolution to resolve resonant peaks. Compute response PSDs, RMS acceleration, RMS displacement and RMS stress at critical locations. Verify the solution by checking that response PSD peaks align with natural frequencies.

RMS/Peak Interpretation

Interpret the RMS and peak response statistically. The RMS is a standard deviation, not a maximum. Use sigma levels (3σ for screening) or duration-specific peak factors for peak estimation. State the statistical basis of any reported peak. Do not treat 3σ as a universal maximum — for long durations, use higher levels or duration-specific estimates.

• RMS = standard deviation, not maximum
• 3σ for screening; duration-specific peak factor for assessment
• State the statistical basis of any reported peak
• For long durations, 3σ is non-conservative — use higher levels

Stress/Fatigue Assessment

Recover stress PSDs at critical locations. For screening, use 3σ stress against yield or ultimate. For fatigue, use the stress PSD with frequency-domain methods (narrow-band, Dirlik) or time-domain rainflow. Run damping sensitivity for fatigue — life can vary by orders of magnitude. Validate frequency-domain results against time-domain for critical components.

• Recover component stress PSDs at critical locations
• Screening: 3σ stress vs yield or ultimate
• Fatigue: stress PSD → spectral methods or rainflow → S-N → Miner
• Run damping sensitivity — fatigue life is extremely damping-sensitive
• Validate frequency-domain fatigue against time-domain for critical components

Qualification Test Planning

If qualification testing is required, plan the test based on the analysis. Define the test PSD, duration, axes and acceptance criteria. Identify critical response locations for accelerometer placement. Determine if notching is needed based on predicted interface loads or response limits. Design and qualify the test fixture. Prepare the test plan and instrumentation list.

Correlation

After the test, correlate the analysis with the test data. Compare modal frequencies, mode shapes and damping. Compare response PSDs and RMS values. If the correlation is poor, investigate and update the model. Document the correlation results and any model updates. The correlation provides the validation evidence for the analysis.

Verification

Verify the analysis at every stage. Check modal frequencies, effective mass, PSD input, response units, RMS integration, damping, reactions and analytical comparison. Document all verification checks. Verification is not optional — it is the process that confirms the analysis is correct. See the random vibration model verification article for the full checklist.

Verification at every stage. Check modal frequencies, effective mass, PSD input, units, RMS integration, damping, reactions. Document all checks. Verification confirms the analysis is correct — it is not optional.

Reporting

Document the complete analysis in a report. The report should include: the engineering question, the environment definition, the PSD and its source, the model description, the modal analysis and adequacy checks, the damping and its basis, the random response results, the stress and fatigue assessment, the verification checks, the correlation results (if available), the conclusions and the recommendations. The report is the evidence — it should be sufficient for an independent reviewer to verify the analysis.

• Engineering question and scope
• Environment definition and PSD (with source)
• Model description and modal analysis (with adequacy checks)
• Damping (with source, values and sensitivity)
• Random response results (PSD, RMS, peaks with statistical basis)
• Stress and fatigue assessment (with method, S-N data, damage)
• Verification checks (all documented)
• Correlation results (if available)
• Conclusions and recommendations

Key Takeaways

• A defensible workflow provides a traceable chain from environment to engineering decision
• Each step produces documented output that feeds the next step and the final report
• Modal analysis is the foundation — verify adequacy before proceeding
• Damping is the most influential parameter — use best available source, run sensitivity
• Verification at every stage, correlation with test, comprehensive reporting — the evidence chain