Forensic Engineering — Engineering Reference

1. At a glance

Forensic engineering applies engineering principles, methods, and analysis to investigate failures, accidents, injuries, and disputes after the fact — typically in support of litigation, insurance, regulatory action, or post-incident corrective action. Where design engineering asks “will it work?”, forensic engineering asks “why did it not work, who is responsible, and what must change?”

Sub-fields:

• Structural failure — building, bridge, tower, dam, scaffolding collapse; foundation movement.
• Materials / fractography — metallurgical and polymer failure, fracture surface interpretation.
• Automotive accident reconstruction — crash dynamics, occupant kinematics, vehicle defect.
• Aviation accident investigation — airframe, powerplant, avionics, human factors.
• Fire and explosion investigation — origin, cause, fuel load, ignition source.
• Product liability — consumer goods, medical devices, machinery, recreational equipment.
• Geotechnical — slope stability, landslide, settlement, retaining wall, dam failure.
• Marine — hull, mooring, propulsion, cargo shift, collision and grounding.
• Process safety — refinery, petrochemical, pharmaceutical, food, mining incidents.
• Electrical / electronic — arc fault, lithium-ion thermal runaway, transformer, switchgear.
• Software / control systems — embedded, autonomous, medical, avionics control failures.

Practitioners often double as expert witnesses in civil and criminal litigation. The dominant professional bodies are the ASCE Forensic Practices Committee, the National Academy of Forensic Engineers (NAFE) (affiliated with NSPE), and ASME’s reliability and pressure-vessel divisions. The core consensus standard for forensic practice is ASTM E2018-15 (Property Condition Assessments), with ASTM E620 governing expert testimony reports.

Forensic engineering is inherently interdisciplinary — a single bridge collapse may require structural, geotechnical, metallurgical, welding, fire, traffic, and human-factors specialists working in parallel. The investigator’s role is to assemble evidence rigorously enough that the conclusions survive cross-examination under the Daubert standard (Section 11).

2. Failure analysis methodology — the 8-step ASM approach

The canonical reference is ASM Handbook Volume 11: Failure Analysis and Prevention (2nd ed, 2021). Its 8-step protocol is industry standard across metallurgical, polymer, ceramic, and composite failures:

(a) Collect background, service history, and design data. Drawings, specifications, material certifications, weld procedures, heat-treatment records, service environment (temperature, pressure, chemistry, cycling), maintenance logs, prior repairs, operating manuals. Interview operators and maintenance personnel before memories degrade.

(b) Preserve evidence — do not damage; photograph at scene. Fracture surfaces must not be touched, oiled, wire-brushed, or fitted back together. Each fragment is bagged, labeled with chain-of-custody, and photographed in situ (overall context, then approach shots, then macro detail with scale bar). Spoliation of evidence is a tort and a discovery sanction risk.

(c) Visual examination and macro-fractography. Stereo microscope at 5–50×. Identify origin, crack-propagation direction (chevron marks point back to origin; river patterns flow downstream from origin), beach marks, ratchet marks, final fast-fracture zone, secondary cracks. Document with annotated photographs.

(d) Material characterization. Chemical composition (OES, XRF, ICP-MS); microstructure (metallography after sectioning a non-critical area); mechanical properties (tensile, hardness, Charpy V-notch impact). Compare measured values to specification — a failed Grade 8.8 bolt that tests as Grade 4.6 indicates a substitution or counterfeit.

(e) Non-destructive testing (NDT). UT (ultrasonic), RT (radiography), PT (dye penetrant), MT (magnetic particle), ET (eddy current), VT (visual), AE (acoustic emission). Choose method by anticipated defect type and material — covered in [[Engineering/Tier3/ndt-methods]].

(f) Mechanical, chemical, and microscopy testing. SEM with EDS for fracture surface, fatigue striation counting, corrosion-product identification. XRD for crystalline phases. TEM for dislocation substructure or precipitate analysis in high-end cases.

(g) Identify failure mode and mechanism. Mode = the how (fatigue, brittle fracture, SCC, creep, wear, overload). Mechanism = the why at the atomic / microstructural level (transgranular cleavage from a stress concentrator at temperature below DBTT, e.g.). These must be distinguished — they are different answers to different stakeholders.

(h) Conclusions and recommendations. Probable cause stated to a reasonable degree of engineering certainty. Contributing factors enumerated. Recommendations for design, material, process, inspection, or operational change. A good forensic report distinguishes fact, opinion, and recommendation, and addresses alternative hypotheses considered and ruled out.

3. Fracture surfaces and fractography

Fracture surfaces are the most information-dense single artifact in a mechanical failure investigation. Interpretation requires both macro (eye, 5–50× stereo) and micro (SEM, 100–10000×) inspection.

Ductile vs brittle fracture

• Ductile — microvoid coalescence around inclusions or second-phase particles; SEM shows equiaxed dimples (tension) or elongated/parabolic dimples (shear). Macro: 45° shear lips, fibrous matte appearance, evidence of plastic deformation.
• Brittle transgranular — cleavage facets along crystallographic planes (typically {100} in BCC ferrite), river patterns converging on origin, tongues and steps. Common in ferritic steel below the ductile–brittle transition temperature (DBTT).
• Brittle intergranular — fracture along grain boundaries; SEM shows rock-candy texture of separated grains. Causes: temper embrittlement, hydrogen embrittlement, SCC, creep cavitation, intergranular corrosion (sensitized stainless).

Fatigue signatures

Fatigue is the single largest cause of in-service mechanical failure. Signatures by scale:

• Origin point — often at a stress concentrator: a notch, weld toe, threaded transition, fretting site, machining mark, corrosion pit, or inclusion.
• Ratchet marks — radial steps between adjacent origin sites under high stress or sharp stress concentration; multiple origins indicate severe loading or poor surface finish.
• Beach marks (macroscale, visible to eye) — concentric arcs around origin marking changes in stress amplitude, load reversal, or environment (e.g., shutdown periods). One beach mark per significant operational event.
• Striations (microscale, SEM 1000–10000×) — one per fatigue cycle, perpendicular to crack-propagation direction. Spacing gives da/dN directly. Not all fatigue produces visible striations (high-strength martensitic steels often do not).
• Final fast-fracture zone — when remaining cross-section reaches Kc, terminal overload occurs (ductile or brittle as appropriate). Ratio of fatigue area to fast-fracture area indicates stress level: small fatigue + large overload = high stress; large fatigue + small overload = low stress.

Environmentally assisted cracking

• Stress corrosion cracking (SCC) — synergy of tensile stress, susceptible alloy, and specific environment. Often branched, intergranular cracks. Classic combinations: 304 stainless + chlorides (e.g., swimming-pool roof collapses, Uster Switzerland 1985 and chloride-induced 316 failures); brass + ammonia; high-strength steel + H2S (sulfide SCC).
• Hydrogen embrittlement (HE) — atomic hydrogen diffuses into high-strength steel (>1000 MPa UTS) or titanium, causing delayed intergranular fracture under sustained load. Sources: cathodic protection over-potential, acid pickling, electroplating without proper bake-out, sour gas service. Distinguishable from SCC by absence of corrosion product and time-delay characteristic.
• Liquid metal embrittlement — mercury + aluminum; gallium + aluminum; zinc + stainless steel (galvanizing-induced cracking).

Creep

Time-dependent deformation at T > ~0.4 Tm. Fracture surface shows intergranular cavitation (voids on grain boundaries, especially perpendicular to maximum principal stress), wedge cracks at triple junctions, and grain-boundary sliding. Sub-divided into primary, secondary (Larson-Miller parameter), and tertiary stages. Critical for power-plant tubing, gas-turbine blades, and chemical reactors.

Wear and tribological failure

• Abrasion — hard particle removes material; grooves parallel to motion.
• Adhesion / galling — surface welding and pull-out; smeared metal, common in stainless–stainless sliding.
• Erosion — impingement of particles or droplets; characteristic angle-dependent patterns.
• Fretting — micro-slip at clamped joints; cocoa-brown debris (Fe2O3) and pitted surface.

SEM examination

Secondary electron (SE) imaging for topography; backscatter (BSE) for compositional contrast; EDS for elemental analysis of inclusions, deposits, and corrosion products. Modern field-emission SEM (Zeiss Sigma, FEI Helios, JEOL JSM-7900F) routinely reaches 1 nm resolution and integrates EBSD for grain-orientation mapping — useful for cleavage-plane identification.

4. Root Cause Analysis (RCA) methods

A fracture surface tells what failed; RCA explains why the system allowed it. The forensic engineer’s deliverable is rarely complete without a systemic root-cause picture.

• 5 Whys — Toyota Production System; iterative “why?” until a controllable root is reached. Best for simple linear causality.
• Fishbone / Ishikawa diagram — categorizes potential causes (Man, Machine, Method, Material, Measurement, Environment — “6 Ms”). Useful brainstorming aid.
• Fault Tree Analysis (FTA) — top-down deductive Boolean logic from undesired top event through AND/OR gates to basic events. Quantifiable with component failure rates. Origin: Bell Labs, Minuteman ICBM, 1962.
• Event Tree Analysis (ETA) — bottom-up inductive from initiating event through success/failure branches of safety functions. Common in nuclear PRA.
• Bow-tie analysis — fault tree (left) joined to event tree (right) at the critical event, with barriers/safeguards labeled. Standard in process safety and aviation safety management.
• Apollo Root Cause Analysis (Dean Gano) — cause-and-effect chart enforcing evidence-backed action/condition pairs.
• TapRooT (System Improvements Inc.) — software-supported, expert-system-guided RCA used by NRC licensees and major refiners.
• Kepner-Tregoe — problem-analysis matrix (is/is not, distinctions, changes) developed in the 1950s.
• MTO (Man + Technology + Organization) — Scandinavian / IAEA framework; explicitly elevates organizational factors equal to technical ones.
• HFACS (Human Factors Analysis and Classification System) — Shappell & Wiegmann; four tiers (unsafe acts, preconditions, supervision, organizational influences) derived from Reason’s Swiss-Cheese model. Standard in US Navy and FAA mishap investigation.

Modern best practice (per NTSB, CSB, IAEA, IOGP) treats single-point human-error conclusions as a failure of the investigation, not an outcome. Reason’s Human Error (1990) and Dekker’s Field Guide to Understanding ‘Human Error’ (3rd ed, 2014) are mandatory reading.

5. Common failure modes by domain

Mechanical

Fatigue dominates: rotating shafts, welded connections, bolted joints (especially preload loss), dissimilar-metal interfaces (galvanic + thermal-expansion mismatch). Brittle fracture in cold service (LNG, arctic pipelines, ship structures). Buckling (Euler + local + lateral-torsional). Galling and wear in unlubricated sliding interfaces. SCC and corrosion fatigue in chemical service.

Civil and structural

Progressive collapse — partial failure propagates through a structure lacking alternate-load-path redundancy:

• Ronan Point (UK, 1968) — gas explosion in 18th-floor flat triggered collapse of one corner of a 22-storey precast concrete tower; led to UK disproportionate-collapse provisions and ultimately to the alternate-load-path requirement in modern codes.
• FIU pedestrian bridge (Florida International University, Miami, March 2018) — accelerated bridge construction (ABC) self-supporting truss; cracked Node 11/12 connection during post-tensioning re-tensioning over live traffic; NTSB HAR-19/02 identified design errors and load-path misjudgment.
• Champlain Towers South / Surfside (Florida, 24 June 2021) — 12-storey residential collapse; 98 fatalities; NIST investigation ongoing through 2024–26; preliminary findings cite long-deferred pool-deck waterproofing, chloride-induced rebar corrosion, concrete spalling, and column connection inadequacies at the parking-garage level.
• Morandi Bridge / Polcevera Viaduct (Genoa, August 2018) — 43 fatalities; cable-stayed prestressed-concrete tower; corroded post-tensioned stay cables inside concrete-encased stays masked from inspection; led to EU bridge inspection reform.

Other structural modes: foundation settlement and differential movement; structural-steel fire damage (post-flashover yield strength loss); concrete spalling under fire; seismic-induced soft-story collapse and pounding.

Aviation

• Aloha Airlines 243 (28 April 1988) — Boeing 737-200 explosive decompression near Maui; cabin roof separation due to multi-site fatigue cracking around fuselage lap joints; cold-bonded cold-bond construction defect; flight attendant lost. NTSB AAR-89/03; drove worldwide aging-aircraft programs and AC 91-56.
• Lauda Air 004 (26 May 1991) — Boeing 767 thrust reverser uncommanded in-flight deployment; 223 fatalities; led to TR redesign.
• ANA 787 / JAL 787 (January 2013) — lithium-ion battery thermal runaway; FAA grounding of the entire 787 fleet; NTSB AIR-14/01.
• Air France 447 (1 June 2009) — Airbus A330 over South Atlantic; pitot icing, AoA confusion, deep-stall descent; BEA report 2012; multi-factor: icing + automation + CRM.
• 787 forward-fuselage fastener / shimming (2019–24 ongoing) — composite delamination and fastener hole quality investigations; FAA-Boeing oversight tightened.

Automotive

• Ford Pinto (1971–80) — fuel-tank rupture in low-speed rear impact; Grimshaw v. Ford Motor Co. (1981) punitive damages; cost-benefit memo controversy.
• Firestone / Ford Explorer tire separation (2000) — tread separation on Wilderness AT tires; ~270 fatalities; congressional hearings; TREAD Act 2000.
• Toyota unintended acceleration (2009–10) — floor-mat entrapment + sticky-pedal mechanical; NASA-led ETC software review found no software cause; $1.2 B DOJ settlement (2014).
• Takata airbag inflator (2014–23) — ammonium-nitrate propellant degrades with humidity/temperature cycling, leading to over-pressure rupture and shrapnel; >100 M inflators recalled; largest automotive recall in history; >30 fatalities worldwide.
• GM ignition switch (2014) — defective Delta detent; engine cutoff during driving, disabling airbags; 124 acknowledged fatalities; $900 M DPA.

Electrical and electronic

• Boeing 787 lithium-ion (2013, Section 5 above).
• Samsung Galaxy Note 7 (2016) — internal separator damage from battery-cell envelope mechanical stress; two design generations failed; global recall, FAA in-cabin ban.
• Tesla Megapack Moss Landing (BESS, 2024 fire) — large-scale lithium-iron-phosphate battery energy storage system fire; thermal-runaway propagation through racks; California PUC and CSB-style review.
• Arc flash in switchgear (per NFPA 70E); semiconductor latch-up; ESD damage to MOS gate oxide; transformer through-fault failure.

Process safety

• BP Texas City refinery (23 March 2005) — isomerization unit blowdown drum overfilled; geyser of hydrocarbon vapor ignited; 15 fatalities; CSB Final Report 2007 (Carolyn Merritt, John Bresland); led to API RP 754 process-safety indicators.
• Deepwater Horizon / Macondo (20 April 2010) — BP/Transocean Gulf of Mexico blowout; 11 fatalities, largest marine oil spill in US history; cement-barrier failure + negative-test misinterpretation + BOP shear-ram failure on drill-pipe joint; National Commission report 2011, CSB volumes 2014–2016.
• West Fertilizer (West, Texas, 17 April 2013) — ammonium-nitrate detonation in fire; 15 fatalities; CSB report 2016.
• Bhopal (Union Carbide India, 2–3 December 1984) — methyl isocyanate release; ~3800 immediate fatalities; cited cause: water ingress to MIC tank, multiple safeguard failures; the worst industrial disaster in history.
• Buncefield (Hertfordshire UK, 11 December 2005) — gasoline tank overfill; vapor cloud explosion; UK Major Incident Investigation Board 2008; revised UK COMAH regulations.

Software and control

• Therac-25 (1985–87) — radiation therapy machine; race condition in mode switching delivered 100× overdose to six patients; foundational case study in safety-critical software (Leveson & Turner, IEEE Computer, July 1993).
• 737 MAX / MCAS (Lion Air JT610, 29 October 2018; Ethiopian ET302, 10 March 2019) — Maneuvering Characteristics Augmentation System tied to single AoA sensor; pilots not trained on MCAS or its disconnect; FAA delegation lapses (ODA); KNKT and Ethiopian AIB reports, JATR report 2019, US House Transportation Committee report 2020, DOT IG audits 2019–2024.
• Ariane 5 Flight 501 (4 June 1996) — inertial reference system reused Ariane 4 software; 64-bit float horizontal-velocity conversion to 16-bit signed integer overflowed at higher velocities; self-destruct at T+39 s; ESA Inquiry Board report 1996.

6. Famous case studies and lessons

• De Havilland Comet (G-ALYP 10 January 1954, G-ALYY 8 April 1954) — world’s first jetliner; square cabin windows; fatigue cracks initiating at rivet holes at window corners under pressurization cycling; led to rounded windows, full-scale pressure-cycle testing of fuselages, and Cohen RAE report 1955.
• Liberty ships (1941–46) — ~1500 of 2710 hulls developed hull cracks, 19 broke in two; brittle fracture of welded mild steel at low temperature with notch effects at hatch corners; established Charpy testing, crack arrestor strakes, riveted seam reintroduction, and ductile-brittle transition temperature in design codes.
• Tacoma Narrows Bridge (Galloping Gertie, 7 November 1940) — aeroelastic torsional flutter at 67 km/h wind in a deck without aerodynamic shaping; led to wind-tunnel testing of all major bridges; modern bridges use box girders, tuned-mass dampers, vortex spoilers.
• Hyatt Regency walkway (Kansas City, 17 July 1981) — suspended skywalks collapsed during tea dance, 114 fatalities, 216 injured; original design used continuous rods through both walkways (each rod nut carries load of one walkway); shop-drawing change split rods into two segments, doubling load on upper-walkway connection; classic case in engineering ethics curricula and design-change communication failure.
• Space Shuttle Challenger / STS-51-L (28 January 1986) — SRB O-ring resilience lost at 31°F; Roger Boisjoly’s pre-launch warning; Rogers Commission Report 1986 + Diane Vaughan’s “normalization of deviance” (The Challenger Launch Decision, 1996).
• Space Shuttle Columbia / STS-107 (1 February 2003) — foam shed from ET bipod ramp at T+82s struck left wing RCC panel 8; re-entry burn-through; Columbia Accident Investigation Board (CAIB) Report 2003 documented organizational drift and “flight readiness review” group-think.
• Deepwater Horizon (2010, Section 5 above) — cement, BOP, organizational; National Commission Report 2011.
• 737 MAX MCAS (2018–19, Section 5 above) — sensor single point + missing training + FAA delegation; 2024 Alaska 1282 door-plug blowout triggered a fresh round of 737 MAX certification and Boeing production-quality reviews.
• Beirut port explosion (4 August 2020) — ~2750 tonnes of ammonium nitrate stored in Warehouse 12 for ~6 years; ignited by adjacent fireworks fire; ~218 fatalities; estimated 1.1 kt TNT-equivalent yield; ongoing Lebanese investigation, no final international report.
• Surfside / Champlain Towers South (2021, Section 5 above) — deferred maintenance, chloride corrosion, waterproofing failure.

7. Automotive accident reconstruction

Scene work

• Survey by total station, robotic theodolite, drone photogrammetry (DJI Mavic 3 Pro, Phantom 4 RTK, WingtraOne), or terrestrial LiDAR (FARO Focus S350, Leica RTC360, Trimble X9, Z+F IMAGER 5016). Modern practice routinely produces sub-cm 3D point clouds in 30–60 min for a typical intersection.
• Reference frame: GNSS RTK to local control; photography with scale bars, north arrow, and overlap for SfM photogrammetry.

EDR — the “black box”

• US FMVSS 405 / 49 CFR Part 563 effectively mandates EDRs in new passenger vehicles. NHTSA’s 2014 rule defines 19 minimum data elements over 5 s pre-crash plus crash-event window: longitudinal Δv, seatbelt status, AOI airbag deployment, throttle %, brake on/off, steering input, ABS activity, RPM, vehicle speed.
• Retrieval via Bosch CDR (Crash Data Retrieval) tool with manufacturer-specific cables; downloaded as hex + report PDF.
• Tesla and other manufacturers store extended telematics (GPS track, Autopilot state, camera frames) in vehicle memory and cloud; subpoena-accessible.

Vehicle dynamics

• Skid mark analysis — Searle (1983) formulation for combined slide and yaw, accounting for braking-and-steering coupling and grade. Modern friction coefficients from drag-sled or accelerometer “drag factor” runs at scene.
• Critical-speed yaw analysis — chord-and-middle-ordinate of curved tire mark gives radius; v = sqrt(μ·g·R). Care with banking and grade.
• Conservation of momentum + energy — 2D linear-momentum analysis for two-vehicle collisions (Marquard / Campbell-Smith); CRASH3 algorithm with crush-stiffness coefficients for energy-equivalent Δv.
• Simulation packages: PC-Crash (Dr. Steffan Datentechnik), HVE (Engineering Dynamics Corp), Virtual CRASH, REC TEC.
• Tire-mark interpretation: ABS scrub vs full-slip skid vs critical-speed yaw vs gouge; surface marks (gouges, scrapes, chips) reveal vehicle interaction post-impact.

Standards

• SAE J224 (collision deformation classification), SAE J670 (vehicle dynamics terminology), SAE J1733 (sign convention for vehicle behavior).
• IIHS and Euro NCAP crash-test data provide benchmark crush stiffness and occupant kinematics.

In court

PE qualifications, ACTAR certification, prior testimony record, peer-reviewed methodology, and conformity to SAE/ASTM are all probed under Daubert.

8. Fire and explosion investigation

Standards

• NFPA 921 Guide for Fire and Explosion Investigations (2024 ed) — the de facto cause-and-origin methodology in US and widely cited internationally. Adopts the scientific method (hypothesis, test, falsify) explicitly.
• NFPA 1033 Standard for Professional Qualifications for Fire Investigator (2024 ed) — investigator competency.
• ASTM E1188 (collection of investigation materials), E1618 (fire-debris GC-MS analysis), E2997 (electrical investigation).

Origin and cause analysis

V-patterns and inverted-cone patterns from buoyant plume; char depth and char patterns; alligator (large-scale blistering — not, per modern NFPA 921, a reliable accelerant indicator); spalling of concrete (heat + moisture); annealing of bedsprings; heat shadows; clean-burn (oxidation of soot from sustained heating); demarcation lines and protected areas.

Modern practice rejects many legacy “indicators” (pour patterns alone, low burn, depth of char alone) as diagnostic in the absence of corroborating laboratory evidence — the Innocence Project and several wrongful-conviction cases (Cameron Todd Willingham, 1992, executed 2004) catalyzed this methodological reform.

Accelerant detection

• Field: hydrocarbon “sniffer” instruments (Arson 2000, Hnu, PID), accelerant-detection canines.
• Lab: GC-MS per ASTM E1618 — identifies gasoline, diesel, kerosene, mineral spirits by characteristic chromatographic patterns (target compounds and groups).

Electrical-fire investigation

• Arc-mapping — survey arc beads on building wiring to localize first arc point and approximate fire travel.
• Distinguishing cause arcing (arc started the fire) from victim arcing (arc resulted from fire) via metallurgy (cause arcs show globular smooth beads; victim arcs are slag-like).
• Service entrance, panel, and receptacle inspection; aged aluminum branch-circuit terminations; FPE Stab-Lok and Zinsco panel hazards.

Explosions

• Vapor cloud (Buncefield 2005, Texas City 2005); dust (Imperial Sugar Refinery, Port Wentworth GA, 7 February 2008, 14 fatalities; CSB report 2009); BLEVE; condensed-phase detonation (West Fertilizer, Beirut port).
• Static electricity and ESD ignition: low-conductivity fluid loading, “static-electricity-induced” tank fires.
• Lithium-ion thermal runaway investigation is now a recognized sub-discipline — failure modes include internal short (Note 7), mechanical abuse (EV crash), external heat (cell-to-cell propagation), overcharge, and dendrite growth. UL 9540A test data characterizes BESS thermal-runaway propagation.

CSB

The US Chemical Safety and Hazard Investigation Board is an independent federal agency modeled on NTSB but for industrial chemical incidents. CSB has no regulatory power; it issues investigation reports and safety recommendations. Major reports include BP Texas City (2007), Deepwater Horizon (2014–16), West Fertilizer (2016), Imperial Sugar (2009), and DuPont La Porte (2019).

9. Geotechnical and structural forensic engineering

Geotechnical

• Site investigation: SPT (Standard Penetration Test), CPT/CPTu (Cone Penetration Test with pore pressure), seismic CPT, geophysical surveys (MASW, resistivity, GPR), borehole sampling for triaxial and consolidation tests.
• Failure modes: bearing-capacity failure, settlement (immediate, primary consolidation, secondary creep), differential settlement, slope failure (rotational, translational, flow), liquefaction (Niigata 1964, Christchurch 2011), retaining-wall failure, dam piping (Teton Dam 1976), karst sinkhole.
• Forensic re-analysis with finite-element (PLAXIS, FLAC) and limit-equilibrium (SLOPE/W, Slide) using as-built geometry and post-failure soil parameters.

Concrete

• ASR (alkali-silica reaction) — reactive aggregate + alkali pore solution + moisture → expansive gel, map cracking, displacement.
• DEF (delayed ettringite formation) — high curing temperature in mass concrete → late ettringite expansion.
• Carbonation depth (phenolphthalein test) and chloride penetration (silver-nitrate spray, RCT) reduce pH and depassivate rebar → rust expansion → spalling. Surfside garage slab and column tops are textbook cases.
• Freeze-thaw spalling in air-entrained vs poorly air-entrained concrete.
• Petrographic examination per ASTM C856.

Steel

• Fatigue at welded details (AASHTO Categories A–E’); corrosion-fatigue (offshore, marine); brittle fracture below DBTT (Liberty ships, Hoan Bridge 2000).
• Welding defects: lack of fusion, lack of penetration, porosity, slag inclusion, undercut, hydrogen-induced cracking, hot cracking. UT (manual + PAUT), RT, and MT inspection per AWS D1.1 acceptance criteria.

Buildings

• Progressive-collapse analysis with alternate-load-path (ASCE 7-22, GSA 2016, DoD UFC 4-023-03).
• Diaphragm action and chord/collector force distribution.
• Connection failure (e.g., welded moment connections in Northridge 1994, leading to FEMA-350 series).

10. Aviation accident investigation

Agencies

• NTSB (US, independent federal, established 1967).
• AAIB (UK Air Accidents Investigation Branch), BEA (France), BFU (Germany), TSB (Canada), JTSB (Japan), CAA systems globally.
• ICAO Annex 13 governs international investigation: state of occurrence has primary responsibility; state of registry, operator, design, and manufacture participate as accredited representatives.

Recorders

• FDR (Flight Data Recorder) — 88+ parameters per FAA for transport-category since 2002; 25 h at 1–32 Hz.
• CVR (Cockpit Voice Recorder) — 2 h (extended to 25 h by EASA 2021 and FAA NPRM 2024).
• QAR (Quick Access Recorder) and operator FOQA (Flight Operational Quality Assurance).
• ACMS / DFDR wreckage decoding with manufacturer support.

Wreckage and materials

• Field mapping and 3D scan of wreckage layout (e.g., TWA 800 reconstruction at Calverton 1996–2000, MH17 Donetsk wreckage assembled in Gilze-Rijen 2014–15).
• Materials examination per Section 3: fatigue, corrosion, composite delamination, fire damage, ballistic damage.

Human factors

• CRM (Crew Resource Management) since United 173 (1978) and KLM/Pan Am Tenerife (1977).
• ATC communication, weather, fatigue, automation surprise (Asiana 214, 2013, SFO).
• Maintenance human factors (ANA 60 / China Airlines 611, 2002 — improper repair of tail strike 1980 → fuselage breakup over Strait of Taiwan).

Composite NDT

Pulse-echo UT, phased-array UT, IR thermography (active flash thermography), acoustic emission, shearography. Increasingly important as composite fraction of airframes grows (787 ~50% by weight, A350 ~52%).

11. Product liability and expert witness

Admissibility standards

• Daubert v. Merrell Dow Pharmaceuticals, 509 U.S. 579 (1993) — SCOTUS set the modern test for admissibility of scientific expert evidence in federal court: (1) testability/falsifiability, (2) peer review/publication, (3) known/potential error rate, (4) standards controlling operation, (5) general acceptance in the relevant community. Codified in Federal Rule of Evidence 702 (amended December 2023 to clarify that the proponent bears the burden of demonstrating admissibility by a preponderance).
• Frye v. United States, 293 F. 1013 (D.C. Cir. 1923) — “general acceptance” test; still followed in some state courts (e.g., New York, Pennsylvania, California for novel science).
• Kumho Tire v. Carmichael, 526 U.S. 137 (1999) — extended Daubert to all expert testimony including engineering, not just “science”.

Engineer-as-expert practice

• Retention letter, qualifications, declaration; preliminary, then final, written report per FRCP Rule 26(a)(2)(B).
• Reports must list: opinions and bases, data considered, exhibits, qualifications, prior testimony (4 years), compensation.
• Cross-examination probes: methodology, alternative hypotheses, daubert factors, fee, prior inconsistent testimony.
• Hot-tubbing (“concurrent expert evidence”) in Australia (Federal Court Practice Note CM7), UK (CPR Part 35), and increasingly other common-law jurisdictions — experts give evidence together, panel-style, before the judge.
• ASTM E620 — standard practice for reporting opinions of scientific or technical experts.

Pitfalls

• “Junk science” — methodology not tested or accepted; General Electric v. Joiner (1997) reinforces gatekeeper role.
• Cherry-picking favorable data; failing to consider alternative explanations; conclusory ipse dixit; advocacy outside expertise.

12. Tools and software

Imaging and capture

• Photography: full-frame DSLR/mirrorless with macro 100 mm + ring/macro flash; calibrated scale bars (ABFO No. 2 for forensic).
• Drone: DJI Mavic 3 Pro, Phantom 4 RTK, Skydio X10 for confined sites and infrastructure.
• 360° capture: Matterport Pro3, Insta360 Titan.

3D scanning

• Terrestrial LiDAR: FARO Focus S/M Premium, Leica RTC360 / BLK360, Trimble X9, Z+F IMAGER 5016.
• Handheld: Artec Eva / Leo, Creaform HandyScan, FARO Freestyle.
• Photogrammetry: RealityCapture, Agisoft Metashape, Pix4Dmapper.

CT and microCT

• For entire failed components: Zeiss Xradia Versa (sub-µm), Nikon XT H 225, GE phoenix v|tome|x. Internal defect mapping without destructive sectioning.

Microscopy

• Stereo (Leica M205, Zeiss Stemi 305); inverted metallographic (Olympus GX, Zeiss Axio); SEM (Zeiss Sigma, FEI Helios, JEOL JSM-7900F, Hitachi SU5000); TEM (Thermo Talos, JEOL JEM-F200); AFM (Bruker Dimension, Park NX). Background in [[MaterialsScience/characterization-methods]].

Mechanical testing

• Instron 5900/8800 series; MTS 810/370 servo-hydraulic; Zwick/Roell Z series; Charpy V-notch (Tinius Olsen IT406); micro-hardness (Wilson Tukon, Buehler Wilson VH1102); drop-weight (NDT, DWTT per ASTM E208).

Software

• Accident reconstruction: PC-Crash, HVE, Virtual CRASH, REC TEC; SMAC, EDSMAC, CRASH4.
• Chemical release: ALOHA (NOAA + EPA), PHAST (DNV), TRACE, FLACS (Gexcon CFD), CANARY (Quest).
• Impact/collapse FEA: LS-DYNA (Ansys), Abaqus/Explicit, Ansys Mechanical, Radioss.
• Fire CFD: FDS (Fire Dynamics Simulator, NIST) with Smokeview; JASMINE, FLUENT combustion models.
• Geotechnical FEA: PLAXIS 2D/3D, FLAC2D/3D, Slope/W, Slide3.
• eDiscovery and document review: Relativity, Everlaw, Disco.

AI augmentation (2024–26)

• Image-based fractography classification — convolutional and transformer models classify fracture mode (ductile dimples, cleavage, fatigue striations, intergranular) at human-expert agreement; commercial offerings emerging from Bruker and Carl Zeiss.
• LLM-assisted document review with semantic search and clustering in Relativity and Everlaw — exhibit identification, deposition prep, contract-clause extraction.
• Robotic 3D capture: autonomous photogrammetry drones (Skydio, Optelos) and SLAM-based mobile mappers (NavVis VLX, Boston Dynamics Spot + RTC360).
• Predictive maintenance ML to prevent failures upstream — vibration, oil-debris, and thermal monitoring with anomaly detection.

13. Standards and best practices

• ASCE Manual of Practice No. 119 — Guidelines for Failure Investigation (and the related ASCE Forensic Practices Committee publications).
• ASCE 7 — Minimum Design Loads (forensic re-analysis baseline).
• NFPA 921 / 1033 — fire and explosion investigation and investigator qualification.
• ASTM E2018 — Property Condition Assessment; E620 — expert reports; E1188, E1618, E1729 — fire-debris; E2332 — investigation of forensic engineering.
• NAFE Journal — peer-reviewed case studies and methodology.
• ICAO Annex 13 — aircraft accident and incident investigation.
• SAE J224, J670, J1733 — collision and vehicle dynamics nomenclature.
• IEC 60079 — explosive atmospheres equipment.
• API RP 1175 — pipeline leak detection program management.
• API 754 — process safety performance indicators.
• OSHA 29 CFR 1910.119 — Process Safety Management of Highly Hazardous Chemicals.

14. Recent trends (2024–26)

• Drone, LiDAR, and photogrammetry are now routine on scenes — UAS Part 107 widely held by forensic engineers in the US; EASA Open / Specific categories in Europe.
• 3D-printed evidence (failed components, scene models) and immersive (VR/AR) jury exhibits — admissibility increasingly accepted with proper foundation.
• AI-assisted case analysis tools (Eve.ai, Harvey, CoCounsel, Caretaker, Relativity aiR) accelerating deposition prep, exhibit organization, and timeline construction; the engineering opinion itself remains human-attributed.
• Telematics and connected-vehicle data: Tesla event log, Autopilot snapshots, Cruise/Waymo robotaxi sensor archives, fleet CMV ELD/GPS — these are now subpoena-routine and require digital-forensics expertise alongside traditional reconstruction.
• EV battery forensics is an emerging discipline: pack disassembly under inert atmosphere, cell-level CT, state-of-charge analytics, propagation analysis per UL 9540A, distinguishing thermal-runaway initiation from external-fire ingress.
• Cybersecurity-related forensic — smart-home actuator misbehavior, connected-vehicle CAN-bus attacks, OT/SCADA incidents (Colonial Pipeline 2021); ISO/SAE 21434 (automotive cybersecurity) and IEC 62443 (industrial) provide audit baselines.
• Climate-driven failures — chloride and freeze-thaw acceleration of concrete deterioration, wildfire-driven WUI structural fires, sea-level corrosion in coastal infrastructure (post-Surfside).

15. Cross-references

• [[Engineering/fatigue-analysis]] — S–N, ε–N, da/dN, Miner’s rule, mean stress effects.
• [[Engineering/fracture-mechanics]] — LEFM, EPFM, J-integral, Kc/Kic, R-curve.
• [[Engineering/mechanics-of-materials]] — stress, strain, yield criteria.
• [[Engineering/structural-dynamics]] — modal analysis, response spectrum, time history.
• [[Engineering/Tier3/welding-processes]] — SMAW, GMAW, GTAW, SAW, FCAW, defects.
• [[Engineering/Tier3/ndt-methods]] — UT, RT, PT, MT, ET, AE, VT.
• [[Engineering/Tier3/steel-grades]] — A36, A572, A992, API 5L, ASTM A516.
• [[MaterialsScience/characterization-methods]] — SEM, EDS, XRD, TEM, AFM.
• [[Engineering/reliability-engineering]] — FMEA, FTA, MTBF, Weibull.
• [[Engineering/Tier3/engineering-codes]] — FAA, NTSB, NFPA, CSB jurisdictions.
• [[Law/contracts-and-ip]] — Daubert, product liability, expert reports.
• [[Engineering/Tier3/battery-chemistries]] — Li-ion, LFP, NMC, thermal-runaway propagation.

16. Citations

• ASM Handbook Volume 11, Failure Analysis and Prevention, 2nd edition, ASM International, 2021.
• Brooks, C. R. and Choudhury, A., Failure Analysis of Engineering Materials, McGraw-Hill, 2002.
• Anderson, T. L., Fracture Mechanics: Fundamentals and Applications, 4th edition, CRC Press, 2017.
• Stamatis, D. H., Failure Mode and Effect Analysis: FMEA from Theory to Execution, 2nd edition, ASQ Quality Press, 2003.
• NFPA 921, Guide for Fire and Explosion Investigations, 2024 edition.
• NFPA 1033, Standard for Professional Qualifications for Fire Investigator, 2024 edition.
• NTSB Investigations Manual (Aviation, Highway, Marine, Pipeline, Hazardous Materials, Railroad).
• ASCE, Manual of Practice No. 119: Guidelines for Failure Investigation.
• Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993); Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999); General Electric Co. v. Joiner, 522 U.S. 136 (1997).
• CSB Final Reports: BP Texas City Refinery Explosion and Fire (Report 2005-04-I-TX, March 2007); Deepwater Horizon (Volumes 1–4, 2014–2016); West Fertilizer (Report 2013-02-I-TX, January 2016); Imperial Sugar (Report 2008-05-I-GA, September 2009).
• CAIB, Columbia Accident Investigation Board Report, Volumes I–VI, August 2003.
• Rogers Commission, Report of the Presidential Commission on the Space Shuttle Challenger Accident, June 1986.
• Boeing 737 MAX investigations: KNKT Final Report on Lion Air JT610 (October 2019); Ethiopian AIB Final Report on ET302 (December 2022); Joint Authorities Technical Review (JATR) Report (October 2019); US House Transportation Committee, The Boeing 737 MAX Aircraft: Costs, Consequences, and Lessons (September 2020); DOT OIG audit reports 2019–2024.
• NIST, Champlain Towers South Investigation (Surfside, FL), interim and final reports 2024+ (ongoing).
• Vaughan, D., The Challenger Launch Decision: Risky Technology, Culture, and Deviance at NASA, University of Chicago Press, 1996.
• Reason, J., Human Error, Cambridge University Press, 1990; Dekker, S., The Field Guide to Understanding ‘Human Error’, 3rd edition, Ashgate, 2014.
• Leveson, N. G. and Turner, C. S., “An Investigation of the Therac-25 Accidents,” IEEE Computer, July 1993.
• Petroski, H., To Engineer Is Human: The Role of Failure in Successful Design, St. Martin’s, 1985; Design Paradigms, Cambridge, 1994.