Mastering Welding Processes
in Manufacturing
A comprehensive engineer's guide to the five most critical welding processes in modern manufacturing — MIG/MAG, TIG, Resistance Spot Welding, Laser Beam Welding, and Friction Stir Welding — covering process physics, metallurgical effects, equipment parameters, weld defects, quality standards, NDT inspection methods, and the process selection matrix every manufacturing engineer needs.
What is Welding — and Why Does Process Selection Matter?
Welding is the process of permanently joining two or more materials — almost always metals — by applying heat, pressure, or both to create a metallurgical bond at the atomic level. Unlike mechanical fastening (bolts, rivets) or adhesive bonding, welding produces a joint that is integral to the parent material — capable of transmitting the full structural load, carrying fluid pressure, conducting electricity, and withstanding the same fatigue, corrosion, and temperature environment as the base metal itself. It is, by any measure, the most important manufacturing joining process in existence: more than 50% of all manufactured goods involve welding, and the global welding equipment and consumables market exceeds USD 30 billion annually.
Welding is a fabrication process that joins materials by causing coalescence — typically by melting the base materials and adding a filler material to form a pool of molten material (the weld pool) that cools to become a strong joint. Not all welding involves melting: solid-state processes like Friction Stir Welding join materials below their melting point through heat and severe plastic deformation, producing joints with superior metallurgical properties in materials that are difficult or impossible to fusion-weld.
Process selection is the most consequential decision in weld engineering. The wrong choice produces welds that are structurally sound on paper but unsuitable in practice — too slow for production volumes, too distorting for precision assemblies, too expensive for the application, or metallurgically inappropriate for the base material. MIG welding, TIG welding, resistance spot welding, laser beam welding, and friction stir welding are not interchangeable alternatives — each occupies a specific domain defined by material type, thickness, joint geometry, production volume, quality requirement, and cost constraint. Understanding those domains is what separates a manufacturing engineer who specifies the right weld from one who creates expensive problems downstream.
The best weld is the one that meets the structural requirement, is producible at the required rate, is inspectable to the required standard, and costs least over the life of the product. Achieving all four simultaneously requires understanding not just the welding process but the material, the joint, the production environment, and the service condition.
— Lincoln Electric Welding Engineering HandbookMIG / MAG Welding — The Workhorse of Manufacturing
MIG (Metal Inert Gas) welding — more precisely called GMAW (Gas Metal Arc Welding) — uses a continuously fed consumable wire electrode that simultaneously acts as both the arc conductor and the filler metal, shielded by a flow of inert gas (argon for MIG) or an active gas mixture (CO₂ or Ar+CO₂ for MAG — Metal Active Gas). The arc melts both the wire and the base metal, creating a weld pool that is protected from atmospheric contamination by the shielding gas. MIG/MAG accounts for approximately 70% of all robotic welding applications globally — a testament to its outstanding balance of speed, versatility, and automation compatibility.
The process offers three distinct metal transfer modes that operate at different current ranges and produce different weld characteristics. Short-circuit transfer (low current, thin materials) produces a small, fast-freezing weld pool — ideal for sheet metal and out-of-position welding. Globular transfer (medium current) produces larger, irregular droplets — high spatter, generally avoided in production. Spray transfer (high current, above spray transition threshold with argon-rich gas) produces a fine, continuous stream of tiny droplets — smooth welds, deep penetration, high deposition rate, but requires flat or horizontal position. Pulsed MIG (electronically switches between high and low current) achieves spray-quality welds at lower average heat input — the preferred mode for thin aluminium and for reduced distortion applications.
For automotive chassis welding, agricultural equipment, and structural steel fabrication — anywhere that high production volumes, medium-to-heavy material thickness (1.5–25mm), and repeatable quality are required — robotic MIG/MAG welding is the definitive choice. A robotic MIG cell welding automotive body components achieves travel speeds of 500–1,200 mm/min with arc-on times exceeding 85%, replacing a welder who might achieve 30–40% arc-on time with significantly more variation. The main limitation is weld appearance — MIG welds typically have spatter deposits requiring post-weld grinding for cosmetic applications, and the process is less suited than TIG to the highest aesthetic standards.
TIG Welding — The Gold Standard of Precision
TIG (Tungsten Inert Gas) welding — formally GTAW (Gas Tungsten Arc Welding) — uses a non-consumable tungsten electrode to create the arc, with the filler metal (when required) added separately by the welder's free hand or by an automated wire feeder. The weld zone is shielded by pure argon or helium. Because the tungsten electrode does not melt, the welder has independent control of heat input (via foot pedal current control) and filler metal addition — a level of control that no other arc welding process can match. The result is the highest-quality, cleanest, most precise weld achievable with arc welding.
TIG welding's non-consumable electrode and separate filler addition allow the welder to control the weld pool with surgeon-like precision — managing heat input to avoid burn-through on thin material, controlling penetration depth, preventing oxidation with comprehensive inert gas coverage, and achieving a smooth, spatter-free, oxide-free weld that needs no post-weld grinding for aesthetic applications. This precision comes at a cost: TIG welding travels at 200–500 mm/min — roughly one-third the speed of MIG — and requires the highest skill level of any fusion welding process. It is categorically the right choice for aerospace components, pharmaceutical and food-grade stainless steel vessels, titanium, inconel, and any application where weld quality, corrosion resistance, and appearance are non-negotiable.
Orbital TIG welding — where the torch rotates automatically around a fixed tube — is the standard process for pharmaceutical, semiconductor, biotechnology, and food industry tube systems, where full-penetration welds on small-diameter tubes must achieve surface roughness Ra <0.8μm on the bore face, with no internal crevices or undercut that could harbour bacteria or contamination. Every semiconductor fab clean room and every pharmaceutical process line uses orbital TIG-welded 316L stainless or high-purity PFA connections as the standard of tube joining quality. Hot Wire TIG passes current through the filler wire to preheat it, increasing deposition rate by up to 100% without increasing heat input to the joint — bridging the speed gap between TIG and MIG for cladding and thick-section applications.
Resistance Spot Welding — The Automotive Body Standard
Resistance Spot Welding (RSW) joins overlapping sheet metal panels by passing a high electrical current through copper electrodes pressed against the sheets. The electrical resistance at the interface between the sheets generates concentrated heat — heating the metal to its melting point at the contact interface only, forming a lens-shaped molten nugget (the spot weld). When the current ceases, the electrode pressure is maintained while the nugget solidifies under compression. The entire cycle — squeeze, weld, hold, release — takes less than one second for most automotive sheet gauges. No filler metal, no shielding gas, no consumables beyond the electrodes themselves — RSW is the most cost-effective, production-efficient welding process for automotive body assembly.
Every passenger vehicle body-in-white (BIW) contains between 3,000 and 6,000 spot welds applied by robotic welding cells operating at cycle rates of one spot every 1–3 seconds. The robots not only weld — they also apply precise electrode force (typically 2,000–6,000 N), control welding current and time to within ±1% accuracy, and monitor the weld signature (dynamic resistance, electrode displacement, welding current) to detect defective welds in real time. Modern RSW controllers using adaptive weld monitoring can detect the electrode coating build-up that shifts the weld current distribution, adjust parameters automatically, and flag electrodes requiring dressing — maintaining consistent nugget size without operator intervention.
The increasing use of Advanced High Strength Steels (AHSS) — dual phase, complex phase, martensitic, and press-hardened steels with tensile strengths from 600 to 1,800 MPa — has made RSW significantly more challenging. AHSS requires higher electrode force, modified weld schedules (multi-pulse or stepped current programmes), and more frequent electrode dressing compared to conventional mild steel. The HAZ of AHSS spot welds can undergo significant microstructural changes — martensite formation and tempering — that must be validated against the OEM's minimum nugget diameter specification and peel/chisel test requirements per AWS D8.1 or the relevant OEM welding standard.
Laser Beam Welding — Speed, Precision, and Minimal Distortion
Laser Beam Welding (LBW) focuses a high-power laser beam (typically from a fibre laser, diode laser, or Nd:YAG laser) to a spot diameter of 0.1–1.0mm, achieving power densities of 10⁵–10⁷ W/cm² — sufficient to vapourise metal and create a keyhole: a narrow vapour capillary surrounded by molten metal that penetrates deep into the material at the laser focal point. As the beam traverses the joint, the keyhole moves forward and the molten metal flows around it and solidifies, creating a narrow, deep weld with an aspect ratio (depth-to-width) of up to 10:1 — achieved at travel speeds of 1,000–8,000 mm/min. No other fusion welding process produces this combination of depth, speed, and minimal heat input simultaneously.
The fundamental advantage of laser welding is its minimal heat-affected zone and negligible distortion. The concentrated heat input melts and solidifies the weld zone so rapidly that the surrounding material sees very little thermal energy — heat-affected zones measured in tenths of a millimetre rather than the centimetres typical of arc welding. For precision assemblies where post-weld straightening or machining would destroy dimensional accuracy — gear assemblies, medical instruments, electronics housings, precision optical devices — laser welding is often the only viable fusion joining process. Fibre lasers (1,070nm wavelength) with output powers of 1–20kW are now the standard production laser for industrial metal welding, offering wall-plug efficiencies of 25–35% (versus 2–5% for CO₂ lasers) and fibre delivery that makes robotic integration straightforward.
Laser-Hybrid Welding — combining a laser beam with a MIG arc in a single weld head — achieves the best of both processes: laser speed and penetration with MIG's bridging ability (tolerance for fit-up gaps and joint misalignment). Laser-hybrid is the dominant process for automotive door and body panel laser welding (Class A visible surfaces where distortion and surface finish are paramount) and for structural steel welding where deep penetration at high speed is required on joints with real-world fit-up variation. The global automotive industry's transition to electric vehicles is dramatically expanding laser welding use: EV battery cell tab welding, hairpin motor stator welding, and battery pack lid sealing are all laser welding applications growing exponentially.
Friction Stir Welding — Joining Without Melting
Friction Stir Welding (FSW), invented by Wayne Thomas at The Welding Institute (TWI) in Cambridge, UK, in 1991, is the most significant welding innovation of the 20th century. Unlike all fusion welding processes, FSW joins metals without melting them. A rotating cylindrical tool with a probe (pin) and shoulder is plunged into the joint line between two rigidly clamped workpieces. Friction between the rotating shoulder and the workpiece surface generates heat (typically 80–90% of the melting temperature of the metal), softening the material into a plasticised, dough-like state. As the tool traverses along the joint, the probe stirs the plasticised material from the leading edge to the trailing edge, where it consolidates under the forge pressure of the tool shoulder, creating a solid-state bond with no solidification defects.
Because FSW never melts the base material, it avoids the full spectrum of fusion welding defects: no porosity (no gas entrapment from solidification), no hot cracking (no solidification shrinkage), no liquation cracking, no hydrogen embrittlement, and no loss of alloying elements through vapourisation. The mechanical properties of FSW joints in aluminium alloys approach or exceed those achievable by any other joining method — typically 80–95% of the base metal tensile strength, compared to 50–70% for MIG-welded joints in the same alloy. This is why FSW has become the dominant joining process for aerospace aluminium structures, EV battery enclosures, ship hulls, and rail vehicle bodies.
The EV manufacturing revolution has made FSW commercially critical. Electric vehicle battery enclosures — large, thin-walled aluminium structures that must be hermetically sealed, dimensionally precise, and resistant to crash energy — are exactly the application where FSW excels over MIG: lower distortion, no porosity (critical for hermetic sealing), higher joint efficiency in the high-strength aluminium alloys used in battery boxes, and no shielding gas requirement. Robotic FSW systems with multi-axis motion and real-time force feedback now enable automated welding of complex battery pack geometries at automotive production rates. Robotic FSW cells from EWI, Grenzebach, and HITACHI now achieve cycle times compatible with body shop takt rates.
Other Key Welding Processes in Industrial Manufacturing
Flux-Cored Arc Welding (FCAW) is a variant of MIG where the electrode wire contains flux materials in its hollow core rather than relying on external shielding gas (self-shielded FCAW) or supplementing it (dual-shielded FCAW). The flux produces slag that covers and protects the weld pool, enabling higher deposition rates than solid wire MIG and excellent performance in out-of-position and outdoor welding where wind disrupts shielding gas coverage. FCAW is the process of choice for heavy structural steel fabrication — shipbuilding, bridge construction, offshore platforms, and thick-section pressure vessels — where deposition rates of 4–8 kg/hour (compared to 1–3 kg/hour for MIG) justify the post-weld slag removal step.
Submerged Arc Welding (SAW) buries the arc beneath a blanket of granular flux, producing the highest deposition rates of any arc process (8–20 kg/hour) with excellent weld quality and minimal fume — making it the preferred process for flat and horizontal heavy-section welding on pipe, plate, and structural members where single-pass welds through 25–75mm thickness are required. Pipe mills use SAW to weld spiral and longitudinal seams in large-diameter steel pipe. Shipyards use SAW for main deck plating. SAW cannot be used out of position.
Plasma Arc Welding (PAW) is a refinement of TIG that constricts the arc through a copper nozzle to create a plasma jet — achieving higher energy density, deeper penetration, and faster travel speed than TIG for the same current. In keyhole mode, PAW can penetrate and weld 8–12mm stainless steel in a single pass without a backing bar — a significant productivity advantage. PAW is used for titanium aerospace components, stainless steel food and pharma vessels, and precision instrumentation.
Electron Beam Welding (EBW) focuses a beam of accelerated electrons in a vacuum chamber, achieving the highest energy densities and greatest penetration depths of any welding process — single-pass welds through 150–300mm of steel are possible. EBW produces the narrowest HAZ of any fusion process and welds refractory metals, titanium alloys, and dissimilar combinations impossible for conventional processes. The vacuum requirement makes it expensive and limited to batch production — but for aerospace turbine discs, nuclear fuel elements, and surgical implants, nothing else delivers the same combination of depth, precision, and metallurgical cleanliness.
Welding Metallurgy — The Heat-Affected Zone
Every fusion welding process — MIG, TIG, laser, plasma, SAW — deposits a heat cycle on the surrounding base metal that transforms its microstructure. Understanding the Heat-Affected Zone (HAZ) and its consequences is fundamental to specifying welding processes, selecting preheat requirements, and designing weld inspection criteria.
The coarse-grained HAZ (CGHAZ) — the zone immediately adjacent to the weld metal that was heated above the grain coarsening temperature (~1,100°C for steel) — is the most metallurgically critical and structurally weakest region of any fusion weld. In carbon and low-alloy steels, the CGHAZ experiences the highest cooling rates and can form untempered martensite — brittle, hydrogen-sensitive microstructure that is the primary site for hydrogen-induced cold cracking (HICC), the most dangerous and insidious weld defect. Preheat, controlled heat input, post-weld heat treatment (PWHT), and low-hydrogen consumables all exist to manage CGHAZ martensite and hydrogen content. In aluminium alloys, the HAZ is a zone of thermal softening where age-hardening precipitates dissolve — the primary reason why MIG-welded 7075-T6 aircraft aluminium achieves only 50–60% of base metal strength in the joint.
FSW's solid-state nature eliminates the weld metal zone entirely and reduces the HAZ width and peak temperature — which is precisely why FSW-welded 7075-T6 achieves 80–90% base metal strength compared to 50–60% for MIG. Heat input control — expressed as kJ/mm = (Volts × Amps × 60) / (1,000 × travel speed in mm/min) — is the single most important welding parameter for managing HAZ width, cooling rate, and joint properties. It is mandatory on all structural, pressure vessel, and pipeline welding procedures per AWS D1.1, ASME IX, EN ISO 15614, and equivalent codes.
Weld Defects & NDT Inspection Methods
Weld defects fall into two categories: surface defects detectable by visual and dye penetrant inspection, and internal (sub-surface) defects that require volumetric NDT methods. Both categories are classified by acceptance criteria in the relevant welding standard — AWS D1.1 for structural steel, ASME Section IX for pressure vessels, EN ISO 5817 for general metalwork — with different acceptance levels (B, C, D corresponding to decreasing quality levels).
Spherical or wormhole gas pockets in the weld metal caused by gas evolved during solidification failing to escape. Sources: moisture in consumables, surface contamination (oil, rust, oxide), inadequate shielding gas coverage, or excessive travel speed. Detected by X-ray, UT, or cross-section. Unacceptable in fatigue-loaded welds.
The most dangerous weld defect — can occur hours to days after welding. Hydrogen absorbed during welding diffuses to the CGHAZ martensite under residual tensile stress, causing delayed brittle fracture. Prevention: preheat (raises interpass temperature above martensite start), low-hydrogen consumables, PWHT. Detected by MT, PT (surface) or UT (sub-surface).
Weld metal that has not fused to the base metal or previous weld pass — a planar defect that acts as a crack initiator under fatigue loading. Caused by incorrect torch angle, insufficient heat input, or contaminated joint faces. The most serious volumetric defect in structural welds. Detected by UT or RT.
A groove melted into the base metal adjacent to the weld toe that is not filled with weld metal. Creates a stress concentration at the weld toe — the critical fatigue crack initiation site in structural welds. Limited to 0.5–1mm depth in most codes. Detected by visual inspection and profilometry.
Failure of the weld metal to penetrate the full depth of the joint — leaving an unfused root. Critical in full-penetration butt welds (pipe, pressure vessels, structural). Causes: insufficient heat input, root gap too small, incorrect electrode angle. Must be detected by UT or RT before hydrostatic testing.
Thermal expansion and contraction during welding creates residual stress and angular, longitudinal, or transverse distortion. Not strictly a defect but a dimensional non-conformance. Controlled by: balanced weld sequencing, pre-setting, pre-bending, clamp fixturing, intermittent welds, low heat input processes. FSW and laser welding produce least distortion of all processes.
| NDT Method | Detects | Surface or Volumetric | Standard | Primary Welding Application |
|---|---|---|---|---|
| Visual Inspection (VT) | Cracks, undercut, porosity (surface), profile | Surface only | ISO 17637 / AWS D1.1 | All welds — mandatory first inspection on every joint |
| Dye Penetrant (PT) | Surface-breaking cracks, porosity, LOF | Surface only | ISO 3452 / ASME V | Non-magnetic materials (Al, Ti, stainless, nickel alloys) |
| Magnetic Particle (MT) | Surface & near-surface cracks (0–5mm depth) | Surface + near-surface | ISO 17638 / ASME V | Ferritic steel welds — very sensitive to surface cracks and HICC |
| Radiographic (RT) | Porosity, inclusions, incomplete penetration (volumetric) | Volumetric | ISO 17636 / ASME V | Pressure vessels, pipelines, structural welds — provides a permanent image record |
| Ultrasonic (UT) | LOF, cracks, incomplete penetration, laminations | Volumetric (full thickness) | ISO 11666 / AWS D1.1 | Thick section structural and pressure vessel welds — faster than RT for thick plate |
| Phased Array UT (PAUT) | All internal defects with precise sizing and location | Volumetric + imaging | ISO 13588 / ASME V | Complex geometries, pipe welds, nozzles — generates cross-sectional images |
| Peel / Chisel Test | Spot weld nugget diameter | Destructive | AWS D8.1 / OEM standards | Resistance spot welding — periodic destructive verification of nugget size and button diameter |
Welding Process Selection Matrix
The single most practical tool for any manufacturing engineer specifying a weld process is a structured selection matrix that maps the key application requirements against process capabilities. The matrix below covers the five primary processes in depth, plus supplementary processes for completeness.
| Process | Best Materials | Thickness Range | Deposition Rate | Joint Quality | Distortion | Automation | Cost | Primary Use Case |
|---|---|---|---|---|---|---|---|---|
| Primary Production Processes | ||||||||
| MIG/MAG (GMAW) | Carbon steel, Al, SS | 1.5–25mm | High 1–6 kg/hr | Good (spatter) | Medium | Excellent | Low–Medium | Automotive structures, general fabrication, robotic production |
| TIG (GTAW) | SS, Al, Ti, Ni alloys | 0.3–8mm | Low 0.5–1.5 kg/hr | Excellent | Low–Medium | Good (orbital) | Medium–High | Aerospace, pharma tube, precision thin-sheet, cosmetic welds |
| Spot (RSW) | Steel sheet, AHSS, Al | 0.5–4mm/sheet | Very high (no filler) | Consistent (robotic) | Very Low | Fully robotic | Very Low | Automotive BIW, white goods, sheet metal assemblies |
| Laser (LBW) | Steel, Al, Cu, Ti, SS | 0.1–25mm | Very high speed | Excellent | Minimal | Excellent (robotic) | High (equipment) | EV batteries, precision assemblies, Class A surfaces, electronics |
| FSW | Al, Cu, Mg, Ti (solid-state) | 0.5–50mm Al | Medium (100–1000 mm/min) | Excellent (no solidif. defects) | Very Low | Robotic FSW cells | High (fixture & machine) | EV battery boxes, aerospace Al panels, rail vehicle bodies |
| Heavy Fabrication & Specialist Processes | ||||||||
| FCAW | Carbon & low-alloy steel | 6–75mm | Very high 4–8 kg/hr | Good (slag removal) | Medium–High | Semi-auto | Low–Medium | Heavy structural steel, shipbuilding, construction, out-of-position |
| SAW | Carbon & alloy steel, SS | 12–100mm+ | Highest 8–20 kg/hr | Excellent | Medium–High | Flat/horizontal only | Low | Pipe seam welding, pressure vessels, heavy plate, bridges |
| EBW | Ti, Ni, refractory metals | 1–300mm | Low–Medium | Exceptional | Minimal | Vacuum chamber | Very High | Turbine discs, nuclear fuel, aerospace critical structures |
- Always specify preheat temperature and interpass temperature on the welding procedure specification (WPS) for carbon and alloy steels >12mm or >0.45% CE
- Qualify welders and welding procedures to the relevant standard (AWS D1.1, ASME IX, EN ISO 15614) before production welding begins
- Control heat input rigorously — record volts, amps, and travel speed on every test plate and monitor in production for AHSS and creep-resistant steels
- Back-purge all TIG tube and pipe root passes with argon — without it, the root side of stainless welds oxidises (sugar) and sacrifices corrosion resistance
- Dress resistance spot welding electrodes on a fixed schedule — electrode degradation is the primary cause of undersized nuggets and weld quality drift
- Use filler metal with matching or over-matching tensile strength and the lowest hydrogen diffusible content available for the application
- Apply post-weld NDT within the required window — HICC requires 24–48 hours before final inspection is conducted on susceptible steels
- MIG welding 7xxx-series aluminium with 4043 filler when 5356 is required for ductility — 4043 solidification cracks in high-magnesium alloys
- Specifying TIG for carbon steel structural joints above 4mm where MIG/MAG would be 3× faster with equivalent quality
- Missing preheat on carbon steel with CE >0.45% and thickness >12mm — leading to HICC within 48 hours of welding
- Using TIG wire without back purge on austenitic stainless tube — oxidised root passes fail corrosion testing
- Welding AHSS (DP1000, PHS) with incorrect parameters — martensite in HAZ softens or becomes brittle depending on cooling rate
- Specifying FSW for a part that cannot be rigidly fixtured — FSW forge forces (5–100kN) distort insufficiently clamped assemblies
- Accepting PPAP weld process without Gauge R&R on the weld monitoring system — non-calibrated weld controllers give false acceptance signals
Summary
Welding is not a single technology — it is a family of profoundly different processes that share only the outcome: a permanent, metallurgically bonded joint. MIG/MAG is fast, versatile, and the backbone of automated production. TIG is slow, precise, and unmatched for quality on difficult materials. Spot welding is the high-volume, no-consumable standard for automotive sheet metal. Laser welding brings speed and minimal distortion to precision assemblies and EV components. Friction Stir Welding eliminates fusion defects entirely in aluminium, enabling joint efficiencies that arc welding cannot approach. Each process has its domain — defined by material, thickness, quality requirement, production volume, and cost. Selecting the right one requires understanding all of them.
The Engineer's Welding Principle
The weld is not a separate component — it is the material itself, altered. When you specify a weld process, you are specifying a thermal cycle that will transform the microstructure of every metal it touches. The Heat-Affected Zone is the engineer's signature: narrow and disciplined in laser and FSW welds, wide and complex in SAW and multi-pass MIG. Understanding what that thermal cycle does to the specific alloy, in the specific thickness, under the specific service load, is the difference between a weld that lasts the life of the product and one that initiates a fatigue crack at 50,000 cycles. Choose the process that matches the metallurgy, not just the geometry.
The future of welding in manufacturing belongs to processes that are fast, clean, and compatible with automation and sustainability goals. Laser welding and FSW are growing exponentially in EV manufacturing — not because they are new, but because the materials and designs of electric vehicles (aluminium battery enclosures, copper hairpin windings, mixed-material body structures) are exactly the applications where these processes excel over conventional arc welding. The engineer who masters all five processes — and knows which one to reach for in which situation — is the one who will specify the right weld the first time, every time.
If you can MIG it reliably, MIG it. If the quality or distortion requirement excludes MIG, try TIG. If the joint is sheet metal in high volumes, spot weld it. If distortion must be negligible and the assembly is precision, use laser. If the material is aluminium and joint efficiency is paramount, use FSW. If none of these fits, go back to the design — because the best welding process is the one that doesn't require a compromise.
Welding Processes · MIG/MAG · TIG · Resistance Spot Welding · Laser Beam Welding · Friction Stir Welding · Weld Metallurgy · NDT · AWS · ASME · RMG Tech

