Steel Design Methodology — AISC, AS, Eurocode, CSA Approach

Q: How do load combinations differ between codes?

AISC 360 references ASCE 7-22 combinations (LRFD: 1.4D, 1.2D+1.6L, 1.2D+1.0W+0.5L, etc.). AS 4100 uses AS 1170.0 (1.2G+1.5Q, 1.2G+Wu+ψcQ, etc.). EN 1993 references EN 1990 (6.10a/b combinations with γG=1.35, γQ=1.5). CSA S16 uses NBCC 2020 load combinations. Each code has different partial safety factors reflecting regional reliability targets.

Q: What are phi factors and why do they vary?

Phi factors (φ) account for material and fabrication variability. Per AISC 360 B3.1: φ=0.90 for tension yielding, φ=0.75 for bolt shear, φ=0.75 for weld metal, φ=0.90 for beam bending (compact), φ=0.85 for column compression (AISC), φ=0.90 for weld base metal. EN 1993 uses γM0=1.0 (cross-section), γM1=1.0 (buckling), γM2=1.25 (connections). AS 4100 uses φ=0.90 (members) and φ=0.80 (connections).

Q: What is the relationship between phi (LRFD) and omega (ASD)?

The LRFD resistance factor φ and ASD safety factor Ω are related through the load statistics assumed in the calibration process. Per AISC 360 B3.2, for a given limit state, φ and Ω are linked as Ω = 1.5/φ (for most common cases with dead-to-live load ratio of approximately 1:3). For example: tension yielding φ=0.90, Ω=1.67 (1.5/0.90=1.67); tension rupture φ=0.75, Ω=2.00 (1.5/0.75=2.00); beam bending φ=0.90, Ω=1.67; bolt shear φ=0.75, Ω=2.00. This relationship ensures approximately equivalent reliability between LRFD and ASD designs across typical load ratios.

Q: Which design methodology is more conservative — LRFD or ASD?

Neither method is consistently more conservative — it depends on the load ratio. LRFD is more conservative for high live-to-dead load ratios (L/D > 3) because the 1.6 live load factor is larger than the ASD equivalent. ASD is more conservative for low live-to-dead load ratios (L/D < 1) because the 1.2 dead load factor in LRFD is relatively low compared to the unified ASD factor of safety. The two methods converge at L/D ≈ 3.2 for typical steel members. EN 1993 and CSA S16 use LRFD-type approaches exclusively, while AISC 360 and AS 4100 offer both options.

Structural steel design follows limit states methodology — ensuring that a structure and its components have adequate strength, stiffness, and stability under all foreseeable loads. This guide compares the design philosophies across four major international codes.

Quick links: Beam capacity → | Column buckling → | Section properties →

Core calculations run via WebAssembly in your browser with step-by-step derivations across AISC 360, AS 4100, EN 1993, and CSA S16 design codes. Results are preliminary and must be verified by a licensed engineer.

Limit States Design Philosophy

Limit states design ensures that a structure remains functional and safe under all expected conditions by checking two categories of limit states:

Strength limit states — Related to safety and load-bearing capacity. These include: flexural failure, shear failure, buckling (local, lateral-torsional, and flexural), connection failure, and overturning. For each strength limit state, the design condition is: Factored Resistance ≥ Required Strength.

Serviceability limit states — Related to functional performance under service loads. These include: excessive deflection, vibration perceptibility, fatigue crack propagation, and corrosion. Serviceability checks use nominal (unfactored) loads.

The fundamental limit state equation is: φRn ≥ ΣγiQi for LRFD, or Rn/Ω ≥ ΣQi for ASD, where φRn is the design strength, ΣγiQi is the required strength from factored loads, Rn is the nominal strength, and Ω is the safety factor.

LRFD vs ASD — Detailed Comparison

Load and Resistance Factor Design (LRFD)

LRFD applies multiple load factors (γ) that vary by load type, reflecting the different variabilities and probabilities of occurrence of different loads. Typical LRFD load combinations (ASCE 7-22):

1.4D
1.2D + 1.6L + 0.5(Lr or S or R)
1.2D + 1.6(Lr or S or R) + (L or 0.5W)
1.2D + 1.0W + L + 0.5(Lr or S or R)
1.2D + 1.0E + L + 0.2S
0.9D + 1.0W
0.9D + 1.0E

The higher factor on live loads (1.6 vs 1.2 for dead load) reflects the greater uncertainty in live loads. Load factors and resistance factors are calibrated to deliver a target reliability index β ≈ 3.0 for typical members (corresponding to approximately 1 in 1,000 probability of exceedance over a 50-year lifetime).

Allowable Stress Design (ASD)

ASD applies a single safety factor to the nominal strength: Ω = 1.67 for bending (equivalent to φ = 0.90), Ω = 2.0 for welds (φ = 0.75), Ω = 2.0 for bolts in shear (φ = 0.75). Loads are unfactored (service-level). ASD combinations (ASCE 7-22):

D
D + L
D + (Lr or S or R)
D + 0.75L + 0.75(Lr or S or R)
D + 0.6W
D + 0.75L + 0.75(0.6W) + 0.75(Lr or S or R)
0.6D + 0.6W
0.6D + 0.7E

When to Use Each Method

LRFD generally produces more economical designs when live load dominates (because the higher uncertainty of live loads is explicitly accounted for). ASD may be preferred when: (1) the project team is more familiar with ASD, (2) the design is governed by serviceability (deflection or vibration), or (3) the client requires ASD for historical consistency. Per AISC 360 B3.2, either method is permitted.

Cross-Code Comparison of Design Parameters

AISC 360-22 (United States)

Method: LRFD or ASD Key phi factors: φb = 0.90 (flexure, compact), φc = 0.90 (compression), φv = 0.90 (shear), φt = 0.90 (tension yielding), φ = 0.75 (tension rupture, bolts, welds) Load combinations: ASCE 7-22 Material specifications: ASTM A36, A572, A992, A588, A514 Design approach: Allowable stress or load and resistance factor

AS 4100:2020 (Australia)

Method: Limit states (capacity factor approach) Key capacity factors: φ = 0.90 (members — flexure, compression, tension, shear), φ = 0.80 (connections — bolts, welds), φ = 0.90 (web bearing) Load combinations: AS 1170.0 Material specifications: AS/NZS 3678, 3679.1, 3679.2 Design approach: Single capacity factor φ applied to nominal capacity. AS 4100 uses nominal loads (not factored loads) combined with capacity factors. Secondary effects are included through the moment amplification factor δb = Cm/(1 - N*/Nomb).

EN 1993-1-1 (European Union)

Method: Limit states (partial factor approach) Key partial factors: γM0 = 1.00 (cross-section resistance), γM1 = 1.00 (member buckling), γM2 = 1.25 (connections — bolts, welds, plates in bearing) Load combinations: EN 1990 (γG = 1.35, γQ = 1.50) Material specifications: EN 10025 (S235, S275, S355, S460, etc.) Design approach: Multiple partial factors applied to material properties and actions. The design resistance is based on characteristic values divided by partial factors.

CSA S16:2019 (Canada)

Method: Limit states (resistance factor approach) Key resistance factors: φ = 0.90 (steel members), φ = 0.67 (welds), φ = 0.80 (bolts in shear), φ = 0.75 (bolts in tension) Load combinations: NBCC 2020 Material specifications: CSA G40.20/G40.21 (350W, 350WT, 480W, etc.) Design approach: Similar to AISC LRFD with differences in specific resistance equations and the treatment of slender elements.

Load Path and Structural Modeling

Understanding the load path is fundamental to structural design methodology. Loads travel from the point of application through the structure to the foundations:

Vertical load path: Roof/Wall/Floor → Deck/Slab → Joists/Beams → Girders → Columns → Foundations → Soil

Lateral load path: Wind/Seismic → Cladding/Walls → Diaphragms → Collectors → Braced frames/Moment frames/Shear walls → Foundations

Each element along the load path must be designed for the accumulated loads from the elements above it. The methodology requires tracing the critical load path and verifying each component: deck shear transfer, beam-to-column connection capacity, column axial and flexural interaction, and foundation bearing and uplift resistance.

Second-Order Analysis

Modern steel design methodology (AISC 360 Chapter C, EN 1993-1-1 Section 5, AS 4100 Clause 4) requires consideration of second-order effects:

P-Δ effects — Global second-order effects from the lateral displacement of the structure under gravity loads. The B2 amplification factor accounts for this: B2 = 1/(1 - Pr × ΔH/HL), where Pr is the total gravity load on the story, ΔH is the first-order story drift, H is the story shear, and L is the story height.

P-δ effects — Local second-order effects from the deflection of the member between its ends. The B1 amplification factor accounts for this: B1 = Cm/(1 - Pr/Pe1) ≥ 1.0, where Cm accounts for the moment gradient and Pe1 is the Euler buckling load about the axis of bending.

When B2 > 1.5 or when the drift ratio exceeds 0.025, a rigorous second-order analysis (direct analysis method per AISC C2) is required instead of the approximate B1/B2 method.

Frequently Asked Questions

What is the difference between LRFD and ASD? LRFD (Load and Resistance Factor Design) applies load factors (>1.0) to nominal loads and resistance factors (<1.0) to nominal strength: φRn ≥ ΣγiQi. ASD (Allowable Stress Design) applies a single safety factor to strength: Rn/Ω ≥ ΣQi. LRFD provides more uniform reliability across different load types because the load factors vary by load type (1.2D + 1.6L vs 1.2D + 1.0W). ASD uses a single factor of safety Ω applied to the nominal strength.

How do load combinations differ between codes? AISC 360 references ASCE 7-22 combinations (LRFD: 1.4D, 1.2D+1.6L, 1.2D+1.0W+0.5L, etc.). AS 4100 uses AS 1170.0 (1.2G+1.5Q, 1.2G+Wu+ψcQ, etc.). EN 1993 references EN 1990 (6.10a/b combinations with γG=1.35, γQ=1.5). CSA S16 uses NBCC 2020 load combinations. Each code has different partial safety factors reflecting regional reliability targets.

What are phi factors and why do they vary? Phi factors (φ) account for material and fabrication variability. Per AISC 360 B3.1: φ=0.90 for tension yielding, φ=0.75 for bolt shear, φ=0.75 for weld metal, φ=0.90 for beam bending (compact), φ=0.85 for column compression (AISC), φ=0.90 for weld base metal. EN 1993 uses γM0=1.0 (cross-section), γM1=1.0 (buckling), γM2=1.25 (connections). AS 4100 uses φ=0.90 (members) and φ=0.80 (connections).

What is the relationship between phi (LRFD) and omega (ASD)? The relationship between LRFD resistance factor φ and ASD safety factor Ω is Ω = 1.5/φ for load combinations governed by dead plus live loads. This relationship ensures approximately equivalent designs between the two methods when the live-to-dead load ratio is approximately 3. For bending: φ = 0.90, Ω = 1.67. For shear: φ = 0.90, Ω = 1.67. For compression: φ = 0.90, Ω = 1.67. For bolt shear: φ = 0.75, Ω = 2.0. For weld metal: φ = 0.75, Ω = 2.0. The conversion provides consistent reliability levels across both methods.

Which design methodology is more conservative — LRFD or ASD? Neither method is universally more conservative — the relative conservatism depends on the load ratio. For typical building structures where live load exceeds dead load, LRFD tends to be more economical (requires less steel) because the probabilistic calibration better reflects the actual load variability. For structures with predominantly dead load (e.g., heavy industrial facilities), ASD may produce lighter members. LRFD provides more uniform reliability across different load scenarios, which is why it has been the preferred method in AISC since 2005. However, both methods are permitted and produce safe designs when properly applied.

Disclaimer (educational use only)

This page is provided for general technical information and educational use only. It does not constitute professional engineering advice. All results must be independently verified by a licensed Professional Engineer.