Boeing Bump Index vs IRI vs straightedge: what each runway roughness measure can and cannot see

The Boeing Bump Index (BBI), the International Roughness Index (IRI) and the straightedge each describe the longitudinal profile of a runway, but they ask different questions: the BBI scores single bumps against an aircraft-based criterion, IRI averages a road-vehicle response along a wheelpath, and a straightedge checks a finished surface over a few metres. FAA Advisory Circular 150/5380-9 (30 September 2009, Active) sets out the BBI (FAA, 2009), and IRI is defined for highways (ASTM, 2021a). This guide is engineering background, not operational guidance: decisions to operate, restrict or close a runway belong to the aerodrome operator under its State's rules.

Key facts

• FAA Advisory Circular 150/5380-9 (30 September 2009) computes the Boeing Bump Index as the largest ratio of bump height to the acceptable-zone limit over straightedges of 0.5 m to 120 m.
• ASTM E1926-08(2021) states its scope as the International Roughness Index (IRI) of highway pavements, so IRI is a road quarter-car index and not an aircraft-response index.
• EASA CS-ADR-DSN Issue 7 GM1 ADR-DSN.B.090(a) states no deviation greater than 3 mm under a 3 m straightedge for a new runway surface.
• At a synthetic ground speed of 70.0 m/s, a 3 m length scale is nominally 23.33 Hz and a 120 m length scale is nominally 0.58 Hz, because frequency equals speed divided by wavelength.
• None of the three measures sees texture or friction, and FAA Advisory Circular 150/5380-9 paragraph 1.2g states that pavement texture is not roughness.

How do the Boeing Bump Index, IRI and a straightedge compare?

They differ in what they compute, the length scale, the speed basis, the defining document and what each leaves unseen. Length scales are nominal; real sensitivity spreads across wavelengths.

Measure	What it computes	Length scale	Speed basis	Document and status	What it cannot see
Boeing Bump Index	Largest ratio of bump height to the acceptable-zone limit at that bump length, per point	Straightedges of 0.5 m to 120 m (394 ft)	None enters the ratio; the zone curve is the Boeing criteria	(FAA, 2009) 2.2, 2.4, Fig 2-3; Active	Repeated bumps, and averages along the runway
IRI	Accumulated quarter-car suspension motion divided by distance travelled, reported in m/km	Road-vehicle response, no single length	Road reference speed of 80 km/h (22.2 m/s) (COTO, 2016a)	(ASTM, 2021a) scope 1.1 to 1.3; highway pavements	Aircraft response; where a single event sits
3 m straightedge	Gap under a 3 m rule on a new surface	3 m	Static	(EASA, 2025) GM1 ADR-DSN.B.090(a); (UK CAA, 2025a) 3.216	Anything longer than a few metres
12 ft straightedge	Contractor smoothness check	12 ft (3.66 m)	Static	(FAA, 2018) 401-5.3g	Long waves; it is a QC check, not acceptance
Cannot see: texture and friction	Not computed by any of the three	Millimetre-scale surface	Not applicable	(FAA, 2009) 1.2g	Macrotexture, microtexture and friction

For the texture and friction row, see the comparison of macrotexture and microtexture. The hub guide holds the limit lookup, including the trap that the FAA's construction document uses 1/4 in in 12 ft while the 2009 roughness circular cites 1/4 in in 16 ft.

Why does one runway give three different answers?

Each measure looks through a different length window, and a vehicle crosses that window at speed. A wavelength λ is the length of one repeat of a surface undulation; a vehicle at ground speed v meets it at frequency f.

The formula, term by term: temporal frequency of a wavelength

Equation 1: f = v / λ

symbol	what it is in words	unit	where its value comes from	typical range
f	rate at which the vehicle meets one repeat of the surface	Hz (cycles/s)	computed	0.58 to 140.00 Hz in the table below
v	ground speed along the runway	m/s (knots)	synthetic here: 70.0 m/s	synthetic
λ	wavelength, taken here as the length scale of a measure	m (ft)	AC 150/5380-9 2.2b, 2.2c; NLR-TP-2009-190 2.1	0.5 m to 120 m

Where it comes from: the formula is derived, not fitted. A point moves along the profile at x = v t. A surface z(x) = A sin(2πx/λ) becomes z(t) = A sin(2π v t/λ) in time. A sine of 2π f t has frequency f, so f = v / λ.

Worked numbers, with the synthetic ground speed v = 70.0 m/s (136.1 kt), each frequency nominal (NLR, 2009):

• 0.5 m straightedge: 70.0 m/s ÷ 0.50 m = 140.00 Hz.
• 3 m straightedge: 70.0 ÷ 3.00 = 23.33 Hz.
• 12 ft (3.66 m) straightedge: 70.0 ÷ 3.6576 = 19.14 Hz.
• 45 m, the reach NLR gives for highway laser profilers: 70.0 ÷ 45.00 = 1.56 Hz.
• 120 m, the longest straightedge: 70.0 ÷ 120.00 = 0.58 Hz.

The same 30 m wavelength gives 0.74 Hz at the road speed of 22.22 m/s (80 km/h) and 2.33 Hz at 70.0 m/s, a ratio of 3.15, so an index tuned to road speed weights wavelengths differently from an aircraft on its take-off roll. The result helps choose a measure and gives no limit. It holds at constant speed only: a take-off roll accelerates, so one wavelength sweeps upward in frequency, and any single frequency is nominal. The script is calc/measure_windows.py. The manual clause that uses the quantity is AC 150/5380-9 2.2b to 2.2e, which fixes the 0.5 m to 120 m straightedge lengths (FAA, 2009).

Narrated film · 2:53

Three ways to read a runway profile

A comparison scene sets the length window of each measure against ground speed, then a chart shows the nominal frequencies from the synthetic 70 m/s table.

Read the film transcript

A runway can pass one roughness measure and fail another, so what is each one looking at? Here are the Boeing Bump Index, the International Roughness Index and the straightedge. The Boeing Bump Index, the BBI, scores single bumps against an aircraft-based criterion. The International Roughness Index, the IRI, averages a road vehicle response along a wheelpath. A straightedge checks a finished surface over a few metres. A runway profile is measured on survey lines: the centreline, plus a left and a right offset. The FAA circular gives offsets of three point zero five or five point two two metres, by airplane design group. They approximate the main-gear paths and are not painted lanes. A vehicle crossing a surface wave meets it at a frequency. Frequency, f, in hertz, equals ground speed, v, in metres per second, divided by wavelength, lambda, in metres. At a synthetic seventy metres per second, a three metre window is twenty-three point three three hertz. The same thirty metre wave is nought point seven four hertz at road speed and two point three three hertz at seventy metres per second. The BBI looks at every point, sampled every nought point two five metres. At each point it tries straightedges from nought point five up to one hundred and twenty metres. It divides each bump height by the acceptable limit for that length, and keeps the largest ratio. A ratio below one is acceptable, and above one is excessive or unacceptable. IRI runs a standard quarter-car over a wheelpath at eighty kilometres per hour, in metres per kilometre. Its standard is written for highway pavements, not for aircraft response. Highway laser profilers reach about forty-five metres, while an aircraft responds to about one hundred and twenty. Gain, G, equals one minus the cosine of pi times L, divided by lambda, where L is the rule length and lambda the wavelength. The largest gap, e, equals amplitude, A, times G. Take a synthetic five millimetre amplitude, a three metre rule and a thirty metre wave. The cosine of nought point three one four two radians is nought point nine five one one, so G is nought point nought four eight nine. The gap is nought point two four five millimetres. A ten millimetre peak to trough wave leaves a quarter of a millimetre under the rule, so the finish test answers a different question. In this idealised model the gain is zero at one point five metres and reaches its maximum of two at three metres. The BBI scores each point on its worst single bump, so repeated bumps are missed. Texture and friction are separate, because the circular states that texture is not roughness. Choose by the decision: the BBI for a single bump, an average index for a level, a straightedge for a new surface finish. This is engineering background; decisions to operate a runway belong to the aerodrome operator. Next, read how to find the bumps that load the landing gear, and the hub guide on thresholds.

What does the Boeing Bump Index see?

The BBI sees single events. At each survey point the AC computes bump height and bump length over every straightedge from 0.5 m up, divides each height by the acceptable-zone limit at that length, and keeps the largest ratio; below 1.0 is acceptable, and above 1.0 is excessive or unacceptable (FAA, 2009). The profile is sampled at 0.25 m (0.82 ft) and the longest straightedge is 120 m (394 ft) (2.2b to 2.2e), so up to 479 straightedges can apply per point. The longest bump length on Figure 2-3 is 60 m, half the longest straightedge. The AC defines airfield roughness by effects on the aircraft, such as fatigue and cockpit vibration, not by passenger ride (1.2b, 1.2d). This guide does not reprint the curve; see the circular explained.

What does IRI see on a runway?

IRI sees the average ride response of a road vehicle. It is computed by simulating a standard quarter-car at 80 km/h over a wheelpath profile and is reported in m/km; ASTM E1926-08(2021) states its scope as highway pavements (ASTM, 2021a) (COTO, 2016a). The road guide on IRI, rut depth and texture depth, the insight on IRI interpretation and the TMH 13 IRI page cover road practice.

The writer's inference, not an FAA statement, is that IRI alone is not aircraft-relevant, because airfield roughness is defined by aircraft effects. NLR-TP-2009-190 section 2.1 records that most highway laser profilers capture wavelengths up to about 45 m, whereas an aircraft responds to wavelengths up to about 120 m (NLR, 2009). Transport Canada AC 302-023 (Issue 02, 16 December 2016), 6.2, lists IRI and the root-mean-square vertical acceleration (RMSVA) as average-roughness indices alongside the bump method; it names ASTM E1364, a rod-and-level method, not the IRI practice E1926, so read its text (Transport Canada, 2016). This guide states no runway IRI limit.

What does a straightedge see?

A straightedge sees surface finish over its own length. The 3 m rule is a construction test: EASA CS-ADR-DSN Issue 7 (16 May 2025), GM1 ADR-DSN.B.090(a), states no deviation greater than 3 mm under a 3 m straightedge for a new surface, and UK CAP 168 (13th edition) 3.216 sets out a 3 m straight edge (EASA, 2025) (UK CAA, 2025a). FAA AC 150/5370-10H (21 December 2018), 401-5.3g, sets contractor quality-control variances at less than 1/4 in in 12 ft (6.35 mm in 3.66 m), a QC check, not acceptance (FAA, 2018).

The formula, term by term: mid-chord offset gain of a straightedge

Equation 2: G = 1 − cos(π L / λ), peak offset e = A · G

symbol	what it is in words	unit	where its value comes from	typical range
G	peak gain: offset under the rule per unit wave amplitude	dimensionless	computed	0 to 2
L	straightedge length	m (ft)	3 m in EASA GM1; 12 ft in AC 150/5370-10H	3 m to 120 m
λ	wavelength of the surface wave	m	synthetic	synthetic: 1.5 m to 120 m
A	amplitude of the wave, peak	mm	synthetic: 5.0 mm	synthetic
e	largest gap between chord and surface at its midpoint	mm	computed	synthetic
k	wavenumber, 2π/λ (intermediate symbol)	rad/m	computed	synthetic
x	position along the profile (intermediate symbol)	m	synthetic	synthetic
z(x)	surface height at x (intermediate symbol)	mm	synthetic	synthetic

Where it comes from: derived, as an idealisation. Lay a chord of length L over z = A sin(kx), k = 2π/λ, resting on the surface at x − L/2 and x + L/2. The midpoint gap is e(x) = z(x) − ½[z(x − L/2) + z(x + L/2)]. Since sin(a + b) + sin(a − b) = 2 sin a cos b, the bracket equals A sin(kx) cos(kL/2), so e = A sin(kx) [1 − cos(πL/λ)], with peak A · G. It is not the AC 150/5380-9 bump-height computation (2.4a).

Worked numbers, synthetic wave amplitude A = 5.0 mm (10 mm peak to trough), L = 3 m, λ = 30 m: π L/λ = 0.3142 rad, cos = 0.9511, G = 0.0489, e = 5.0 × 0.0489 = 0.245 mm. The 12 ft rule gives 0.362 mm. In this fixed-chord model the mid-chord offset is zero at λ = 1.5 m (G = 0.0000), and G reaches its maximum of 2 at λ = 3 m (G = 2.0000), and again at λ = L/(2n+1), such as 1 m and 0.6 m. Equation 2 is a mid-chord-offset filter, not a rule resting on crests: a real straightedge on a 1.5 m sinusoid rests on the crests and shows the full 2A gap. The script is calc/chord_gain.py.

This shows a 10 mm peak-to-trough 30 m wave, in the range NLR says an aircraft responds to, leaves 0.245 mm under the rule against a 3 mm finish test: the test answers a different question. It assumes a sinusoid and a rigid chord whose ends stay on the profile (a mid-chord-offset idealisation, not a rule resting on crests).

What do all three miss?

Multiple events, and texture. The BBI scores each point on its worst single bump. Transport Canada AC 302-023 6.1(6) notes that an isolated-bump analysis does not account for multiple or sequential bumps (Transport Canada, 2016). An FAA research report, DOT/FAA/TC-23/09 (January 2023), which does not constitute FAA policy, describes Type 1 shock, Type 2 short-wavelength roughness (struts react independently) and Type 3 long-wavelength roughness (pitch and plunge, coupling main and nose gear), with speed of encounter a primary factor (FAA, 2023). Its research parameters (100 ft and 25 ft simulated profilographs, a 12 ft rolling straightedge) are not FAA limits. Texture is a separate matter: AC 150/5380-9 1.2g says texture is not roughness (FAA, 2009).

Which measure answers which question?

Choose by the decision, not by habit.

• A single bump against an aircraft criterion: the BBI, with the interval and survey lines in AC 150/5380-9 2.2.
• An average level along the runway: an index such as IRI or RMSVA (Transport Canada 6.2), read as a road-derived or acceleration average.
• Finish of a new surface: a straightedge at the length the governing document names.
• Repeated events, texture, friction: none of the three.

Common errors are treating IRI and the BBI as one scale, quoting road IRI limits for runways, and mixing 12 ft with 16 ft.

Frequently asked

What is the difference between the Boeing Bump Index and IRI?

The Boeing Bump Index (BBI) in FAA Advisory Circular 150/5380-9 scores single bumps against an aircraft-based criterion, using straightedges of 0.5 m to 120 m. The International Roughness Index (IRI) in ASTM E1926 averages a road quarter-car response at 80 km/h along a wheelpath and is defined for highway pavements. They are different scales and should not be converted into one another.

Is IRI used on runways?

Transport Canada AC 302-023 (Issue 02, 2016), section 6.2, lists the International Roughness Index (IRI) among the indices that express average runway roughness, beside the bump method. FAA Advisory Circular 150/5380-9 defines airfield roughness by effects on the aircraft rather than passenger ride. This guide states no runway IRI limit, because none appears in the documents catalogued here.

Why can a surface pass a 3 m straightedge test and still be rough?

A 3 m straightedge sees wavelengths of a few metres. In a synthetic idealised model, a 30 m wave of 5.0 mm amplitude leaves only 0.245 mm under a 3 m rule, far below the 3 mm finish test in EASA CS-ADR-DSN GM1 ADR-DSN.B.090(a). The test checks surface finish, so long waves, which NLR says aircraft respond to, need a longer-window measure.

Why does ground speed matter when comparing roughness measures?

Frequency equals ground speed divided by wavelength (f = v / λ). A 30 m wavelength meets a vehicle at 0.74 Hz at 22.22 m/s (80 km/h) and at 2.33 Hz at a synthetic 70.0 m/s, a ratio of 3.15. A road-speed index therefore weights wavelengths differently from an aircraft on its take-off roll, and every single frequency per measure is nominal.

Do the Boeing Bump Index, IRI and a straightedge measure texture or friction?

No. FAA Advisory Circular 150/5380-9, paragraph 1.2g, states that pavement texture is not roughness. Macrotexture, microtexture and friction need their own methods and are compared in a separate guide. The three measures here describe the longitudinal profile at different length scales and say nothing about wet-surface friction or water drainage.

Where this connects

The guide on how to find the bumps that load the landing gear takes the single-event idea into a profile, and the hub on runway surface thresholds and the documents that set them holds the limit lookup. For checking any roughness survey, see validation sections. Last reviewed 3 November 2026 against the editions named.

References

1. COTO (2016a). TMH13: Automated Road Condition Assessments. Part C: Roughness. Committee Draft Final, May 2016. Original source.Sections C.2–C.5; Table C.5 is for calibration-section selection
2. FAA (2009). AC 150/5380-9, Guidelines and Procedures for Measuring Airfield Pavement Roughness. 30 Sep 2009, no revision letter, Active. Original source.Foreword; para 1.2b-g; para 2; 2.1; 2.2a-g; 2.3a-b; 2.4a-d; 2.5a-b; Fig 2-3; Fig 2-4; Ch 3 (3.1, 3.2)
3. ASTM (2021a). Standard Practice for Computing International Roughness Index of Roads from Longitudinal Profile Measurements (ASTM E1926). E1926-08(2021). Original source.Scope 1.1-1.3
4. FAA (2018). AC 150/5370-10H, Standard Specifications for Construction of Airports. 21 Dec 2018, Active; errata 19 Aug 2020 (draft 10J not final). Original source.401-5.3g (p. 280); 401-6.2d; 401-6.2e; 401-8.1d; 501-4.12; 501-5.3g; 501-6.6b(4)
5. EASA (2025). Certification Specifications and Guidance Material for Aerodrome Design (CS-ADR-DSN). Issue 7, 16 May 2025 (ED Decision 2025/004/R; applicable 24 May 2025; PDF corrected 16 Jun 2025). Original source.CS ADR-DSN.B.090 (a)-(d); GM1 ADR-DSN.B.090(a); GM1 ADR-DSN.B.191 (m)-(p)
6. UK CAA (2025a). CAP 168, Licensing of Aerodromes. 13th edition, July 2025, incorporating Amendment 15 (page version 13, re-issued 2 Dec 2025). Original source.3.210-3.217 (3.216 3 m straight edge); 3.214-3.215; 3.232-3.254
7. FAA (2023). Method to Identify and Quantify Multiple-Event Roughness, DOT/FAA/TC-23/09 (T. Gerardi, M. Gerardi, APR Consultants). Final Report, January 2023 (does not constitute FAA policy). Original source.Exec summary pp. ix-x; Sec 2 p. 2; Sec 2.1-2.3 pp. 2-3; Sec 3 p. 4; Sec 3.1 p. 9; Table 1 p. 7; App E pp. E-3/E-4, E-8/E-9
8. Transport Canada (2016). AC 302-023, Measurement and Evaluation of Runway Roughness. Issue 02, effective 16 Dec 2016 (current per TC page, checked 3 Nov 2026). Original source.4.3(3); 5.1-5.3; 6.1 with Fig 1 and Table 1; 6.2 (IRI, RMSVA, RCI)
9. NLR (2009). Evaluation Methods for longitudinal evenness of runway pavements: An overview of recent Dutch experiences, NLR-TP-2009-190 (P.A. van Gelder, M.J.A. Stet). June 2009. Original source.Sec 1 pp. 5-6; Sec 2.1 p. 7 and Table 1; 2.2; Elaborate Survey pp. 9-10; 3.1 p. 11; 3.3 p. 13; 3.5 p. 16; 4 p. 18

Use the manual edition and project requirements that apply to your work. Worked examples use synthetic values and are not site-specific design advice.