NASA/TM—2019-220153

Calculating Factors of Safety and Margins of

Safety From Interaction Equations
John K. Ramsey
Glenn Research Center, Cleveland, Ohio

September 2019
 NASA STI Program . . . in Profile

Since its founding, NASA has been dedicated • CONTRACTOR REPORT. Scientific and
to the advancement of aeronautics and space science. technical findings by NASA-sponsored
The NASA Scientific and Technical Information (STI) contractors and grantees.
Program plays a key part in helping NASA maintain
this important role. • CONFERENCE PUBLICATION. Collected
papers from scientific and technical
conferences, symposia, seminars, or other
The NASA STI Program operates under the auspices
meetings sponsored or co-sponsored by NASA.
of the Agency Chief Information Officer. It collects,
organizes, provides for archiving, and disseminates • SPECIAL PUBLICATION. Scientific,
NASA’s STI. The NASA STI Program provides access technical, or historical information from
to the NASA Technical Report Server—Registered NASA programs, projects, and missions, often
(NTRS Reg) and NASA Technical Report Server— concerned with subjects having substantial
Public (NTRS) thus providing one of the largest public interest.
collections of aeronautical and space science STI in
the world. Results are published in both non-NASA • TECHNICAL TRANSLATION. English-
channels and by NASA in the NASA STI Report language translations of foreign scientific and
Series, which includes the following report types: technical material pertinent to NASA’s mission.

• TECHNICAL PUBLICATION. Reports of For more information about the NASA STI
completed research or a major significant phase program, see the following:
of research that present the results of NASA
programs and include extensive data or theoretical • Access the NASA STI program home page at
analysis. Includes compilations of significant http://www.sti.nasa.gov
scientific and technical data and information
deemed to be of continuing reference value. • E-mail your question to [email protected]
NASA counter-part of peer-reviewed formal
• Fax your question to the NASA STI
professional papers, but has less stringent
Information Desk at 757-864-6500
limitations on manuscript length and extent of
graphic presentations.
• Telephone the NASA STI Information Desk at
757-864-9658
• TECHNICAL MEMORANDUM. Scientific
and technical findings that are preliminary or of • Write to:
specialized interest, e.g., “quick-release” reports, NASA STI Program
working papers, and bibliographies that contain Mail Stop 148
minimal annotation. Does not contain extensive NASA Langley Research Center
analysis. Hampton, VA 23681-2199
NASA/TM—2019-220153

Calculating Factors of Safety and Margins of

Safety From Interaction Equations
John K. Ramsey
Glenn Research Center, Cleveland, Ohio

National Aeronautics and

Space Administration

Glenn Research Center

Cleveland, Ohio 44135

September 2019
 Level of Review: This material has been technically reviewed by technical management.

Available from
NASA STI Program National Technical Information Service
Mail Stop 148 5285 Port Royal Road
NASA Langley Research Center Springfield, VA 22161
Hampton, VA 23681-2199 703-605-6000

This report is available in electronic form at http://www.sti.nasa.gov/ and http://ntrs.nasa.gov/

 Calculating Factors of Safety and Margins of
Safety From Interaction Equations
John K. Ramsey
National Aeronautics and Space Administration
Glenn Research Center
Cleveland, Ohio 44135

Summary
This report presents the derivation of the relationship between the factor of safety, the margin of
safety, and a given interaction equation in two- or three-dimensional space. Factors of safety and margins
of safety can be calculated from these interaction equations using either the closed-form solutions
provided or the numerical methods mentioned in this report. The dual definition of the factor of safety is
presented.

Introduction
For structures under combined loading, interaction equations have been used to predict structural
failure without having to determine the principal stresses. The interaction equations account for the
simultaneous effect of one stress (or load, or moment) component on another, for all possible component
stress (or load, or moment) combinations.
In aerospace engineering, margins of safety are used to indicate the strength capability remaining in a
structure. This report presents closed-form solutions of, and methods to determine, factors of safety and
consequently margins of safety from interaction equations.

Nomenclature
a exponent in interaction equation
b exponent in interaction equation
C1,2,3 constants in Equations (11) to (16) representing trigonometric functions for a given stress state
(cf. Figure 1)
c exponent in interaction equation
F allowable stress
FS factor of safety, definition 1. Calculated ratio S2/S1. See Appendix A.
Fsu allowable ultimate shear stress
Fsy allowable yield shear stress
Ftu allowable tensile ultimate stress
Fty allowable tensile yield stress
f calculated stress
 factor of safety, definition 2. Specified multiplying factor applied to calculated stresses (or
corresponding calculated loads and or moments) for the purpose of creating a margin of safety.
See Appendix A.
u ultimate factor of safety, definition 2. A specified multiplying factor applied to calculated stresses
for the purpose of creating a margin of safety for the ultimate failure mode. See Appendix A.

NASA/TM—2019-220153 1
y yield factor of safety, definition 2. A specified multiplying factor applied to calculated stresses for
the purpose of creating a margin of safety for the yield failure mode. See Appendix A.
MS margin of safety. A measure of a structure’s predicted reserve strength in excess of the product of
the load or stress under consideration and the applicable factor of safety (definition 2 of factor of
safety)
MSu margin of safety for the tensile, compressive, or shear ultimate condition
MSy margin of safety for the tensile, compressive, or shear yield condition
P point in stress-ratio space that is interior to the failure surface or failure curve
P′ point in stress-ratio space on the failure surface or failure curve
R stress ratio, f /F or () f / F
Rb bending stress ratio
Rs shear stress ratio
Rsu shear stress ratio for the ultimate condition
Rsy shear stress ratio for the yield condition
Rt tensile stress ratio
Rtu tensile stress ratio for the ultimate condition
Rty tensile stress ratio for the yield condition
S1 magnitude of stress state corresponding to point P, interior to the failure surface in stress-ratio
space. See Figure 1 and Figure 4 to Figure 9.
S2 magnitude of stress state corresponding to point P′ on the failure surface in stress-ratio space. See
Figure 1 and Figure 4 to Figure 9.
SF safety factor, used synonymously for the factor of safety definition 2 ()
α angle defined in Figure 1
∆ distance between points P′ and P
φ angle defined in Figure 1
( )′ quantity at structural failure
( )i,j,k indices representing different stress types, stress states, or failure modes

Stress Ratios
Interaction equations characterize combinations of stress, loads, or moments that cause structural
failure (Refs. 1 to 7). These equations have been expressed in terms of load ratios, moment ratios, or
stress ratios, where these ratios consist of loads, moments, or stresses of the same character. These
interaction equations manifest themselves in the form of two-dimensional (2D) curves, three-dimensional
(3D) surfaces, or boundaries in multidimensional space, and they can be created using theory (e.g., Ref. 3)
or experiment (e.g., Ref. 1).
Reference 7 is one of the earlier publications describing the use of interaction equations to determine
the allowable loads for a structure under combined loading. The approach was denoted therein as the
“stress-ratio method.” In this report, stress ratios will be used exclusively. However, the same outcomes
would result when using load or moment ratios. The stress ratio R, consisting of stresses of the same
character is defined as

f
R= (1)
F
where f is the calculated stress and F is the allowable stress.

NASA/TM—2019-220153 2
 The factor of safety 1 () can be incorporated into the stress ratios, and in those cases f is multiplied
by the  in the numerator of the stress ratio. Here  is used to indicate a specified quantity, not to be
confused by FS, which is a calculated quantity. Incorporating the  into Equation (1), the stress ratio
takes on the following general form:

(  ) f
R= (2)
F

The ratio takes on the following form for the yield stress condition:

Rty =
( y ) f (3)
Fty

where y is the specified yield factor of safety, and Fty is the yield tensile stress allowable. A similar
ratio can be constructed for the ultimate stress condition:

Rtu =
( u ) f
(4)
Ftu

where u is the specified ultimate factor of safety, and Ftu is the ultimate tensile stress allowable.
When a stress state involves shear, the following shear stress ratio for the yield condition is
customarily expressed as

Rsy =
( y ) f (5)
Fsy

where Fsy is the yield shear stress allowable. 2 The shear stress ratio for the ultimate condition is
customarily expressed as

( u ) f
Rsu = (6)
Fsu

where Fsu is the ultimate shear stress allowable.2

1
The use of the “factor of safety” terminology can be confusing because it has a dual definition. The different symbols
FS and , corresponding to the two different definitions of the factor of safety, are used in this report for clarity,
although this is typically not done in practice. Please see the definitions in the Nomenclature and Appendix A.
2
If the yield or ultimate shear stress allowables are not available in the literature, it is customary to approximate the
yield allowables as Fsy = 0.6Fty for alloy or carbon steels, and Fsy = 0.55Fty for stainless steels, and the ultimate
allowables as Fsu = 0.6Ftu for alloy or carbon steels, and Fsu = 0.55Ftu for stainless steels (Ref. 6).

NASA/TM—2019-220153 3
 Interaction Equations
The interaction of one stress ratio with another at failure can be characterized using an interaction
equation expressed as

( ΣRi′ )a + ( ΣR′j ) + (=
ΣRk′ ) 1
b c
for i ≠ j ≠ k (7)

where the exponents a, b, and c are determined from experimental test results as demonstrated, for
example, in Reference 1 and/or theory as demonstrated in Reference 3. The stress ratios Ri′ , R′j , and Rk′
characterize the failure stress state and do not include the specified  because doing so would not truly
reflect a failure stress state (unless  = 1). The summations and corresponding indices permit the
combination of stress ratios as dictated by experiment or theory. With the inclusion of three separate
summations, Equation (7) is more general than typically encountered, but the intent is to address some
possible as-yet-to-be-encountered interaction equations. Any stress combinations not covered by
Equation (7) can be addressed using and extending the techniques presented in this report.
Equation (7) may be plotted as a surface in 3D stress-ratio space as shown schematically in Figure 1.
This surface is sometimes referred to as a “failure surface,” or it could be described as an interaction
surface. Some interaction equations involve three stress ratios, with two of them being summed together,
and in these situations Equation (7) can be plotted in 2D stress-ratio space as shown in Figure 2, where
typically one would only encounter two stress ratios. It is up to the discretion of the analyst to decide
which axes represent a given stress ratio or sum of stress ratios. Figure 3 is a plot of a generic interaction
equation with two stress ratios.
Referring to Figure 1, the stress state at point P has coordinates (ΣRi, ΣRj, and ΣRk), and it lies within
the bounds of the failure (or interaction) surface (or equation), which indicates that the structure has
reserve strength. The stress state at point P′ has coordinates ( ΣRi′ , ΣR′j , and ΣRk′ ) and lies on the failure
(or interaction) surface, indicating that the structure does not have reserve strength at point P′.
Figure 2 shows a plot of interaction Equation (7), where the indices i = 1 and j = 2, 3 and where
Rk′ = 0. The stress state at point P has coordinates (R1, R2 + R3) and it lies within the bounds of the
interaction equation, which indicates that the structure has reserve strength. The stress state at point P′ has
coordinates ( R1′ , R2′ + R3′ ) and lies on the interaction curve, indicating that the structure has no reserve
strength at point P′.
Figure 3 shows a generic plot of interaction Equation (7) involving two stress ratios, where i = 1
and j = 2 and where Rk′ = 0. The stress state at point P has coordinates (R1, R2) and just as in the case for
Figure 2, it lies within the bounds of the interaction equation, which indicates that the structure has
reserve strength. The stress state at point P′ has coordinates ( R1′ , R2′ ) and lies on the interaction curve,
indicating that the structure has no reserve strength at point P′.

NASA/TM—2019-220153 4
 Figure 1.—Generic failure surface (in 3D stress-ratio space) illustrating
magnitude S1 of current stress-ratio state P and magnitude S2 of stress-
ratio state at failure P′ for case where all stress-ratio sums ΣR are
proportional up to failure.

Figure 2.—Generic interaction curve (in 2D stress-ratio

space) involving three stress ratios R, illustrating
current stress-ratio state P and stress-ratio state at
failure P′.

NASA/TM—2019-220153 5
 Figure 3.—Generic interaction curve (in 2D stress-
ratio space) involving two stress ratios R,
illustrating current stress-ratio state P and stress-
ratio state at failure P′.

Interaction equations describe the relationships between the stress ratio combinations that exist at
failure, but they do not necessarily describe the relationships between the stress ratios prior to failure.
Figure (A.3.2.0-1) of Reference 3, and Figure 12 on page 2.11-6 of Reference 8 present three scenarios
for the relationships between stress ratios prior to and at failure as depicted in Figure 4, Figure 5, and
Figure 6 herein. Figure 4 presents the scenario where the stress ratio sum (R2 + R3) and R1 increase
proportionately up to structural failure at point P′. Figure 5 presents the case where the stress ratio sum
(R2 + R3) remains constant as R1 increases up to structural failure at point P′. Figure 6 presents the
scenario where the stress ratio R1 remains constant as the stress ratio sum (R2 + R3) increases up to
structural failure at point P′.

Margin of Safety
In aerospace engineering, the margin of safety (MS) is used to indicate the degree to which the
structure satisfies the strength requirements. One of the most concise definitions of the MS is given in
Reference 4 as “the ratio of excess strength to the required strength,” where, in terms of stress, the excess
strength is the difference between the allowable stress and the required stress, and the required stress is
the product of the  and the calculated stress. The required stress is also known as the design stress.
This is expressed mathematically as

( allowable stress ) − ( required stress ) ( allowable stress )

=MS = −1 (8)
( required stress ) (  )( calcluated stress )
A MS ≥ 0 is a prediction of adequate strength for the stress state at hand. In certain situations, a
minimum positive MS may be imposed to account for unknowns not covered by the  in Equation (8).

NASA/TM—2019-220153 6
 Figure 4.—Interaction curve (in 2D stress-ratio space)
involving three stress ratios R, illustrating magnitude
S1 of current stress-ratio state P and magnitude S2 of
stress-ratio state at failure P′, for case where stress
ratio R1 is proportional to stress ratio sum (R2 + R3) up
to failure.

Figure 5.—Interaction curve (in 2D stress-ratio space), illustrating magnitude S1

of current stress-ratio state P and magnitude S2 of stress-ratio state at failure
P′, for case where stress ratio sum (R2 + R3) remains constant up to failure.

NASA/TM—2019-220153 7
 Figure 6.—Interaction curve (in 2D stress-ratio space),
illustrating magnitude S1 of current stress-ratio state P
and magnitude S2 of stress-ratio state at failure P′, for
case where stress ratio R1 remains constant up to
failure.

For the tensile yield condition the margin of safety takes on the following form:

MS y=
( allowable stress ) − 1= Fty
− 1=
1
−1
( y ) ( calcluated stress ) ( y ) f Rty
(9a)

For the tensile ultimate condition the margin of safety takes on the following form:

( allowable stress ) Ftu 1

MSu= −=
1 −=
1 −1 (9b)
( u )( calcluated stress ) ( u ) f Rtu

For the shear yield condition the margin of safety takes on the following form:

MS =
( allowable stress ) −=1 Fsy
−=
1
1
−1
y
( y ) ( calcluated stress ) ( y ) f Rsy
(9c)

For the shear ultimate condition the margin of safety takes on the following form:

( allowable stress ) Fsu 1

MS= −=
1 −=
1 −1 (9d)
u
( u )( calcluated stress ) ( u ) f Rsu

The margin of safety Equations (8) and (9) are typical forms the analyst may encounter. The product
of the yield factor of safety and the calculated stress is known as the design yield stress. The product of
the ultimate factor of safety and the calculated stress is known as the design ultimate stress.
In addition to tension and shear stress states, Equations (9a) and (9b) may also be used for the
compressive stress state as well, for the yield and ultimate conditions, respectively, where the
compressive yield and ultimate strengths are substituted for Fty and Ftu, respectively, here and in

NASA/TM—2019-220153 8
Equations (3) and (4). The above equations assume that the yield factor of safety (definition 2) applies to
the tensile, compressive, or shear yield condition, and that the ultimate factor of safety (definition 2)
applies to the tensile, compressive, or shear ultimate condition.
Reference 5 presents interaction equations for combined load systems and their corresponding MS
equations, but it does not explicitly show how to obtain the MS equation from its corresponding
interaction equation. Reference 2 presents the MS equation and the interaction equation in terms of a
common factor in 2D stress-ratio space, creating a system of two equations and two unknowns that can be
solved for the common factor and consequently, the MS. Reference 9 extends the coverage of the MS
topic presented in Reference 5, illustrating this same system of two equations and two unknowns, but
extending the presentation in Reference 2 to that of a multidimensional stress-ratio space. One feature of
the current report is to present the derivation of the system of two equations with the common factor.
Referring to Figure 1 to Figure 6, the allowable stress combination corresponds to point P′, and the
required stress combination corresponds to point P, and Equation (8) becomes

P′ − P
MS = (10)
P

Referring to Figure 1, the following relationships may be observed:

S1 cos ϕ cos α = ΣRi → S1C1 = ΣRi (11)

S1 cos ϕ sin α = ΣR j → S1C2 = ΣR j (12)

S1 sin ϕ = ΣRk → S1C3 = ΣRk (13)

S2 cos ϕ cos α = ΣRi′ → S2C1 = ΣRi′ (14)

S2 cos ϕ sin α = ΣR′j → S2C2 = ΣR′j (15)

S2 sin ϕ = ΣRk′ → S2C3 = ΣRk′ (16)

where the trigonometric terms are constants for a given stress state, and for conciseness these are denoted
as C1, C2, and C3 as shown above.
Although the following mathematics leading up to Equation (21) could have been omitted here, by
inspection of Figure 1, it was decided to be more rigorous. Utilizing the equations for the distance
between two points in 3D space, the ratio in Equation (10) becomes

( ΣRi′ − ΣRi )2 + ( ΣR′j − ΣR j ) + ( ΣRk′ − ΣRk )

2 2
P′ − P
= (17)
( ΣRi − 0 ) + ( ΣR j − 0 ) + ( ΣRk − 0 )
P 2 2 2

Substituting Equations (11) to (16) into Equation (17) gives

P′ − P ( S2C1 − S1C1 )2 + ( S2C2 − S1C2 )2 + ( S2C3 − S1C3 )2

= (18)
P
( S1C1 )2 + ( S1C2 )2 + ( S1C3 )2

NASA/TM—2019-220153 9
Factoring out terms in Equation (18) yields

P′ − P ( S2 − S1 )2 ( C12 + C22 + C32 )

= (19)
P (
S12 C12 + C22 + C32 )
Rearranging Equation (19) simplifies to

P′ − P ( C12 + C22 + C32 )

( S2 − S1 ) S2 − S1
= = (20)
P S1 ( C12 + C22 + C32 ) S1

Substituting Equation (20) into Equation (10), the MS becomes

S 2 − S1 S 2
=
MS = −1 (21)
S1 S1

As can be seen from Equation (21), the MS is expressed in terms of the distances S1 and S2 as defined in
Figure 1, Figure 4, Figure 5, and Figure 6.

Factor of Safety
Reference 3 employs both definitions (mentioned previously) of the factor of safety. In regard to
interaction equations, and in terms of variables used in this report, Reference 3 defines the factor of safety
to be the calculated ratio of S2 to S1. The factor of safety in this regard will be denoted as FS, thus
differentiating it from the specified factor of safety, . Therefore,

S2
FS = (22a)
S1

It should be noted as mentioned previously, that on the interaction curve or surface, the stress ratios do
not incorporate the specified  (or equivalently  could be thought of as equaling 1 there). Therefore S2,
which lies on the interaction curve or surface and is a function of stress ratios per Equations (14) to (16),
does not incorporate a . However, because S1 is not on the interaction or failure surface and is a function
of stress ratios per Equations (11) to (13), it may incorporate a . This is shown symbolically in
Equation (22b), where the subscript indices i could be replaced with index j or k. The derivation of
Equation (22b) is presented in Appendix A.

f′
S2 ∑ Fii
= =
FS
S1 (  )i fi
(22b)
∑ Fi

Equation (22b) shows that the factor of safety definition 1 is a function of the factor of safety
definition 2. To avoid confusion between the factor of safety symbols in Equation (22b), the terminology

NASA/TM—2019-220153 10
“safety factor” (SF) may be employed in place of the factor of safety ( ) definition 2. Please see


Appendix A for a more detailed presentation on this.

Substituting Equation (22a) into Equation (21) gives the margin of safety in terms of the factor of
safety (definition 1):

= FS − 1
MS (23)

Again, it can be seen from comparing Equation (23) with Equation (8) that the FS is a function of .

Interaction Surfaces
In the following mathematical development through Equation (30), the stress ratios are assumed to be
proportional to each other. This means that ϕ′ = ϕ and α′ = α in Figure 1. Multiplying both sides of
Equation (22a) by cos ϕ cos α and rearranging gives

( FS ) S1 cos ϕ cos α= S 2 cos ϕ cos α (24)

and upon substituting Equations (11) and (14) into Equation (24),

( FS ) ΣRi =ΣRi′ (25)

Multiplying both sides of Equation (22a) by cos ϕ sin α and rearranging leads to

( FS ) S1 cos ϕ sin α= S2 cos ϕ sin α (26)

and upon substituting Equations (12) and (15) into Equation (26),

( FS ) ΣR j =
ΣR′j (27)

Multiplying both sides of Equation (22a) by sin ϕ and rearranging gives

( FS ) S1 sin
= ϕ S 2 sin ϕ (28)

and upon substituting Equations (13) and (16) into Equation (28),

( FS ) ΣRk =
ΣRk′ (29)

Substituting Equations (25), (27), and (29) into Equation (7) generates Equation (30), which describes the
relationships between stress ratios in terms of the FS for the scenario that all stress ratios maintain
proportionality prior to and at failure.

( FS ) ΣRi  + ( FS ) ΣR j  + ( FS ) ΣRk  =

a b c
1 (30)

Equation (30) is valid for all failure surfaces in 3D space, including exponents with noninteger values,
and corresponds to the failure path from S1 to S2 shown in Figure 1. This type of equation describing the
relationships between the stress ratios prior to, and at failure, in terms of the factor of safety FS will be
denoted from here on as the “factored interaction equation.”

NASA/TM—2019-220153 11
 For the case where ΣRi is constant,

ΣRi =ΣRi′ (31)

Substituting Equations (31), (27), and (29) into Equation (7), the factored interaction equation for the case
where ΣRi = constant and where ΣRj and ΣRk remain proportional to each other is

( ΣRi )a + ( FS ) ΣR j  + ( FS ) ΣRk  =

b c
1 (32)

and corresponds to the failure path from S1 to S2 shown in Figure 7.

For the case where ΣRj = constant and ΣRk = constant,

ΣR′j
ΣR j = (33)

and

ΣRk′
ΣRk = (34)

Figure 7.—Generic failure surface (in 3D stress-ratio space), illustrating

magnitude S1 of current stress-ratio state P and magnitude S2 of stress-
ratio state at failure P′, for case where stress ratio sum ΣRi, remains
constant up to failure.

NASA/TM—2019-220153 12
 Figure 8.—Generic failure surface in (3D stress-ratio space) illustrating
the magnitude S1 of the current stress-ratio state P and magnitude S2
of stress-ratio state at failure P′, for the case where stress ratio sum
ΣRi, and sum ΣRk remain constant up to failure.

Substituting Equations (25), (33), and (34) into Equation (7) the factored interaction equation is

( FS ) ΣRi  + ( ΣR j ) + ( ΣRk ) =1

a b c
(35)

and corresponds to the failure path from S1 to S2 shown in Figure 8.

The next section will consider failure or interaction curves in 2D stress-ratio space.

Interaction Curves
Interaction equations involving only two of the three summations in Equation (7), where Rk′ = 0 , are
of the following form:

( ΣRi′ )a + ( ΣR′j )
b
=1 for i ≠ j (36)

Equation (36) may be plotted in 2D stress-ratio space. For combined loading scenarios consisting of only
two stress ratios, the summation signs are not required. For example, when only tension and shear
loadings are considered, the interaction Equation (36) becomes

( Rt′ )a + ( Rs′ )b =
1 (37)

NASA/TM—2019-220153 13
 Equation (36) is also often used for the case where tension, bending, and shear are considered for
beam-like structures, in which case the interaction Equation (36) often is of the form

( Rt′ + Rb′ )a + ( Rs′ )b =

1 (38)

where the first summation in Equation (36) is utilized, since the tension and bending stress act normal to
the structural cross section, and thus the tension and bending stress ratios being of similar character may
be added. For the case where both stress ratio sums are proportional to each other we substitute
Equations (25) and (27) into Equation (36), yielding the factored interaction equation

( FS ) ΣRi  + ( FS ) ΣR j  =

a b
1 (39)

which could have also been obtained by deleting the last term on the left hand side of Equation (30).
For the case where one of the stress ratio sums remains constant up to failure, for example when ΣRj
is constant, then

ΣR′j
ΣR j = (40)

Substituting Equation (40) and Equation (25) into Equation (36) yields the factored interaction equation
for the case where ΣRj is constant,

( FS ) ΣRi  + ( ΣR j ) =1
a b
(41)

which could have also been obtained be deleting the last terms on the left hand side of Equation (35).
Equation (41) is equivalent in form to the case where ΣRi is constant, which is

( ΣRi )a + ( FS ) ΣR j 
b
=
1 (42)

For the rest of this report, Equation (41) will be used to cover 2D factored interaction equations for the
scenarios when one stress ratio sum remains constant up to failure.

Calculating the Factor of Safety and Consequently the Margin of Safety

Equations (30), (32), (35), (39), or (41) individually and Equation (23) constitute a system of two
equations and two unknowns (FS and MS) that may be solved for the FS and subsequently the MS.
The FS may be determined from factored interaction Equations (30), (32), (35), (39), or (41),
knowing the exponents a, b, and c and the stress ratios, Ri, Rj, and Rk. Once the FS is obtained, the MS
may be calculated using Equation (23).

NASA/TM—2019-220153 14
 In many cases a closed-form solution for the FS may be obtained directly or by using an analytical
approach such as Cardano’s Method 3 (Ref. 10) for equations up to order 3, which is outlined here. Table I
to Table III provide these closed-form solutions for the FS as well as the constants used in Cardano’s
Method for interaction equations of the forms of Equations (30), (32), and (35). By subtracting 1 from the
FS, the MS is obtained, as shown in Equation (23). Appendix B, Appendix C, and Appendix D present the
derivations of the closed-form solutions presented in the tables along with the constants used in Cardano’s
Method.
To implement Cardano’s Method, transform the factored interaction equation into the form

( FS )3 + d ( FS )2 + e ( FS ) + g =
0 (43)

Equation (43) may be solved for the factor of safety FS using Cardano’s Method as follows:

d2
1. Calculate p =
− +e
3
3
 d  de
2. Calculate=
q 2  − + g
3 3
3 2
 p q
=
3. Calculate Q   + 
 3 2

q
4. Calculate A = 3 − + Q
2

q
5. Calculate B = 3 − − Q
2

6. Calculate y1= A + B

A+ B A− B
7. Calculate y 2,3 =
− ±i 3
2 2

8. Calculate the factors of safety (roots): ( FS )=

i
yi − d 3 , where the (FS)i must be a real number

=
9. Calculate the minimum MS, MS min ( FS )i  − 1

3
Sometimes referred to in English as Cardan’s Method.

NASA/TM—2019-220153 15
 TABLE I.—FACTORS OF SAFETY FS AND CONSTANTS USED IN CARDANO’S
METHOD FOR FACTORED INTERACTION EQUATIONS OF THE FORM

( FS ) ΣRi  + ( FS ) ΣR j  + ( FS ) ΣRk  =

a b c
1
[Where a, b, and c are integers and MS = FS – 1.]

c=1
b 1 2 3
a
1
1 ------------------------------------------------ ------------------------------
ΣRi + ΣR j + ΣRk

− ( ΣR j + ΣRk ) + ( ΣR j + ΣRk ) + 4 ( ΣRi ) ( ΣRk )2 + 4 ( ΣRi )2 + ( ΣR j )

2 2 2
−ΣRk +
2  
2 ( ΣRi )
2 ------------------------------
2 ( ΣRi ) + ( ΣR j ) 
2 2

 
Cardano’s Method Cardano’s Method Cardano’s Method

( ΣR j )
2

d=0 d= d=0
( ΣRi )3
3 ΣR j + ΣRk ΣRk ΣRk
e= e= e=
( ΣRi ) ( ΣRi ) ( ΣRi ) + ( ΣR j )
3 3 3 3

1 1 1
g= − g= − g= −
( ΣRi ) ( ΣRi ) ( ΣRi ) + ( ΣR j )
3 3 3 3

c=2
1
2
( ΣRi ) + ( ΣR j ) + ( ΣRk )2
---------------------------------------------------- 2 2 ------------------------------

Cardano’s Method Cardano’s Method

( ΣR j ) + ( ΣRk ) ( ΣRk )2
2 2

d= d=
( ΣRi )3 ( ΣRi )3 + ( ΣR j )3
3 ----------------------------------------------------
e=0 e=0
1 1
g= − g= −
( ΣRi ) ( ΣRi ) + ( ΣR j )
3 3 3

c=3
1
3
( ΣRi )3 + ( ΣR j ) + ( ΣRk )3
---------------------------------------------------- ------------------------------------------------ 3
3

NASA/TM—2019-220153 16
 TABLE II.—FACTORS OF SAFETY FS AND CONSTANTS USED IN CARDANO’S
METHOD FOR FACTORED INTERACTION EQUATIONS OF THE FORM
( ΣRi )a + ( FS ) ΣR j  + ( FS ) ΣRk  =
b c
1
[Where a, b, and c are integers and MS = FS – 1.]
c 1 2 3
b

1 − ( ΣRi )
a
1 ----------------------- ---------------------
ΣR j + ΣRk

( ΣRk )2 + 4 ( ΣR j ) 1 − ( ΣR )a 
2
−ΣRk + 1 − ( ΣRi )
a

2  i 
 ---------------------
( ΣR j ) + ( ΣRk )2
2
2 ( ΣR j )
2

Cardano’s Method Cardano’s Method

d=
( ΣRk )2
d=0
( ΣR j )
3

1 − ( ΣRi )
a

3 ΣRk
e= 3
( ΣR j ) + ( ΣRk )3
3

( ΣR j )
3 e=0

g=
( ΣRi )a − 1 g=
( ΣRi )a − 1
( ΣR j ) ( ΣR j )
3 3

TABLE III.—FACTORS OF SAFETY FS FOR FACTORED

INTERACTION EQUATIONS OF THE FORM
( FS ) ΣRi  + ( ΣR j ) + ( ΣRk ) =1
a b c

[Where a, b, and c are integers and MS = FS – 1.]

a 1 − ( ΣR ) − ( ΣRk )c
b
j

ΣRi

Integer Exponents of Order 3 or Higher, or Noninteger Exponents

The previous sections presented closed-form solutions, including Cardano’s Method for equations up
to order 3, for solving for the FS from the factored interaction Equations (30), (32), and (35) (shown again
below), and subsequently the MS. Closed-form solutions also exist that can be used to solve for the FS
when the factored interaction equation is of integer order 4. However, it is more useful to adopt a general
numerical procedure to solve for the FS from the factored interaction equations of any order, including
cases where the exponents are not integers, as shown in Equations (44) and (45) (interaction equations
from Ref. 11), for example:

( FS ) Rs  + ( FS )( Rt + Rb ) 
2.5 1.5 (44)
=
1

( FS ) Rs  + ( FS ) Rt  + ( FS ) Rb =
2.5 1.5
1 (45)

NASA/TM—2019-220153 17
Additionally, factored interaction equations may consist of the three separate summations of the following
forms as presented previously in this report:

( FS ) ΣRi  + ( FS ) ΣR j  + ( FS ) ΣRk  =

a b c
1 (30)

( ΣRi )a + ( FS ) ΣR j  + ( FS ) ΣRk  =

b c
1 (32)

( FS ) ΣRi  + ( ΣR j ) + ( ΣRk ) =1

a b c
(35)

To solve for the FS in these cases, it is useful to use a numerical root-finding algorithm to find the
roots, the FS, of the rearranged factored interaction Equations (30), (32), and (35) of the following forms,
respectively:

( FS )a ( ΣRi )a + ( FS )b ( ΣR j ) + ( FS ) ( ΣRk ) − 1 =0
(46)
b c c

( ΣRi )a + ( FS )b ( ΣR j ) + ( FS ) ( ΣRk ) − 1 =0
b c c (47)

( FS )a ( ΣRi )a + ( ΣR j ) + ( ΣRk ) − 1 =0
b c
(48)

Root-finding algorithms exist in the literature and may be programmed into the computer. One such
algorithm is presented in Reference 12, where a combination of Newton-Raphson and bisection methods
are employed and has been used successfully at the NASA Glenn Research Center. Software packages
exist that inherently contain numerical root-finding algorithms, relieving the user of the need to write the
root-finding code. It is beyond the scope of this report to further address the topic of root finding. The
reader is encouraged to search out the appropriate algorithm from literature or on-line resources.

Conservative Approach in 2D Stress-Ratio Space

To calculate the minimum possible margin of safety, MSmin; that is, the shortest distance between the
current stress state P and the interaction curve, as shown in Figure 9, a suggested method is as follows:

(1) For a selected set of points P′ on the interaction curve, calculate ∆, the distance between points P′
and P, using the standard equation for the distance between two points in a plane:

( ΣRi′ − ΣRi )2 + ( ΣR′j − ΣR j )

2
∆= (49)

( ΣRi )2 + ( ΣR j )
2
(2) Calculate S1 =
(3) For the shortest distance ∆, calculate MSmin = ∆ S1 . It should be noticed from Figure 9, that the
∆ S1 ≤ ( S2 − S1 ) S1 .
minimum ∆ ≤ S2 – S1, and consequently MSmin =

NASA/TM—2019-220153 18
 Figure 9.—Conservative interpretation of margin
of safety (MS) (in 2D stress-ratio space), where
MS = ∆/S1. Current stress-ratio state P and closest
stress-ratio state at failure P′, are illustrated.

Conclusion
This report derives the relationship between factors of safety and interaction equations, denoted as the
“factored interaction equation,” enabling the determination of the corresponding margins of safety.
Factored interaction equations were considered for cases where all or some stress ratios remain
proportional up to failure and the other stress ratios remain constant. Closed-form solutions including the
constants used in Cardano’s Method, for the factor of safety and consequently the margin of safety, were
presented in terms of stress ratios. Numerical root-finding methods are useful in situations where the
interaction equations, and consequently the factored interaction equations, are of order 4 or higher, or they
contain noninteger exponents.

NASA/TM—2019-220153 19
 Appendix A.—Factor of Safety and Safety Factor
The use of the “factor of safety” (FS) terminology can be confusing because it has a dual definition of
being a calculated or a specified quantity, as explained below in definitions 1 and 2, respectively, and
because at times the terminology “safety factor” is used synonymously for the “factor of safety”
definition 2 as well. This has been the situation even in the earliest classical texts on aeronautical or
aerospace structural analysis. Below are a few examples from the classical texts that illustrate this.
References 3 and 4 use the terms “factor of safety” and “safety factor” interchangeably. In
Reference 3, Section B4, pages 44, 46, 47, and 52, the factor of safety is defined as a specified multiplier
of stresses or moments in margin of safety (MS) equations. In Section B4.5, pages 5, 7, 8, and 13, this
type of specified multiplying factor used in MS equations is referred to as “safety factor.”
In Reference 4, page 287, the factor of safety is defined as a specified multiplying factor of 1.5, while
this same multiplying factor of 1.5 is denoted as a “safety factor” on pages 289 and 292.
Reference 5 defines/implies that the factors of safety can be either a calculated or a specified quantity.
In Reference 5, page C1.6, the factor of safety is defined as the ratio of the strength of the structure to the
limit loads. On page C1.7, it further defines the yield and ultimate factors of safety as ratios of yield and
ultimate strength respectively, to limit load. In these cases the sense is that the factor of safety is a
calculated quantity. However, on the same page C1.7, both the yield and ultimate factors of safety are
also referred to in the sense of specified multiplying factors on the limit load.
The engineer needs to be aware of the context in which the terminology “safety factor” and “factor of
safety” are used, and/or their symbolic placement in equations, to determine if they are a specified or
calculated quantity.

A.1 Factor of Safety (FS), Definition 1

FS is a calculated ratio, typically of two quantities of the same character. These ratios can be loads,
stresses, or moments:

strength allowable stress allowable stress

FS = , , or (A1)
load required stress design stress

It can be seen from Figure 1 and therefore from Equations (25), (27), and (29) that

S2 ΣRi′ ΣR′j ΣRk′

FS = = = = ( ϕ = ϕ′, α = α′ ) (A2)
S1 ΣRi ΣR j ΣRk

and from Figure 7 that

S2 ΣR′j ΣRk′
FS = = = ( φ = φ′, α = α′ = 90° ) (A3)
S1 ΣR j ΣRk

and lastly from Figure 8 that

S2 ΣRi′
FS = = ( ϕ = ϕ′ = α = α′ = 0 ) (A4)
S1 ΣRi

NASA/TM—2019-220153 21
 Substituting Equation (2) into Equations (A2) to (A4), and keeping in mind that the stress ratios on
the interaction curve or surface R′ do not have a specified  (or equivalently,  could be thought of as
equaling 1 there), the relationship between the calculated FS and the specified  in terms of stress
becomes

f′
S2 ∑ Fii
=
FS =
S1 (  )i fi
(A5)
∑ Fi

where the stress subscript indices i, could be replaced by j or k and where Equation (A5) is applicable to
the stress states characterized by Equations (A2) to (A4).
If the allowable stress F and the specified factor of safety  can be taken outside of the summation
in Equation (A5), the factor of safety in its simplified form becomes

f′
∑F ∑
1
f′
=
=FS
S2
= = F ∑f′
(  ) f (  ) f (  ) ∑ f
(A6)
∑ ∑
S1
F F

If in some cases it is not permissible to take one or both of the stress F and the specified factor of
safety  outside of the summations, Equation (A5) can be simplified accordingly, but the fact remains
that the calculated factor of safety is a function of the specified factor of safety.
The margin of safety may be expressed in terms of the calculated factor of safety as derived in the
main text as Equation (23) and repeated here for completeness:

= FS − 1
MS (23)

A.2 Factor of Safety (), Definition 2

 is a specified multiplying factor applied to loads or stresses for purposes of calculating margins of
safety MS, and at times is referred to as a “safety factor”:

allowable stress
=MS −1 (A7a)
( ) calculated stress

allowable load
=MS −1 (A7b)
( ) calculated load

This definition of the  is usually associated with yield and ultimate strengths or stresses; namely, the
yield and ultimate factors of safety are often denoted symbolically as y and u, respectively.

NASA/TM—2019-220153 22
 A.3 Safety Factor
The terminology “safety factor” has been used synonymously for the factor of safety definition 2, as
previously mentioned. Applying this convention to Equations (A5) and (A7) and utilizing the symbol SF
for safety factor avoids any ambiguity as shown in Equations (A8) and (A9):

f′
∑ Fii
FS =
( SF )i fi
(A8)
∑ Fi

and

allowable stress
MS= − 1= FS − 1 (A9a)
( SF ) calculated stress

allowable load
MS= − 1= FS − 1 (A9b)
( SF ) calculated load

where Equation (23) has been incorporated into Equation (A9).

NASA/TM—2019-220153 23
 Appendix B.—Stress Ratios Maintaining Proportionality Up to Failure
This appendix derives the closed-form equations for the factor of safety (FS) and consequently the
margin of safety (MS) for several examples of factored interaction equations in the form of Equation (30):

( FS ) ΣRi  + ( FS ) ΣR j  + ( FS ) ΣRk  =

a b c
1 (B1)

involving different integer exponent values (1, 2, or 3).

B.1 a=b=c=1
The factored interaction Equation (B1) becomes

( FS ) ΣRi  + ( FS ) ΣR j  + ( FS ) ΣRk  =

1 1 1
1 (B2)

and after factoring out the factor of safety FS, it may be obtained directly:

1
FS = (B3)
ΣRi + ΣR j + ΣRk

By substituting Equation (B3) into Equation (23), the MS is

1
=MS −1 (B4)
ΣRi + ΣR j + ΣRk

B.2 a = 2, b = c = 1
The factored interaction Equation (B1) becomes

( FS ) ΣRi  + ( FS ) ΣR j  + ( FS ) ΣRk  =

2 1 1
1 (B5)

Rearranging Equation (B5) into a quadratic equation in FS,

( ΣRi )2 ( FS )2 + ( ΣR j + ΣRk ) ( FS ) − 1 =0 (B6)

The FS may be obtained using the quadratic formula

− ( ΣR j + ΣRk ) + ( ΣR j + ΣRk ) + 4 ( ΣRi )

2 2
FS = (B7)
2 ( ΣRi )
2

where the negative square root in the numerator has been omitted because the FS must be positive. If you

( ) ( ΣR j + ΣRk ) + 4 ( ΣRi ) ,
2 2
multiply the numerator and denominator of Equation (B7) by ΣR j + ΣRk +
the FS becomes

NASA/TM—2019-220153 25
 2
FS =
( ΣR j + ΣRk ) + ( ΣR j + ΣRk )
(B8)
+ 4 ( ΣRi )
2 2

which, ignoring the summation signs and considering a 2D stress-ratio space with Rk = 0, is equivalent to
the FS presented in Table A3.5.0-1 of Reference 3.
Substituting Equation (B7) into Equation (23), the MS is

− ( ΣR j + ΣRk ) + ( ΣR j + ΣRk ) + 4 ( ΣRi )

2 2
=MS −1 (B9)
2 ( ΣRi )
2

B.3 a = 3, b = c = 1
The factored interaction Equation (B1) becomes

( FS ) ΣRi  + ( FS ) ΣR j  + ( FS ) ΣRk  =

3 1 1
1 (B10)

Rearranging Equation (B10) into the form of Equation (43), repeated here for convenience,

( FS )3 + d ( FS )2 + e ( FS ) + g =
0 (43)

gives

( FS )3 ( ΣRi )3 + ( FS ) ( ΣR j + ΣRk ) − 1 =0 (B11)

Then dividing Equation (B11) by (ΣRi)3 results in

( ΣR j + ΣRk ) − 1
( FS )3 + ( FS ) =
0 (B12)
( ΣRi )3
( ΣRi )3
which, upon comparing Equations (B12) and (43), results in the following constants for Equation (43):

d=0 (B13a)

e=
( ΣR j + ΣRk )
(B13b)
( ΣRi )3
1
g= − (B13c)
( ΣRi )3
which may be used in Cardano’s Method (see section “Calculating the Factor of Safety and Consequently
the Margin of Safety”) to solve for the safety factor FS, and subsequently the MS using Equation (23).

NASA/TM—2019-220153 26
 B.4 a = b = 2, c = 1
The factored interaction Equation (B1) becomes

( FS ) ΣRi  + ( FS ) ΣR j  + ( FS ) ΣRk  =

2 2 1
1 (B14)

Rearranging Equation (B14) into a quadratic equation in FS,

( i ) ( j )  ( )
 ΣR 2 + ΣR 2  FS 2 + ΣR FS − 1 =0
k( ) (B15)

the factor of safety FS may be obtained using the quadratic formula

( ΣRk )2 + 4 ( ΣRi )2 + ( ΣR j )

2
−ΣRk +
 
FS = (B16)
2 ( ΣRi ) + ( ΣR j ) 
2 2
 

Substituting Equation (B16) into Equation (23), the MS is

( ΣRk )2 + 4 ( ΣRi )2 + ( ΣR j )

2
−ΣRk +
 
=MS −1 (B17)
2 ( ΣRi ) + ( ΣR j ) 
2 2
 

B.5 a = 3, b = 2, c = 1
The factored interaction Equation (B1) becomes

( FS ) ΣRi  + ( FS ) ΣR j  + ( FS ) ΣRk  =

3 2 1
1 (B18)

and after rearranging Equation (B18) into the form of Equation (43), it is,

( FS )3 ( ΣRi )3 + ( FS )2 ( ΣR j ) + ( FS ) ΣRk − 1 =0
2
(B19)

Then dividing Equation (B19) by (ΣRi)3 it gives

( ΣR j )
2
ΣRk 1
( FS )3
+ ( FS )
2
+ ( FS ) − =
0 (B20)
( ΣRi )3
( ΣRi )3
( ΣRi )3

NASA/TM—2019-220153 27
which, upon comparing Equations (B20) and (43), results in the following constants for Equation (43):

( ΣR j )
2
d= (B21a)
( ΣRi )3
ΣRk
e= (B21b)
( ΣRi )3
1
g= − (B21c)
( ΣRi )3
which may be used in Cardano’s Method (see section “Calculating the Factor of Safety and Consequently
the Margin of Safety”) to solve for the factor of safety FS, and subsequently the MS, using Equation (23).

B.6 a = 3, b = 3, c = 1
The factored interaction Equation (B1) becomes

( FS ) ΣRi  + ( FS ) ΣR j  + ( FS ) ΣRk  =

3 3 1
1 (B22)

Rearranging Equation (B22) into the form of Equation (43) gives

( FS )3 ( ΣRi )3 + ( ΣR j )
3
+ ( FS ) ΣRk − 1 =0 (B23)
 

Then dividing Equation (B23) by ( ΣRi ) + ( ΣR j )  gives

3 3
 

ΣRk 1
( FS )3 + ( FS ) − =
0
( ΣRi )3 + ( ΣR j ) ( ΣRi )3 + ( ΣR j )
3 3 (B24)

which upon comparing Equations (B24) and (43), results in the following constants for Equation (43):

d=0 (B25a)

ΣRk
e=
+ ( ΣR j )
(B25b)
( ΣRi )3 3

1
g= −
+ ( ΣR j )
(B25c)
( ΣRi ) 3 3

which may be used in Cardano’s Method (see section “Calculating the Factor of Safety and Consequently
the Margin of Safety”) to solve for the factor of safety FS, and subsequently the MS, using Equation (23).

NASA/TM—2019-220153 28
 B.7 a=b=c=2
The factored interaction Equation (B1) becomes
( FS ) ΣRi  + ( FS ) ΣR j  + ( FS ) ΣRk  =
2 2 2
1 (B26)

After factoring out the factor of safety FS, it may be obtained directly:

1
FS =
( ΣRi )2 + ( ΣR j )
(B27)
+ ( ΣRk )
2 2

Substituting Equation (B27) into Equation (23), the MS is

1
=MS −1
+ ( ΣR j ) + ( ΣRk )
(B28)
( ΣRi ) 2 2 2

B.8 a = 3, b = c = 2
The factored interaction Equation (B1) becomes

( FS ) ΣRi  + ( FS ) ΣR j  + ( FS ) ΣRk  =

3 2 2
1 (B29)

and after rearranging Equation (B29), it is as follows:

( FS )3 ( ΣRi )3 + ( FS )2 ( ΣR j ) + ( ΣRk )  − 1 =0

2 2
 (B30)


Dividing Equation (B30) through by (ΣRi)3 then gives

( ΣR j ) + ( ΣRk )2 −
2
1
( FS )3
+ ( FS )
2
=
0 (B31)
( ΣRi ) 3
( ΣRi ) 3

By comparing Equations (B31) and (43), the following constants for Equation (42) are obtained:

( ΣR j ) + ( ΣRk )
2 2
d= (B32a)
( ΣRi )3
e=0 (B32b)

1
g= − (B32c)
( ΣRi )3
which may be used in Cardano’s Method (see section “Calculating the Factor of Safety and Consequently
the Margin of Safety”) to solve for the safety factor FS, and subsequently the MS using Equation (23).

NASA/TM—2019-220153 29
 B.9 a = b = 3, c = 2
Again, starting with the factored interaction Equation (B1),

( FS ) ΣRi  + ( FS ) ΣR j  + ( FS ) ΣRk  =

3 3 2
1 (B33)

and rearranging Equation (B33) gives

( FS )3 ( ΣRi )3 + ( ΣR j )
3
+ ( FS ) ( ΣRk )2 − 1 =0
2
 (B34)


Then dividing Equation (B34) through by ( ΣRi ) + ( ΣR j )  gives

3 3
 

( ΣRk )2 1
( FS )3 + ( FS )2 − =
0 (B35)
( i) ( j) ( i) ( j)
3 3 3 3
Σ R + Σ R Σ R + Σ R

By comparing Equations (B35) and (43), the following constants for Equation (43) are obtained:

d=
( ΣRk )
2
(B36a)
( ΣRi )3 + ( ΣR j )
3

e=0 (B36b)

1
g= −
+ ( ΣR j )
(B36c)
( ΣRi ) 3 3

which may be used in Cardano’s Method (see section “Calculating the Factor of Safety and Consequently
the Margin of Safety”) to solve for the factor of safety FS and subsequently the MS using Equation (23).

B.10 a = b = c =3
The factored interaction Equation (B1) becomes

( FS ) ΣRi  + ( FS ) ΣR j  + ( FS ) ΣRk  =

3 3 3
1 (B37)

and after factoring out the factor of safety FS, it may be obtained directly:

1
FS =
( ΣRi )3 + ( ΣR j )
(B38)
+ ( ΣRk )
3 3 3

Substituting Equation (B38) into Equation (23), the MS is

1
=MS −1
+ ( ΣR j ) + ( ΣRk )
(B39)
( ΣRi )
3 3 3 3

NASA/TM—2019-220153 30
 Appendix C.—One Stress Ratio Sum Remaining Constant up to Failure
This appendix derives the closed-form equations for the factor of safety (FS) and consequently the
margins of safety (MS) for several examples of factored interaction equations in the form of Equation (32):

( ΣRi )a + ( FS ) ΣR j  + ( FS ) ΣRk  =

b c
1 (C1)

when one exponent (here, a) is an integer constant and the others are different integer exponent values
(1, 2, or 3).

C.1 b=c=1
The factored interaction Equation (C1) is

( ΣRi )a + ( FS ) ΣR j  + ( FS ) ΣRk  =

1 1
1 (C2)

Factoring out the factor of safety FS,

FS ( ΣR j + ΣRk ) = 1 − ( ΣRi )
a
(C3)

and solving for the FS gives

1 − ( ΣRi )
a
FS = (C4)
ΣR j + ΣRk

Upon substituting Equation (C4) into Equation (23), the MS becomes

1 − ( ΣRi )
a
=MS −1 (C5)
ΣR j + ΣRk

C.2 b = 2, c = 1
The factored interaction Equation (C1) becomes

( ΣRi )a + ( FS ) ΣR j  + ( FS ) ΣRk  =

2 1
1 (C6)

Recognizing that Equation (C6) is a quadratic equation in FS,

( FS )2 ( ΣR j ) + ( FS )1 ( ΣRk )1 + ( ΣRi )a − 1 =0
2
(C7)

NASA/TM—2019-220153 31
and solving for FS results in

( ΣRk )2 + 4 ( ΣR j ) 1 − ( ΣR )a 
2
−ΣRk +
 i 

FS = (C8)
2 ( ΣR j )
2

where the negative square root in the numerator has been omitted because the FS must be positive.
Substituting Equation (C8) into Equation (23), the MS becomes

( ΣRk )2 + 4 ( ΣR j ) 1 − ( ΣR )a 
2
−ΣRk +  i 

=MS −1 (C9)
2 ( ΣR j )
2

C.3 b = 3, c = 1
The factored interaction Equation (C1) becomes

( ΣRi )a + ( FS ) ΣR j  + ( FS ) ΣRk  =

3 1
1 (C10)

transforming Equation (C10) into the form of Equation (43), repeated here for convenience,

( FS )3 + d ( FS )2 + e ( FS ) + g =
0 (43)

it follows that

( FS )3 ( ΣR j ) + ( FS )( ΣRk ) + ( ΣRi ) − 1 =0
3 a
(C11)

( )
3
Dividing Equation (C11) by ΣR j gives

( ΣRk ) + ( ΣRi )a − 1 =
( FS ) 3
+ ( FS ) 0 (C12)
( ΣR j ) ( ΣR j )
3 3

By comparing Equations (C12) and (43), the following constants for Equation (43) are obtained:

d=0 (C13a)

( ΣRk )
e= (C13b)
( ΣR j )
3

( ΣRi ) − 1
a
g= (C13c)
( ΣR j )
3

NASA/TM—2019-220153 32
which may be used in Cardano’s Method (see section “Calculating the Factor of Safety and Consequently
the Margin of Safety”) to solve for FS and consequently, the MS using Equation (23).

C.4 b=c=2
The factored interaction Equation (C1) becomes

( ΣRi )a + ( FS ) ΣR j  + ( FS ) ΣRk  =

2 2
1 (C14)

Rearranging and factoring out the FS gives

( FS )2 ( ΣR j ) + ( ΣRk )  = 1 − ( ΣRi )

2 2 a
(C15)
 

and solving for the FS results in

1 − ( ΣRi )
a
FS = (C16)
( ΣR j ) + ( ΣRk )2
2

Substituting Equation (C16) into Equation (23), the MS is

1 − ( ΣRi )
a
=MS −1 (C17)
( ΣR j ) + ( ΣRk )2
2

C.5 b = 3, c = 2
The factored interaction Equation (C1) becomes

( ΣRi )a + ( FS ) ΣR j  + ( FS ) ΣRk  =

3 2
1 (C18)

Transforming Equation (C18) into the form of Equation (43), as follows,

( FS )3 ( ΣR j ) + ( FS ) ( ΣRk ) + ( ΣRi ) − 1 =0
3 2 2 a
(C19)

( )
3
Dividing Equation (C19) by ΣR j gives

( ΣRk )2 + ( ΣRi )a − 1 =
( FS )3
+ ( FS )
2
0 (C20)
( ΣR j ) ( ΣR j )
3 3

NASA/TM—2019-220153 33
By comparing Equations (C20) and (43), the following constants for Equation (43) are obtained:

( ΣRk )2
d= (C21a)
( ΣR j )
3

e=0 (C21b)

g=
( ΣRi ) − 1
a
(C21c)
( ΣR j )
3

which may be used in Cardano’s Method (see section “Calculating the Factor of Safety and Consequently
the Margin of Safety”) to solve for FS and consequently, the MS using Equation (23).

C.6 b = 3, c = 3
The factored interaction Equation (C1) becomes

( ΣRi )a + ( FS ) ΣR j  + ( FS ) ΣRk  =

3 3
1 (C22)

Rearranging and factoring out FS in Equation (C22) gives

( FS )3 ( ΣR j ) + ( ΣRk )  = 1 − ( ΣRi )

3 3 a
 (C23)


and solving for FS yields

1 − ( ΣRi )
a
FS = 3 (C24)
( ΣR j ) + ( ΣRk )
3 3

Substituting Equation (C24) into Equation (23), the MS becomes

1 − ( ΣRi )
a
=MS 3 −1 (C25)
( ΣR j ) + ( ΣRk )3
3

NASA/TM—2019-220153 34
 Appendix D.—Two Stress Ratio Sums Remaining Constant up to Failure
This appendix derives the closed-form equation for the factor of safety (FS) and consequently the
margin of safety (MS) for factored interaction equations in the form of Equation (35):

( FS ) ΣRi  + ( ΣR j ) + ( ΣRk ) =1

a b c
(D1)

involving any integer exponent.

By rewriting Equation (D1) in the form

( FS )a ( ΣRi )a = 1 − ( ΣR j ) − ( ΣRk )c
b
(D2)

and solving for the FS, the result is

( ΣR j ) − ( ΣRk )c
a 1− b

FS = (D3)
ΣRi

Substituting Equation (D3) into Equation (23), the MS becomes

( ΣR j ) − ( ΣRk )c
a 1− b

=MS −1 (D4)
ΣRi

NASA/TM—2019-220153 35
 References
1. Steeve, B.E.; and Wingate, R.J.: Aerospace Threaded Fastener Strength in Combined Shear and
Tension Loading. NASA/TM—2012-217454, 2012. http://ntrs.nasa.gov
2. Sarafin, Thomas P.; and Larson, Wiley: Spacecraft Structures and Mechanisms—From Concept to
Launch. Springer, Netherlands, 1995, p. 244.
3. Astronautic Structures Manual. Vol. 1, Sect. A3, NASA TM X–73305, 1975.
4. Peery, David J.: Aircraft Structures. McGraw-Hill Book Co., New York, NY, 1950.
5. Bruhn, E.F.: Analysis and Design of Flight Vehicle Structures. S.R. Jacobs & Associates, Inc., Tri-
State Offset Company, 1973, pp. C1.6–C1.7.
6. Barrett, Richard T.: Fastener Design Manual. NASA RP–1228, 1990, p. 21. http://ntrs.nasa.gov
7. Shanley, F.R.; and Ryder, E.I.: Stress Ratios: The Answer to the Combined Loading Problem.
Aviation, 1937.
8. Blodgett, Omer W.: Design of Welded Structures. James F. Lincoln Arc Welding Foundation,
Cleveland, OH, 1966.
9. McCombs, William F.: A Supplement to Analysis & Design of Flight Vehicle Structures, Bruhn, For
Increased Scope and Usefulness. Datatec, Dallas, TX, 1998.
10. Korn, Granino A.; and Korn, Theresa M.: Mathematical Handbook for Scientists and Engineers:
Definitions, Theorems, and Formulas for Reference and Review. McGraw-Hill, New York, NY,
1961, p. 23.
11. Requirements for Threaded Fastening Systems in Spaceflight Hardware. NASA–STD–5020A, 2018.
12. Press, William H., et al.: Numerical Recipes: The Art of Scientific Computing. University Press,
Cambridge, MA, 1990, p. 258.

NASA/TM—2019-220153 36