Stress analysis is a primary task for civil, mechanical and aerospace engineers involved in the design of structures of all sizes, such as tunnels, bridges and dams, aircraft and rocket bodies, mechanical parts, and even plastic cutlery and staples. Stress analysis is also used in the maintenance of such structures, and to investigate the causes of structural failures.
Typically, the input data for stress analysis are a geometrical description of the structure, the properties of the materials used for its parts, how the parts are joined, and the maximum or typical forces that are expected to be applied to each point of the structure. The output data is typically a quantitative description of the stress over all those parts and joints, and the deformation caused by those stresses. The analysis may consider forces that vary with time, such as engine vibrations or the load of moving vehicles. In that case, the stresses and deformations will also be functions of time.
In engineering, stress analysis is often a tool rather than a goal in itself; the ultimate goal being the design of structures and artifacts that can withstand a specified load, using the minimum amount of material (or satisfying some other optimality criterion).
Stress analysis may be performed through classical mathematical techniques, analytic mathematical modelling or computational simulation, through experimental testing techniques, or a combination of methods.
General principles 
Stress analysis is specifically concerned with solid objects. The study of stresses in liquids and gases is the subject of fluid mechanics.
Stress analysis adopts the macroscopic view of materials characteristic of continuum mechanics, namely that all properties of materials are homogeneous at small enough scales. Thus, even the smallest particle considered in stress analysis still contains an enormous number of atoms, and its properties are averages of the properties of those atoms.
In stress analysis one normally disregards the physical causes of forces or the precise nature of the materials. Instead, one assumed that the stresses are related to strain of the material by known constitutive equations.
By Newton's laws of motion, any external forces that act on a system must be balanced by internal reaction forces, or cause the particles in the affected part to accelerate. In a solid object, all particles must move substantially in concert in order to maintain the object's overall shape. It follows that any force applied to one part of a solid object must give rise to internal reaction forces that propagate from particle to particle throughout an extended part of the system. With very rare exceptions (such as ferromagnetic materials or planet-scale bodies), internal forces are due to very short range intermolecular interactions, and are therefore manifested as surface contact forces between adjacent particles — that is, as stress.
Fundamental problem 
The fundamental problem in stress analysis is to determine the distribution of internal stresses throughout the system, given the external forces that are acting on it. In principle, that means determining, implicitly or explicitly, the Cauchy stress tensor at every point.
The external forces may be body forces (such as gravity or magnetic attraction), that act throughout the volume of a material; or concentrated loads (such as friction between an axle and a bearing, or the weight of a train wheel on a rail), that are imagined to act over a two-dimensional area, or along a line, or at single point. The same net external force will have a different effect on the local stress depending on whether it is concentrated or spread out.
Types of structures 
In civil engineering applications, one typically considers structures that can practically static equilibrium: that is, are either unchanging with time, or are changing slowly enough for viscous stresses to be unimportant. By Newton's laws of motion, any external forces that act on such a system must be balanced by internal reaction forces. It follows that any force applied to one part of a solid object must give rise to internal reaction forces that propagate from particle to particle throughout an extended part of the system. With very rare exceptions (such as ferromagnetic materials or planet-scale bodies), internal forces are due to very short range intermolecular interactions, and are therefore manifested as surface contact forces between adjacent particles — that is, as stress.
In mechanical and aerospace engineering, however, stress analysis must often be performed on parts that are far from equilibrium, such as vibrating plates or rapidly spinning wheels and axles. In those cases, the equations of motion must include terms that account for acceleration of the particles.
In structural design applications, one usually tries to ensure the stress is everywhere well below the yield strength of the material. In the case of dynamic loads, the material fatigue must also be taken into account. However, these concerns lie outside the scope of stress analysis proper, being covered in materials science under the names strength of materials, fatigue analysis, stress corrosion, creep modeling, and other.
Experimental methods 
Stress analysis can be performed experimentally by applying forces to a test element or structure and then determining the resulting stress using sensors. In this case the process would more properly be known as testing (destructive or non-destructive). Experimental methods may be used in cases where mathematical approaches are cumbersome or inaccurate. Special equipment appropriate to the experimental method is used to apply the static or dynamic loading.
There are a number of experimental methods which may be used:
- Tensile testing is a fundamental materials science test in which a sample is subjected to uniaxial tension until failure. The results from the test are commonly used to select a material for an application, for quality control, and to predict how a material will react under other types of forces. Properties that are directly measured via a tensile test are ultimate tensile strength, maximum elongation and reduction in area. From these measurements properties such as Young's modulus, Poisson's ratio, yield strength, and strain-hardening characteristics can be determined.
- Strain gauges can be used to experimentally determine the deformation of a physical part. A commonly used type of strain gauge is a thin flat resistor that is affixed to the surface of a part, and which measures the strain in a given direction. From the measurement of strain on a surface in three directions the stress state in the part can be calculated.
- Neutron diffraction is a technique that can be used to determine the subsurface strain in a part.
- The photoelastic method relies on the fact that some materials exhibit birefringence on the application of stress, and the magnitude of the refractive indices at each point in the material is directly related to the state of stress at that point. The stresses in a structure can be determined by making a model of it in a photoelastic material.
- Dynamic mechanical analysis is a technique used to study and characterize viscoelastic materials, particularly polymers. The viscoelastic property of polymer is studied by dynamic mechanical analysis where a sinusoidal force (stress) is applied to a material and the resulting displacement (strain) is measured. For a perfectly elastic solid, the resulting strain and the stress will be perfectly in phase. For a purely viscous fluid, there will be a 90 degree phase lag of strain with respect to stress. Viscoelastic polymers have the characteristics in between where some phase lag will occur during DMA tests.
Mathematical methods 
While experimental techniques are widely used, most stress analysis is done by mathematical methods, especially during design.
Differential formulation 
The basic stress analysis problem can be formulated by Euler's equations of motion for continuous bodies (which are consequences of Newton's laws for conservation of linear momentum and angular momentum) and the Euler-Cauchy stress principle, together with the appropriate constitutive equations.
These laws yield a system of partial differential equations involving the stress tensor field and the strain tensor field as unknown functions to be determined. Solving for either one will yield the other through the constitutive equations. Both fields will normally be continuous within each extended part of the system that does not contain any concentrated loads and that can be regarded as a continuous medium with smoothly varying constitutive equations.
The external body forces will appear as the independent ("right-hand side") term in the differential equations, while the concentrated forces appear as boundary conditions. An external surface force, such as ambient pressure or friction, can be incorporated as an imposed value of the stress tensor across that surface External forces that are specified as line loads (such as traction on a concrete block plastic applied by an embedded rebar) or point loads (such as the weight of a person standing on a roof) introduce singularities in the stress field, and may be handled by assuming that they are spread over a small arbitrary volume or surface patch. The basic stress analysis problem is therefore a boundary-value problem.
Elastic and linear cases 
A system is said to be elastic if any deformations caused by external forces will spontaneously and completely disappear once the external forces are removed. Stress analysis for such systems is based on the theory of elasticity and infinitesimal strain theory. When the applied loads cause permanent deformation, one must use more complicated constitutive equations, that can account for the physical processes involved (plastic flow, fracture, phase change, etc.)
Engineered structures are usually designed so that the maximum expected stresses are well within the range of linear elasticity (the generalization of Hooke’s law for continuous media); that is, the deformations caused by internal stresses are linearly related to them. In this case the differential equations that define the stress tensor are linear, too. Linear equations are much better understood than non-linear ones; for one thing, their solution (the stress at any given point) will also be a linear function of the external forces. For small enough stresses, even non-linear systems can usually be assumed to be linear.
Built-in stress 
The mathematical problem represented is typically ill-posed because it has an infinitude of solutions. In fact, in any three-dimensional solid body one may have infinitely many and infinitely complicated) non-zero stress tensor fields that are in stable equilibrium even in the absence of external forces.
Such built-in stress may occur due to many physical causes, either during manufacture (in processes like extrusion, casting or cold working), or after it (for example because of uneven heating, or changes in moisture content or chemical composition). However, if the system can be assumed to be linear, then the built-in stress can be ignored in the analysis, since it will merely add to the solution without interfering with it.
If linearity cannot be assumed, however, any built-in stress may affect the distribution of externally-induced stress (for example, by changing the effective stiffness of the material) or even cause the unexpected material failure. For these reasons, a number of techniques have been developed to avoid or reduce built-in stress, such as annealing of cold-worked glass and metal parts, expansion joints in buildings, and roller jointss for bridges.
Stress analysis is simplified when the physical dimensions and the distribution of loads allow the structure to be treated as one- or two-dimensional. In the analysis of trusses, for example, the stress field may be assumed to be uniform and uniaxial over each member. Then the differential equations reduce to a finite set of equations (usually linear) with finitely many unknowns.
If the stress distribution can be assumed to be uniform (or predictable, or unimportant) in one direction, then one may use a plane stress and plane strain formulation where the stress field is a function of two coordinates only, instead of three.
Even under the assumption of linear elasticity, the relation between the stress and strain tensors is generally expressed by a fourth-order stiffness tensor with 21 independent coefficients. This complexity may be required for general anisotropic materials, but for many common materials it can be simplified. For orthotropic materials like wood, whose stiffness is symmetric about three orthogonal planes, nine coefficients suffice to express the stress-strain relationship. For isotropic materials, these reduce to only two.
One may be able to determine a priori that, in some parts of the system, the stress will be of a certain type, such as uniaxial tension or compression, simple shear, isotropic compression or tension, torsion, bending, etc. In those parts the stress field may then be represented by fewer than six numbers, possibly just one.
Solving the equations 
In any case, for two- or there-dimensional domains one must solve a system of partial differential equations with boundary conditions. Analytical or closed-form solutions to the differential equations can be obtained when the geometry, constitutive relations, and boundary conditions are simple enough. Otherwise one must generally resort to numerical approximations such as the finite element method, the finite difference method, and the boundary element method.
Factor of safety 
The factor of safety (also "margin of safety" or "design factor") is a design requirement for the structure based on the uncertainty in loads, material strength, and consequences of failure. In design of structures, calculated stresses are restricted to be less than a specified allowable stress, also known as working, designed or limit stresses, that is chosen as some fraction of the yield strength or of the ultimate strength of the material which the structure is made of. The ratio of the ultimate stress to the allowable stress is defined as the factor of safety.
Laboratory test are usually performed on material samples in order to determine the yield strength and the ultimate strength that the material can withstand before failure. Often a separate factor of safety is applied to the yield strength and to the ultimate strength. The factor of safety on yield strength is to prevent detrimental deformations and the factor of safety on ultimate strength is to prevent collapse.
The factor of safety is used to calculate a maximum allowable stress:
Load transfer 
The evaluation of loads and stresses within structures is directed to finding the load transfer path. Loads will be transferred by physical contact between the various component parts and within structures. The load transfer may be identified visually or by simple logic for simple structures. For more complex structures more complex methods, such as theoretical solid mechanics or numerical methods may be required. Numerical methods include direct stiffness method which is also referred to as the finite element method.
The object is to determine the critical stresses in each part, and compare them to the strength of the material (see strength of materials).
For parts that have broken in service, a forensic engineering or failure analysis is performed to identify weakness, where broken parts are analysed for the cause or causes of failure. The method seeks to identify the weakest component in the load path. If this is the part which actually failed, then it may corroborate independent evidence of the failure. If not, then another explanation has to be sought, such as a defective part with a lower tensile strength than it should for example.
Uniaxial stress 
If two of the dimensions of the object are very large or very small compared to the others, the object may be modelled as one-dimensional. In this case the stress tensor has only one component and is indistinguishable from a scalar. One-dimensional objects include a piece of wire loaded at the ends and a metal sheet loaded on the face and viewed up close and through the cross section.
When a structural element is subjected to tension or compression its length will tend to elongate or shorten, and its cross-sectional area changes by an amount that depends on the Poisson's ratio of the material. In engineering applications, structural members experience small deformations and the reduction in cross-sectional area is very small and can be neglected, i.e., the cross-sectional area is assumed constant during deformation. For this case, the stress is called engineering stress or nominal stress. In some other cases, e.g., elastomers and plastic materials, the change in cross-sectional area is significant, and the stress must be calculated assuming the current cross-sectional area instead of the initial cross-sectional area. This is termed true stress and is expressed as
- is the nominal (engineering) strain, and
- is nominal (engineering) stress.
The relationship between true strain and engineering strain is given by
In uniaxial tension, true stress is then greater than nominal stress. The converse holds in compression.
Graphical representation of stress at a point 
Mohr's circle, Lame's stress ellipsoid (together with the stress director surface), and Cauchy's stress quadric are two-dimensional graphical representations of the state of stress at a point. They allow for the graphical determination of the magnitude of the stress tensor at a given point for all planes passing through that point. Mohr's circle is the most common graphical method.
Mohr's circle 
Mohr's circle, named after Christian Otto Mohr, is the locus of points that represent the state of stress on individual planes at all their orientations. The abscissa, , and ordinate, , of each point on the circle are the normal stress and shear stress components, respectively, acting on a particular cut plane with a unit vector with components .
Lame's stress ellipsoid 
The surface of the ellipsoid represents the locus of the endpoints of all stress vectors acting on all planes passing through a given point in the continuum body. In other words, the endpoints of all stress vectors at a given point in the continuum body lie on the stress ellipsoid surface, i.e., the radius-vector from the center of the ellipsoid, located at the material point in consideration, to a point on the surface of the ellipsoid is equal to the stress vector on some plane passing through the point. In two dimensions, the surface is represented by an ellipse (Figure coming).
Cauchy's stress quadric 
The Cauchy's stress quadric, also called the stress surface, is a surface of the second order that traces the variation of the normal stress vector as the orientation of the planes passing through a given point is changed.
Graphical representation of the stress field 
The complete state of stress in a body at a particular deformed configuration, i.e., at a particular time during the motion of the body, implies knowing the six independent components of the stress tensor , or the three principal stresses , at each material point in the body at that time. However, numerical analysis and analytical methods allow only for the calculation of the stress tensor at a certain number of discrete material points. To graphically represent in two dimensions this partial picture of the stress field different sets of contour lines can be used:
- Isobars are curves along which the principal stress, e.g., is constant.
- Isochromatics are curves along which the maximum shear stress is constant. This curves are directly determined using photoelasticity methods.
- Isopachs are curves along which the mean normal stress is constant
- Isostatics or stress trajectories are a system of curves which are at each material point tangent to the principal axes of stress.
- Isoclinics are curves on which the principal axes make a constant angle with a given fixed reference direction. These curves can also be obtained directly by photoelasticity methods.
- Slip lines are curves on which the shear stress is a maximum.
See also 
- Forensic engineering
- Rockwell scale
- Structural analysis
- Worst case circuit analysis
- List of finite element software packages
- Donald Ray Smith and Clifford Truesdell (1993) "An Introduction to Continuum Mechanics after Truesdell and Noll". Springer. ISBN 0-7923-2454-4
- I-Shih Liu (2002), "Continuum Mechanics". Springer ISBN 3-540-43019-9
- Fridtjov Irgens (2008), "Continuum Mechanics". Springer. ISBN 3-540-74297-2
- John Conrad Jaeger, N. G. W. Cook, and R. W. Zimmerman (2007), "Fundamentals of Rock Mechanics" (4th edition) Wiley-Blackwell. ISBN 0-632-05759-9