An Overview of Safety Factors in the Engineering Design Process
Posted on February 6, 2018
Many considerations are included in a valid engineering design process. Typically decisions that provide for safe service and expected reliability in the given component or structure are the most critical. Determining the numerical value of an appropriate safety factor (or factor of safety) during design is vital to achieving mechanical or structural integrity.
Incorrect specification of the safety factor in a given application can produce premature failures. Consequences then may include loss of reputation or market status, costly property losses, human injuries or even fatalities. The responsible engineer or his/her employer may be found to be legally liable for the failure.
A safety factor (SF) is the numerical ratio of the maximum mechanical stress-carrying capability of the given material used divided by the stress value employed to specify critical dimensions, e.g., thickness, of the product or structure. The latter value is often known as the working stress (also known as allowable stress). For ductile materials, the safety factor is specifically defined as follows:
SF = nominal yield strength of the material / working stress used to design the product,
Brittle materials, such as gray cast iron, have no clearly defined yield strength and in those cases, the ultimate tensile strength (UTS) of the material is used in the numerator of the SF ratio instead of yield strength. However, UTS values for this material are very different for tension versus compression loading. Therefore this material characteristic must be considered in establishing a SF.
Safety factors must have values greater than 1.0. In this way the largest working stress in the final produce is never more than the yield (or ultimate) strength of the material used. SF values of 2.0 to 3.0 are common but they may vary from just above 1.0 to more than 4.0 depending on application variables. The particular SF used during design often is dictated by a recognized code for the given product or type of structure. For example, SF’s for design and/or working stresses are defined in the American Institute for Steel Construction (AISC) “Steel Construction Manual” for structures and in the American Society of Mechanical Engineers (ASME) “Boiler and Pressure Vessel Code”. In many situations the specific product is not clearly covered by a code or standard. In that case the choice of a suitable SF is a design decision left up to the engineer.
A safety factor is necessary because multiple uncertainties can exist. Some common examples are as follows:
- Maximum service loads and how they are applied to the part or structure may not be known exactly,
- Unexpected service conditions such high temperature, corrosion, wear or fatigue loading may be encountered in service. Each of these influences can accelerate failure,
- Materials used are generally assumed to be homogenous and free of internal defects. These characteristics may not be true and consequently the normal stress-carrying capabilities of the material can be reduced,
- The product manufacturing, fabrication or installation process may introduce detrimental features. Some examples are surface defects, poorly made welds and residual stresses left in hte material. These former two results can create stress concentration points that lower fatigue resistance. Residual stresses in the material are additive to the normal working stress level that was assumed,
- Competent, regular in-service inspection and maintenance may have been assumed during design of the product. The reality is that these failure preventative measures are not always accomplished.
Like many engineering design decisions, the choice of the value of the SF (if not defined by a relevant code) becomes a trade-off. For example, a higher strength material or thicker cross sections might be used, more thorough manufacturing quality control and inspection could be specified or more thorough design analysis via finite element analysis (FEA) might be justified. However each of these actions will likely add to the cost, size, operating weight or time-to-market of the final mechanical product or structure. The consequences of failure in each application must dictate the proper mix of safety factor choice and related actions.