Hazard Analysis
What Is Hazard Analysis?
Hazard analysis is a structured discipline concerned with the systematic identification, assessment, and prioritization of hazards in engineered systems, processes, or products. Its central purpose is to support decisions about risk reduction before accidents occur, rather than after. The discipline draws from reliability engineering, systems theory, occupational safety, and chemical process engineering, and its formal methods are embedded in international standards governing industries from aviation and automotive to nuclear power and pharmaceuticals.
Hazard analysis distinguishes between the hazard itself, which is a potential source of harm, and risk, which combines the probability of that hazard manifesting with the severity of the consequences. A thorough analysis characterizes both dimensions, enabling engineers and safety managers to allocate resources to the hazards that pose the greatest expected loss.
Systematic Analysis Methods
Several structured methods are in widespread professional use. Failure Modes and Effects Analysis (FMEA) works forward from component-level failures to identify how each failure mode propagates through a system and what harm it could cause. It is particularly suited to hardware-intensive products and is required by standards such as IEC 61508 for functional safety assessments. Fault Tree Analysis (FTA) works in reverse: starting from an undesired top-level event, it decomposes contributing causes using Boolean logic until individual component failures or human errors are reached.
Hazard and Operability Studies (HAZOP) were developed for continuous-process industries such as petrochemicals and apply a guide-word approach, examining each parameter of a process design with deviations such as "more," "less," "reverse," and "other than" to surface hazard scenarios that designers may not anticipate. The ACM Digital Library contains extensive literature on applying these hazard analysis methods to software-intensive systems, where causal chains are often harder to enumerate than in purely physical systems.
Safety Integrity and Risk Criteria
Hazard analysis feeds directly into safety integrity assessments, which determine how much risk reduction a safety function must provide. The IEC 61508 functional safety standard defines four Safety Integrity Levels (SIL 1 through SIL 4), with each level corresponding to a target range for the probability of dangerous failure per hour. Hazard analysis provides the input data: once a hazard scenario is characterized by its severity and tolerable risk, the required SIL follows from the gap between that tolerable level and the residual risk without the safety function.
In automotive systems, the equivalent framework is the Automotive Safety Integrity Level (ASIL), defined by ISO 26262, which applies hazard analysis specifically to road vehicles and adds an exposure factor to account for how often a vehicle operates in conditions where the hazard is relevant.
Software and Systems Hazard Analysis
Software hazard analysis has grown in importance as the proportion of safety-critical functionality implemented in software has increased. Dedicated techniques such as Software Fault Tree Analysis (SFTA) and System-Theoretic Process Analysis (STPA), developed at MIT's Partnership for a Systems Approach to Safety, extend traditional hardware-centric methods to cover software control logic and system-level emergent behaviors. The hazard function, a statistical measure of instantaneous failure rate over time, is commonly employed in reliability-linked hazard assessments when time-to-failure data are available.
Applications
Hazard analysis has applications in a wide range of regulated and safety-critical industries, including:
- Aerospace and aviation system certification
- Automotive safety under ISO 26262 and related standards
- Nuclear power plant safety case development
- Medical device risk management per ISO 14971
- Chemical and process plant safety under ATEX and IEC standards
- Product safety assessment in consumer electronics manufacturing
- Six Sigma defect-reduction programs in high-volume production