What is HAZOP? The hazard and operability study explained

HAZOP is short for hazard and operability study. IEC 61882 describes it as a structured and systematic examination of a defined system with two aims: identifying risks to people, equipment and the environment in operation and maintenance, and identifying operability problems that can cause disturbances or lost production. Hence the pair of words: hazard and operability.

Its defining feature is the examination session: led by a study leader (facilitator), a multidisciplinary team goes through the design part by part and asks, prompted by guide words, what happens if a process parameter deviates from its intended value. IEC 61882 treats HAZOP as an enhancement to sound design based on codes and experience – not as a substitute for them.

In German-speaking countries the same method is also known as the PAAG method (Prognose von Abweichungen, Auffinden der Ursachen, Abschätzen der Auswirkungen, Gegenmaßnahmen).

HAZOP was developed at the British chemical company ICI in the 1960s; Trevor Kletz dates it to 1963. The first paper appeared in 1974, and in 1977 the Chemical Industries Association published a guide. Kletz did not invent the method, but he became its best-known advocate.

The international standard is IEC 61882, “Hazard and operability studies (HAZOP studies) – Application guide”. Edition 1 appeared in 2001, the current Edition 2.0 in 2016. It is an application guide: it describes good practice rather than creating a legal obligation.

1. Definition: initiate the study, define scope and objectives, assign roles and responsibilities.
2. Preparation: plan the study, collect documents (P&IDs, process description, substance data, design data), and agree the guide words and deviations to be examined.
3. Examination: in team sessions, examine the nodes one after another.
4. Documentation and follow-up: record the results, sign off the report, track recommendations and responsibilities.

Examining one node

1. The design engineer presents the section on the P&ID.
2. For the selected node, the purpose, operating conditions and instrumentation are explained – the design intent.
3. The team combines parameters and guide words into deviations, such as “more flow”.
4. For each credible deviation: identify causes, assess consequences without safeguards, evaluate existing safeguards, and record a recommendation with a responsible person where needed.
5. When all guide words are done, the team moves to the next node.

The system is divided into sections that are examined in turn. In practice they are called nodes; IEC 61882 speaks of “parts”. A node can be physical, such as a line between two major equipment items, or logical, such as one step of a batch procedure. The more complex the system and the more severe the potential consequences, the smaller the nodes should be.

Each node has a design intent: what it is supposed to do, with its operating values – substance, flow, temperature, pressure, timing, sequence, location and direction. A precise design intent is the basis for everything that follows, because only a clearly stated intent can be tested for deviations.

Guide words structure the search for deviations. Each guide word is applied to the relevant parameters of a node; each combination is a potential deviation. The basic set in IEC 61882:

Further guide words may be used if they are defined before the study starts. A full reference with parameters and example deviations is in our guide to HAZOP guide words.

HAZOP is teamwork. Typical roles are the study leader (facilitator or chair), the recorder (scribe), design and process engineers, operations, maintenance, instrument and control engineers and safety specialists, with other experts or equipment suppliers as needed. Guidance puts the team at three to eight people, most often five to seven.

The facilitator should ideally be independent of the design team, so that assumptions are challenged without bias. The plant knowledge, however, has to come from the operator’s own people. More on the role in our guide to the HAZOP facilitator.

Results are recorded in a worksheet. Common columns are node and design intent, parameter and guide word, deviation, cause, consequence, existing safeguards, recommendation, owner and due date. Many companies add risk ranking before and after safeguards using their own risk matrix.

While tracing causes and consequences, existing safeguards are deliberately set aside and only then assessed for effectiveness – so the unmitigated risk stays visible. Safeguards should be recorded even where no action is needed; otherwise they may later be removed out of ignorance. A free Excel worksheet with these columns is available as our HAZOP template, and worked rows for six process units are in HAZOP examples.

Effort depends on plant size, number of nodes, operating modes and preparation. Guideline figures from the Center for Chemical Process Safety (CCPS), as quoted by the US EPA:

Because concentration drops over long sessions, guidance recommends shorter session days, for example half days. In one published example of a large offshore facility, the team averaged one to two nodes per day. The biggest cost driver is the team’s session time.

Regulations usually require a systematic hazard analysis rather than HAZOP by name – and HAZOP is one of the most widely accepted methods for it:

• United States: OSHA’s process safety management standard (29 CFR 1910.119) requires a process hazard analysis for covered processes and lists HAZOP among the accepted methods, alongside What-If, checklists, FMEA and fault tree analysis. The analysis must be updated and revalidated at least every five years. The EPA Risk Management Program mirrors these requirements.
• European Union: under the Seveso III Directive, upper-tier establishments must demonstrate in their safety report that major-accident hazards have been identified. In Germany, authorities accept PAAG/HAZOP as a method for the systematic hazard identification the Störfall-Verordnung requires, without prescribing it.
• United Kingdom: COMAH 2015 requires major accident hazards to be identified; HSE guidance uses HAZOP as an example method but the regulations do not name it.
• Functional safety: HAZOP results commonly feed LOPA and the determination of safety integrity levels (SIL) under IEC 61511.

The most common comparison is covered in detail in HAZOP vs HAZID.

Many studies are still recorded in spreadsheets. Software helps once several editors, P&ID revisions and action tracking come together. AI can take over preparation work – capturing P&ID components and drafting deviations per node – while the risk judgement stays with the team. See AI HAZOP for what works, what research shows and where the limits are, and the HAIZOP platform for the feature set.