Loading...

Please Wait...

 

The hazard and operability (HAZOP) study is the most commonly used process hazard analysis (PHA) method (for details, click here). However, there are many other PHA methods available which may be more suitable depending on the circumstances.

This page describes a variety of PHA methods and provides a comparison of their advantages and disadvantages, Comparison of PHA Methods.

These methods are described:

Preliminary Hazard Analysis (PrHA)  

PrHA identifies the hazards of a process and the hazardous situations they may produce. Possible causes, consequences and recommendations for protective measures are addressed. A criticality ranking may be assigned and used to prioritize protective measures.

Typically, PrHA is used to evaluate and prioritize hazards early in the life of a process as a precursor to more detailed hazard analysis studies. Generally, it is applied during conceptual design or at the R&D stage when there is little information available on design details or operating procedures. Commonly, it is used as a design review tool before a P&ID is developed. It is useful in making site selection decisions and in analyzing large facilities when circumstances prevent other techniques from being used.

Checklist

A checklist used as a hazard evaluation procedure employs prepared lists of questions relating to process safety to identify concerns and prompt the analysts to determine whether existing safeguards are adequate. Checklists are used to identify common hazards and ensure compliance with procedures, codes of practice, regulations, etc. Checklist questions are based on experience and knowledge of safety issues for the process and applicable codes, standards and regulations.

Checklists can be applied to virtually any aspect of a process such as equipment, materials, procedures, etc. Their application requires knowledge of the process and its procedures and an understanding of the meaning of the checklist questions. Checklists may become outdated and they should be audited and updated regularly.

What-If (WI) and What-If Checklist (WIC)

WI studies involve posing questions relating to initiating events to identify hazard scenarios for a process. The PHA team brainstorms questions in a WI study. The team starts with a prepared list of questions in a WIC study, although almost always additional questions are added as a study proceeds. Sometimes PHA teams develop questions based on the HAZOP thought process by thinking through what questions would arise if a HAZOP study were being performed.

WI methods are well-suited to examining the impacts of proposed changes in Management of Change (MOC) PHA studies because the questions can be tailored to the change and the areas affected by it. They can be used to study virtually any aspect of a process such as equipment, procedures, control systems, management practices, etc. Team leaders should be experienced with the technique since it is provides less structure than other PHA methods.

Hazard and Operability (HAZOP) Study

The HAZOP method is used to identify hazard scenarios with impacts on people and the environment as well as operability scenarios where the concern is the capacity of the process to function. Originally, it was developed for fluid processes but it has also been applied to non-fluid systems such as materials handling, drilling operations, aerospace systems, etc. Currently, it is the most commonly used technique in the process industries.

The HAZOP method focuses on investigating deviations from design intent such as “no flow” at a location in the process where flow is intended or “high pressure” in a vessel which should not exceed a pressure limit. By definition, deviations are potential problems, e.g., no flow in a transfer line or overpressuring a vessel. Deviations from design intent are generated by applying guide words to process parameters at different locations (nodes) throughout the process, e.g., for an inlet line to a vessel, No + Flow = No Flow, or for a vessel, High + Pressure = High Pressure.

A standard list of seven guide words is used: No, More, Less, As Well As, Part Of, Reverse, and Other Than. The team chooses appropriate parameters for each node, e.g., flow, pressure, temperature, composition, level, addition, cooling, location, etc. The use of guide words with parameters provides the opportunity to explore deviations from design intent in every conceivable way thus helping to ensure completeness of the PHA study.

Failure Modes and Effects Analysis (FMEA)

FMEA is a hazard evaluation procedure in which failure modes of system components, typically, process equipment, are considered to determine whether existing safeguards are adequate. Failure modes describe how components fail (e.g., open, closed, on, off, leaks, etc.). The effects of each failure mode are the process responses or incident resulting from the component failures, i.e., hazard scenario consequences. A FMEA becomes a FMECA (Failure Modes and Effects and Criticality Analysis) when a criticality ranking is included for each failure mode and effect. A criticality ranking is the same as a risk ranking.

FMEA is used extensively in the aerospace, nuclear, and defense industries. Typically, it is used in the process industries for special applications such as Reliability Centered Maintenance (RCM) programs and the analysis of control systems.

FMEA can be conducted at different levels of resolution. For PHA purposes, usually it is conducted at the equipment level, e.g., valves, pumps, lines, etc. For RCM purposes, usually it is conducted at the equipment component level, e.g., motor, shaft, impeller, casing, seal, bearings, etc. for a pump.

Major Hazard Analysis (MHA)

MHA was developed specifically to support process safety studies [A1, A2]. It is used to identify major hazard scenarios involving fires, explosions, toxic releases and reactivity excursions. MHA employs a structured approach to identify loss of containment scenarios. Causes of loss of containment can be direct, e.g., valves left open or ruptures in lines or vessels, or indirect, e.g., runaway reactions resulting in releases through pressure relief devices or vessel and piping rupture. MHA constrains brainstorming to such scenarios within a structured framework to guide the identification of initiating events using standard checklists. Brainstorming focuses on specific categories of initiating events to focus the team’s brainstorming without narrowing their vision. The checklists provide guidance to the team and help assure completeness. They can be customized for specific facilities or types of processes. The method prompts consideration of items not already in the checklists. MHA uses a process subdivision similar to other PHA methods.

Process Hazard Review (PHR)

PHR was developed for use with operating plants as an alternative to HAZOP [A3]. It addresses major hazards. There are variants that address other types of hazards and environmental releases. It is based on the premise that most major hazard process incidents involve loss of containment. PHR uses prompts covering the range of mechanisms for loss of containment to identify hazard scenarios. The method has been extended to address other hazard types (Operational Hazard Review) and environmental releases (Environmental Hazard Review).

Fault Tree Analysis (FTA)

FTA is not really comparable to standard PHA methods. It does not identify a full set of hazard scenarios for a process. Rather, it is used to identify the causes of a particular incident (called a top event) using deductive reasoning. Often, it is used when other PHA techniques indicate that a particular type of accident is of special concern and a more thorough understanding of its causes is needed. Thus, it is a useful supplement to other PHA techniques. Sometimes FTA is used in the investigation of incidents to deconstruct what happened. FTA is also used to quantify the likelihood of the top event. It is best suited for the analysis of highly redundant systems.

FTA identifies and graphically displays the combinations of equipment failures, human failures and external events that can result in an incident. Computer programs are used to provide graphical representations of fault trees and to calculate top event likelihoods. FTA is not a technique that lends itself to a team-based study. Typically, one or two people construct a fault tree. It requires different training and resources than other PHA techniques.

Event Tree Analysis (ETA)

ETA is not really comparable to standard PHA methods. It does not identify a full set of hazard scenarios for a process. Rather it is used to identify the possible outcomes following the success or failure of protective systems after the occurrence of a given starting event and, optionally, to calculate the frequencies of the outcomes. Event trees graphically display the progression of event sequences beginning with a starting event, proceeding to control and safety system responses, and ending with the event sequence consequences.

ETA helps analysts to determine where additional safety functions will be most effective in protecting against the event sequences. Typically, ETA is used to analyze complex processes that have several layers of safety systems or emergency procedures to respond to starting events. ETA is not a technique that lends itself to a team-based study. Typically, one or two people construct an event tree.

Cause-Consequence Analysis (CCA)

CCA is a blend of fault tree analysis and event tree analysis that produces a CCA diagram combining fault and event trees. It is used to identify causes and consequences of hazard scenarios. The CCA diagram displays the relationships between the incident outcomes (consequences) and their causes and it can depict and evaluate multiple scenario outcomes, including recovery paths where the operator, or system, recovers or mitigates the consequences, as well as the worst consequence path. CCA is commonly used when the failure logic of hazard scenarios is simple.

Bow-Tie Analysis (BTA)

BTA is a less formal variation of Cause-Consequence Analysis. It uses a combination of high-level fault and event trees to produce a diagram resembling a bow tie. Hazards and initiating events appear on the pre-event side (left side) and impacts (consequences) appear on the post-event side (right side). The focal point of the diagram is the specific loss event that ties together the initiating events and consequences. There is a time progression from the left to the right of the diagram. Associated prevention and mitigation safeguards are shown on either side of the loss event and they are viewed as barriers, some of which may apply to more than one cause.

BTA is used for screening hazards of well-understood processes and to perform an initial analysis for existing processes or in the middle stages of process design.

Comparison of PHA Methods   

Method Advantages Disadvantages
PrHA Easy to understand

Fast to perform
Requires careful judgment

Not a detailed PHA Method
Checklist Easy to use and provides results quickly

Level of detail can be varied

Communicate information well

Effective way to take  advantage of lessons learned
Does not help in identifying new, or unrecognized hazards

May overlook unusual hazards or novel elements of a process

No cause and effect analysis 

Usually, requires some subjective interpretation 

Limited to the experience of the author

Repetitive nature can lead to errors

May not apply to the particular situation 

Provides a minimum level of hazard evaluation 
WI and WIC Easily understood

Flexible

Less effort / time

Can help to identify scenarios that involve interactions between different parts of the process
Loose structure

Results particularly dependent on the skill, experience and thoroughness of user  

No assurance that the breadth or depth of the questions considered is adequate  


HAZOP Viewed as the most effective of traditional PHA methods 

Provides assurance that hazard scenarios have been identified

Addresses both safety and operability 

Difficult to exclude operability scenarios 

Difficult to consider all aspects of intention in a reasonable time period   

Effort involved can be significant 

Focuses on individual nodes and may miss some hazard scenarios that involve interactions between nodes 

FMEA Systematic, element-by-element procedure that helps ensure completeness 

Easily understood and used by engineers 

Easily updated for design changes or facility modifications  

Not efficient for identifying combinations of equipment failures 

Human failures are not generally examined although the effects of misoperation can be described by an equipment failure mode or by the causes of a failure    

External events are not easily addressed   

MHA Focuses exclusively on hazard scenarios, i.e., does not address operability scenarios 

Time required is substantially less than traditional methods

Structured approach 

Readily understood by PHA teams

All hazard scenarios for a node appear in a single worksheet

Current PHA studies can be converted easily into MHA format 
Does not address operability scenarios   


PHR Structured method 

Quickly identifies and assesses major hazard scenarios 

Operations personnel can share their experience effectively  

Focuses more on what team members know, not what they don’t know 

Generates more general recommendations rather than specific ones  

Proprietary method   

FTA Thorough and systematic

Can be time consuming 

Binary representation of faults (either success or failure, no partial failures) 
ETA Easy to understand

Can be time consuming

Binary representation of failures (either success or failure, no partial failures) 
CCA Provides a detailed graphical depiction of hazard scenarios CCA diagram can become complex 
BTA Easy to understand  Provides only a simple analysis

Does not provide a formal way to identify loss events 

Can become complex for larger processes