Guidelines for Nodes Selection in HAZOP Study

Node selection is a critical early step in a HAZOP study because it directly determines study duration, schedule accuracy, and workshop effectiveness. The number, size, and complexity of nodes provide the most reliable basis for estimating the effort required and strongly influence how smoothly the study is facilitated.

A node is a discrete part of a process or system with a defined function and clear boundaries, selected for detailed HAZOP analysis. A node may be a physical component (vessel, pipe, valve, pump) or a logical element (control loop, process step, phase change, or operating condition).

Why Node Selection Matters?

Node selection is important for two key reasons:

It is the most accurate method for estimating total HAZOP study time
Node size and complexity directly affect facilitation quality, study speed, and hazard identification effectiveness

Practical Guidelines for Defining HAZOP Nodes

1. Node Selection for Continuous Processes

For continuous processes, nodes are typically selected by following the direction of flow on the Process Flow Diagram (PFD). Major vessels are treated as individual nodes, while connecting process lines—including minor equipment such as pumps—are grouped accordingly.

2. Defining Node Boundaries in Continuous Processes

Node boundaries are commonly defined at:

A change in design intent
A phase or composition change
A major equipment item

To avoid ambiguity (e.g., when applying “more flow”), nodes should be defined so deviations apply uniformly across them. Single-line nodes are often preferred, even if this increases the total node count.

3. Node Selection for Complex Systems

Complex systems such as flare systems, drainage, fuel gas, inert gas, instrument air, cooling water, firewater, and chemical dosing networks span multiple P&IDs and connections and must be divided into multiple nodes.

4. Interfaces Between Process and Complex Systems

Connections into complex systems should first be reviewed within their respective process nodes, with node boundaries typically ending at the upstream isolation valve.

5. Internal Node Structure of Complex Systems

The complex system itself should be divided into logical nodes based on:

Headers and manifolds
Major equipment

For collection systems, separate nodes are defined for each header up to the collection vessel, with knock-out drums and flare stacks treated as individual nodes. Supply systems are divided into interface nodes, header/manifold nodes, return system nodes (where applicable), and equipment nodes such as compressors, receivers, or dosing skids.

6. Familiarization Before Node Definition

Before identifying nodes, the HAZOP leader should thoroughly understand the process using available information such as Process Flow Diagrams (PFDs), Piping and Instrumentation Diagrams (P&IDs), layout drawings, operating manuals, and expert knowledge. Adequate familiarization ensures technically sound and practical node selection.

7. Initial Node Identification Using PFDs

Nodes should initially be identified on the PFD, which is a highly effective tool for defining logical and functional sections of the process, as well as system interfaces. Node selection typically progresses in the direction of process flow, using the PFD for overall guidance.

8. Functional Basis for Node Selection

Nodes may be physical (vessels, pipes, valves, pumps) or logical (control loops, process steps, phase changes), but should always be selected based on function (e.g. transfer, heat, cool, react) so that the design intent of each node is clear and consistently understood by the team.

9. Defining Node Boundaries

Once nodes are identified, their boundaries must be clearly defined. Boundaries may be physical (pipes, flanges, fittings) or conceptual (design limits, operating ranges, setpoints). Clear boundaries help isolate each node and define its inputs and outputs for focused analysis.

10. Marking Nodes on P&IDs

After defining nodes on the PFD, they should be color-marked on the P&IDs, with different colors used to distinguish different nodes. Marked-up P&IDs should be shared with the HAZOP team before the workshop to support preparation and efficient discussions.

11. Use of Compound Nodes

In practice, compound nodes are commonly used. For example, feed piping from a vessel, a pump, a control valve set, and a heat exchanger supplying a reactor may be treated as a single node if all elements share a common function.

12. Optimizing Node Size

With increasing experience and confidence in the HAZOP methodology, node size may be expanded. An optimum node may include multiple items of equipment, provided they perform a common function and deviations can be applied uniformly.

13. Typical Node Progression Along the Process

Node selection generally follows the process flow:

Start with feed lines entering the P&ID (without including vessels)
Define the first major vessel as a separate node
Extend outlet lines from the vessel until the next major vessel is reached

14. Factors Influencing Node Size and Complexity

The size and complexity of a node depend on several factors, including the experience of the HAZOP leader and team, the severity of process hazards, and the complexity of the control system.

15. Avoiding Overly Small or Large Nodes

Very small nodes increase study time due to repetition of guide word applications, while overly large nodes can create confusion and increase the risk of missed hazards. As a rule of thumb, limit each node to a maximum of five valves or pumps (main and standby pumps counted as one).

16. Use of Super Nodes

Some organizations use “super nodes” that combine multiple lines and major vessels to accelerate studies. While this provides a broader overview, it complicates analysis and increases the risk of missing credible scenarios, so it should be applied with caution.

17. Global Nodes

A global node is often included at the end of the HAZOP to address initiating events that can affect multiple nodes, such as external events (e.g. flooding) or total loss of utilities (e.g. electrical power failure).

18. Responsibility for Node Selection

Nodes should be selected by the HAZOP leader, with input from team members. For greenfield projects, the leader may define nodes independently based on experience, while for brownfield projects, node selection should be done in consultation with the process design engineer.

19. Node Selection Around Major Vessels

Major vessels with level control (such as towers, drums, and tanks) are often treated as separate nodes. Typically, the vessel and its bottom system are one node, while side draws and overhead systems are separate nodes, unless no major downstream equipment exists.

20. Nodes with Multiple Operating Modes

If a node has more than one operating mode (e.g. normal operation and regeneration), each operating mode should be treated as a separate HAZOP node due to differences in hazards and deviations.

21. Parallel Trains

Parallel process trains may be reviewed independently or “by difference.” When using a by-difference approach, the trains must be carefully compared to ensure all differences in design, control, instrumentation, and layout are identified and assessed.

22. Study Boundaries and Interfaces

Upstream and downstream operations outside the formal study scope should not be ignored. Deviations may be applied beyond node boundaries, and any credible downstream impacts should be captured, with recommendations raised for follow-up with external parties where required.

23. Node Selection for Batch and Procedural HAZOP

For batch and procedural HAZOPs, nodes should represent major process or procedural steps rather than individual actions. Grouping steps (e.g. charging, reaction, cooling, transfer) avoids repetition. Hierarchical task analysis helps identify safety-critical steps so HAZOP effort is focused where risk is highest.