Consequence Modelling Inputs And Assumptions are the quiet levers behind every credible risk contour, thermal radiation map, and toxic dispersion footprint that HSE and process safety teams rely on. When a plot plan changes, when a new storage tank is added, or when ADNOC Code of Practice updates reshape expectations for Quantitative Risk Assessment (QRA), the numbers and choices feeding a model determine whether the results are decision-grade or misleading. This article unpacks what those inputs and assumptions are, why they matter in the UAE’s regulatory and climatic context, and how to treat them with discipline so that results earn stakeholder confidence from the control room to the boardroom.
Rather than cataloging software outputs, we focus on the logic behind the data: how inventory and thermophysical properties drive source terms; how meteorology and terrain shape dispersion; how ignition probability affects jet fire and vapor cloud explosion outcomes; and how equipment geometry and elevations alter impingement, escalation, and potential domino effects. Along the way, we relate the discussion to methods like HAZID, HAZOP, LOPA, Bowtie, and SIL allocation under IEC 61511, and to common UAE regulatory touchpoints with OSHAD, Dubai Municipality, and Civil Defence approvals. The aim is simple: help you scrutinize the modelling backbone with the same rigor you bring to plant modifications, emergency response planning, and assurance against major accident hazards.
Who Does This Affect?

While consequence modeling is often associated with QRA specialists, its inputs and assumptions ripple across disciplines and decisions. In practice, the following teams are directly impacted:
- HSE managers and process safety engineers stewarding Major Accident Hazard (MAH) frameworks, Bowtie barriers, and ALARP demonstrations for refineries, gas processing plants, petrochemical complexes, and LNG/LPG terminals.
- Plant managers and operations leaders responsible for safe, reliable production at onshore and offshore facilities, tank farms, truck loading gantries, and pipeline networks.
- Design and projects teams shaping plot plans, siting decisions, and firewater system layouts for new units, debottlenecking, or brownfield tie-ins.
- Maintenance, inspection, and integrity teams whose RBI, SIS testing intervals, and isolation strategies influence release frequency assumptions and barrier availability.
- Emergency response and business continuity coordinators using thermal radiation, overpressure, and toxic concentration contours to define muster points, shelter-in-place criteria, and evacuation triggers.
- Compliance officers navigating ADNOC requirements, OSHAD-SF obligations, and Dubai Civil Defence permitting for hazardous materials storage, transfer, and processing.
Suppliers, EPCs, and even insurers also engage with models, scrutinizing underlying assumptions when reviewing facility siting or domino-effect controls. In short, anyone whose decision relies on magnitude, footprint, or likelihood of hazardous effects has a stake in what goes into—and is assumed within—these models.
Why This Topic Matters

Because models condense complex physics into tractable scenarios, they are only as strong as the data and decisions entered at the start. Underestimation of inventory or mischaracterization of weather stability can swing results by orders of magnitude. Overly optimistic ignition assumptions may mask flash fire or vapor cloud explosion potential in congested pipe racks. Under-recognized confinement factors can downgrade overpressure predictions in compressor buildings or enclosed process areas. And conservative but unrealistic assumptions can distort CAPEX and OPEX by over-driving fireproofing, spacing, or passive fire protection.
In the UAE, the stakes are heightened by climatic and regulatory realities. High ambient temperatures, humidity, and frequent stable night-time conditions promote dense-gas behavior for many releases. Coastal and island terrain can create sea-breeze loops affecting dispersion and backflow. ADNOC Codes of Practice and regulator expectations for ALARP require transparent, defendable input selection and sensitivity checks. Civil Defence approvals and municipal planning around industrial zones rely on credible impact distances when setting land-use buffers and emergency response arrangements. Getting inputs and assumptions right is not academic; it influences siting, safeguards, operations, and business continuity.
Key Points to Understand

Building on the “why,” let’s examine the building blocks that shape consequence predictions and how to handle them rigorously.
1) Source Terms: Inventory, Properties, and Process Conditions

Everything starts with the source term—the mass or energy rate entering the environment after a loss of containment. It depends on:
- Fluid properties: molecular weight, vapor pressure, heat of vaporization, density, viscosity, flammability limits, and toxicity thresholds (e.g., AEGLs, ERPGs).
- Process conditions: operating pressure and temperature, phase, and upstream/downstream equipment that may act as flow restrictions.
- Release geometry: hole size distributions, nozzle orientation, elevation, and equipment internals impacting choked versus subsonic flow.
Assumptions often enter when plant records vary, when summer versus winter operations differ, or when upset conditions are plausible. It is prudent to anchor values in engineering data (P&IDs, PFDs, heat and material balances, historian trends) and test key variations in sensitivity runs.
2) Release Scenarios: Frequencies and Sizes

Consequence calculations typically pair with frequency analysis, but even for pure consequence studies (no frequencies), the range of hole sizes and event types chosen drives outcomes: small continuous leaks, catastrophic rupture, pump seal failures, PSV lift to atmosphere, or hose failures at loading stations. These choices should correlate with inspection intervals, design codes, and integrity data. A common pitfall is selecting textbook hole sizes without reflecting the facility’s actual equipment population and damage mechanisms.
3) Dispersion: Meteorology, Terrain, and Heavy-Gas Behavior

UAE meteorology needs careful treatment. Prevailing winds, diurnal sea-breeze cycles, and stable night-time conditions (Pasquill F or E) can elongate toxic plumes. High humidity and heat can alter buoyancy for hot vapors and flare gases. Terrain matters too: berms, buildings, dunes, or coastal topography can channel flow, promote recirculation, or create shadow zones. For dense or heavy-gas releases (common with LPG, ammonia, or cooled hydrocarbons), using a dense-gas capable model and verifying stability class definitions becomes essential.
4) Ignition and Fire/Explosion Behavior

Ignition probability is a prime assumption, and it should account for congestion, confinement, and ignition source density (motors, hot surfaces, electrical equipment) in the affected area. Jet fires are driven by momentum and flame length; pool fires by spreading rates and burning rates; flash fires by flammable cloud size; and vapor cloud explosions by congestion and confinement indices. Overpressure damage thresholds and thermal radiation criteria for people, equipment, and structures should match recognized guidance and internal corporate standards, ensuring alignment with ALARP arguments.
5) Safeguards and Barriers: Consistency with LOPA and IEC 61511

When crediting safeguards, align with your LOPA and Safety Instrumented System (SIS) design basis. Availability, proof-test intervals, and common cause failures should be consistent and not double-counted. Physical barriers (dikes, curbing), detection and isolation response times, deluge effectiveness, and fireproofing durations must be realistic and evidence-backed. Keep credits traceable to maintenance and operations performance indicators.
6) Receptors and Occupancy
Personnel distribution, occupancy hours, and sheltering capabilities define the difference between an equipment damage case and a Potential Loss of Life case. Consider contractor presence during turnaround, night shifts with reduced staffing, and public receptors at fence lines or along shared corridors. Protective measures—refuge rooms, breathing apparatus, and evacuation routes—should be represented faithfully.
7) Documentation and Traceability
Transparent traceability of inputs and assumptions is the most reliable way to earn stakeholder trust and expedite regulator queries. Maintain a clear basis of calculation, version control, scenario register, and sensitivity record. Where suitable, organize your inputs in a centrally managed repository such as engineering document control or a curated knowledge base like CABINET NPM so teams can consistently retrieve the same datasets.
| Input/Assumption Category | Typical Data Sources | Impact of Uncertainty |
|---|---|---|
| Inventory and fluid properties | P&IDs, H&MBs, vendor datasheets, MSDS | Directly affects source term and toxic/flammable footprint magnitude |
| Hole sizes and release types | Integrity records, RBI, failure databases, design codes | Changes event severity bands and escalation potential |
| Meteorological data | Site met masts, regional stations, long-term wind roses | Shifts downwind distances and frequency-weighted contours |
| Terrain and congestion | 3D models, drone surveys, plot plans | Drives recirculation, flame acceleration, and overpressure |
| Ignition probability | Area classification, equipment density, historical incidents | Alters split among jet fire, flash fire, and VCE outcomes |
| Safeguard credits | LOPA, IEC 61511 SIL verifications, test records | Modifies duration, escalation, and vulnerability windows |
| Receptors and occupancy | HR rosters, contractor plans, shift schedules | Influences risk to life and emergency planning basis |
How This Works in Practice
Turning principles into robust modeling begins with data fidelity. Teams typically consolidate P&IDs, UFDs, heat and material balances, and recent operating envelopes from the DCS or historian to capture credible pressures, temperatures, and compositions. Integrity and maintenance records help anchor likely failure modes and event sizes—pump seal leaks, flange failures, pinhole corrosion, or vessel ruptures. For storage and loading, hose types, couplings, and transfer rates define realistic release rates.
Meteorology deserves special attention. Where possible, use site-specific mast data covering multiple years to generate stability-class-weighted wind roses. If regional datasets are used, confirm representativeness regarding sea-breeze patterns and nocturnal stability. For coastal refineries and island terminals, consider directional diurnal cycles for day versus night cases, as night-time stable conditions can substantially extend toxic endpoints.
Model selection follows scenario complexity. Screening-level models handle most continuous or instantaneous releases well; dense-gas capable models are mandatory for heavier-than-air clouds; and computational fluid dynamics (CFD) is warranted for high-stakes congestion/confinement cases, such as enclosed compressor buildings or compact manifolds, where explosion overpressure sensitivity is acute. Documentation should flag when simplified models are used, and why they are adequate for the decision at hand.
Safeguard representation is often the tightest coordination point with operations. Gas detectors, ESD valves, deluge zones, and fireproofing thicknesses map directly to assumptions like isolation time, flame exposure duration, or structural survivability. Verify that the credited response times align with tested values and that power and instrument dependencies are accounted for during loss-of-utility scenarios. If shelter-in-place is a protection layer, ensure HVAC shutdown times and tightness ratings are supported by test results.
Throughout, modelers should work iteratively with HAZOP, LOPA, and emergency response teams. Bowties can be a helpful bridge, showing how initiating events flow into hazardous outcomes and where prevention and mitigation alter the consequence envelope. The final deliverables—maps, distance tables, and equipment survivability notes—are only as defendable as the trail of input decisions that underpin them.
Benefits and Practical Use Cases
Disciplined treatment of inputs and assumptions is not bureaucracy; it is how organizations avoid blind spots and overbuild alike. Benefits crystallize in tangible, day-to-day decisions:
- Plot plan optimization: reliable jet fire and explosion envelopes inform separation distances, access ways, and siting of shelters and substations.
- Emergency response realism: consistent toxic endpoints and radiation levels define muster points, egress routes, and command post locations that actually remain tenable.
- Barrier integrity alignment: coherent credit for detectors, isolation, and deluge drives test intervals and spares strategies per IEC 61511 and internal performance standards.
- Regulatory confidence: transparent bases help streamline discussions with OSHAD, Dubai Municipality, Civil Defence, and industry operators’ HSE functions, reducing rework and delays.
- Change management: when MOCs alter inventories or layouts, a traceable model basis enables fast, accurate re-runs to confirm ALARP remains demonstrated.
Concrete examples abound across the region’s high-hazard industries:
- Refinery expansions where revised tower pressures change source terms for C3/C4 systems, requiring adjusted jet fire contours and new fireproofing extents.
- LPG bullet farms adjacent to logistics corridors, where dense-gas dispersion governs fence-line exposure and shelter-in-place criteria for neighboring facilities.
- Pipelines crossing residential buffers, where ignition probability and congestion assumptions affect flash fire and VCE modeling near valve stations.
- Offshore production platforms, where confined module layouts demand refined explosion modeling and equipment survivability checks tied to deluge performance.
- Ammonia storage in power generation sites, where toxicity endpoints, stability classes, and night-work occupancy define risk to personnel and contractors.
Best Practices and Points of Caution
Having seen how this plays out, a set of disciplined practices can keep teams aligned and models defensible:
- Use consistent data hierarchies. Prioritize site-specific engineering data, then validated corporate standards, then recognized public sources. Flag any surrogate values and track their replacement plan.
- Model UAE-relevant meteorology. Incorporate diurnal sea-breeze effects and stable night-time conditions. Test sensitivity to stability classes and humidity for dense gases and flare gases.
- Represent congestion and confinement faithfully. Use plot plans, 3D models, and site walks to tune VCE inputs; escalate to CFD where simplified indices fail to capture critical geometry.
- Document safeguard credits transparently. Align with LOPA and IEC 61511; avoid double counting; trace credit to test records and performance indicators; include common cause considerations.
- Calibrate ignition probabilities to reality. Tie assumptions to area classification, ignition source density, and historical incidents. Don’t borrow values unsuitable for offshore vs onshore or for open vs enclosed areas.
- Define receptor and occupancy credibly. Account for shifts, contractors, and shutdowns; validate shelter performance; align with emergency response doctrine.
- Adopt effect criteria consistently. Use recognized thresholds for thermal radiation, overpressure, and toxicity. Ensure consistency with corporate risk matrices, ALARP narratives, and permitting expectations.
- Run sensitivities. Where uncertainty is material—inventory range, weather extremes, response times—show the envelope. Use the results to prioritize data improvement and barrier reinforcement.
- Peer review and version control. Independent scrutiny and disciplined change logs reduce model drift and help when regulators or insurers ask hard questions.
- Link to MOC and assurance cycles. When equipment, inventories, or layouts change, trigger targeted re-runs and keep your emergency planning current.
Finally, remember that Consequence Modelling Inputs And Assumptions are not static. As operating envelopes shift, as new barriers come online, and as standards like ISO 45001 and ISO 14001 reinforce integrated risk thinking, your model basis should evolve in step, not lag behind.
Frequently Asked Questions About Consequence Modelling Inputs And Assumptions
How should we select meteorological data for dispersion modeling in the UAE?
Use multi-year, site-specific mast data when available to capture wind roses, stability classes, and diurnal sea-breeze cycles. If using regional datasets, validate representativeness and consider separate day/night cases. Because stable night-time conditions (E/F) can lengthen toxic plumes, include them explicitly and ensure humidity and temperature ranges reflect seasonal extremes.
What’s the best way to choose hole sizes and release types?
Let integrity and RBI insights drive event size distributions: common small leaks for corrosion/erosion mechanisms, medium leaks for flange/gasket issues, and rare large ruptures for catastrophic failures. For rotating equipment, include pump/compressor seal failures; for storage, address nozzle and manway leaks, plus catastrophic tank failures only where credible. Avoid generic textbook values detached from your asset’s equipment population and damage mechanisms.
When is CFD necessary for explosion modeling?
Use CFD when congestion and confinement dominate explosion severity and simplified indices cannot capture the geometry: compressor buildings, enclosed modules, tight pipe racks, or complex manifolds. For open, low-congestion areas, validated simplified models may suffice. Document why CFD is or isn’t warranted, and align the choice with the consequence criticality of the decision at hand.
Which toxicity endpoints should we use—AEGL, ERPG, IDLH, or corporate limits?
Select endpoints that match decision context and regulatory expectations. For public receptors, AEGL or ERPG values are common; for worker protection and emergency planning, corporate or jurisdictional criteria may apply. Maintain internal consistency, justify choices in your basis of calculation, and verify alignment with emergency response doctrine and relevant regulator expectations.
How do we credit safeguards without overstating protection?
Match credits to proven availability and response times. Align with LOPA and IEC 61511 verification, avoid double counting (e.g., don’t count both gas detection and SIS trip for the same function if they’re not independent), and include power/instrument dependencies. Where uncertainty exists, run sensitivity cases with and without credit to test impact on outcomes.
Do we need different models for heavy gases like LPG and ammonia?
Yes. Dense-gas capable models are necessary for heavier-than-air releases where gravity spreading and pooling occur. Standard Gaussian models may under-predict distances for these cases. Reflect temperature, humidity, and terrain, which significantly affect plume behavior for heavy gases, especially at night.
Should occupancy include contractors during turnarounds?
Absolutely. Turnarounds can substantially alter personnel density and distribution. If the risk study supports emergency planning or risk to life metrics, include turnaround occupancy scenarios and verify muster capacity and routes under the modeled effects.
How often should we update consequence model inputs?
Tie updates to management of change (MOC). Any change to inventories, operating envelopes, equipment layout, or safeguards should trigger a review. Periodic assurance—annually or aligned with major shutdowns—helps catch drift in assumptions versus current reality.
What’s the relationship between QRA and consequence modeling?
Consequence modeling estimates the physical effects footprints (e.g., heat flux, overpressure, concentration). QRA combines those effects with initiating event frequencies and conditional probabilities (e.g., ignition) to quantify risk metrics. Good consequence inputs are the foundation of trustworthy QRA results.
Can we use corporate standard values if site data is limited?
Yes, but document the hierarchy: site data first, then corporate standards, then recognized public sources. If you must use surrogate values, run sensitivities and create a plan to replace them with measured or verified site-specific inputs when available.
Conclusion
Across refineries, petrochemical complexes, offshore platforms, terminals, pipelines, and power generation sites, the quality of decision-making on siting, safeguards, and emergency planning hinges on the discipline applied to inputs and assumptions. We have seen how inventories and thermophysical properties inform source terms; how meteorology, terrain, and congestion shape dispersion, fire, and explosion behavior; and how ignition probability, safeguards, and occupancy transform outcomes. Treating these elements transparently, aligning them with HAZOP, LOPA, and IEC 61511, and testing uncertainty through sensitivities ensures that models are fit for purpose and defendable to internal and external stakeholders. As operating contexts evolve, keeping your modelling basis synchronized with reality is the surest way to sustain ALARP and maintain regulatory confidence.
To go further on this topic, NPM’s broader Consequence Modelling Services UAE page covers this in more depth.
Our References
NPM’s track record with consequence modeling spans complex, high-hazard industries worldwide, from integrated refineries and gas processing hubs to chemical clusters, marine terminals, and power generation sites handling toxic and flammable inventories. In these engagements, our teams have supported both greenfield and brownfield decisions: shaping plot plans with jet fire and explosion envelopes; validating shelter-in-place strategies against toxic dispersion; informing deluge zoning and fireproofing durations; and reconciling LOPA and SIL assumptions with modelled response times and survivability. What distinguishes NPM’s body of work is the sustained emphasis on input transparency and assumption discipline. We invest in building a clear basis of calculation, documenting scenario selection logic, and aligning with standards such as IEC 61511, ISO 45001, ISO 14001, and widely recognized industry practices for effect criteria and barrier performance. This documentation focus enables operations, engineering, and emergency response teams to use the same modeling backbone confidently, and it streamlines exchanges with regulators and insurers who expect clarity on data origins, version control, and sensitivity bounds. Whether the context is a dense-gas dispersion challenge near a community boundary or an explosion sensitivity in a congested module, the objective remains the same: equip decision-makers with defensible, decision-grade models they can trust to manage major accident hazards effectively over the asset life cycle.
For more information about our references, visit Our References.
Speak With NPM
If you would like to discuss how the points above relate to your facility or to clarify any specifics about inputs, assumptions, and effect criteria, we are happy to help. A brief conversation can often identify the data or sensitivities that will add the most value to your next study. For any further questions or to speak with our team, Contact Us.