Scenario analysis helps a team test decisions against different possible futures. It explores how changes in technology, operations, regulation or society could affect risks and opportunities. The aim is to find actions that remain effective when conditions change, and to identify when the plan needs a review.
What scenario analysis does
A scenario describes a plausible future and the sequence of changes that could produce it. It gives the team a shared basis for asking what could happen and how a proposed decision would perform. The Government Office for Science describes scenarios as possible futures, rather than predictions. [1]
A best case, expected case and worst case can help test assumptions. However, three variations of the same forecast can miss important changes. A useful scenario set can also explore different operating conditions, dependencies or stakeholder needs. State what each label means. A “worst case” within the study is not proof that no more severe outcome is possible.
When to use the method
Use scenario analysis when a decision must remain effective over time. Examples include a major equipment purchase, a new operating model, a maintenance strategy or a long-term infrastructure plan. It can support risk identification, analysis and evaluation. It can also reveal opportunities, such as equipment that can serve several future uses.
The method is useful when benefits and harm affect different people, places or generations. Record these differences. An overall benefit can hide a severe impact on one community or a future maintenance team.
Prepare the study
Define the decision, system boundary and time horizon. State the objectives and the criteria that will be used to compare options. Include safety, environmental effects, service performance and cost where relevant.
Select people with knowledge of the system, its users and its external context. Include operations, maintenance and relevant specialists. Use research, operating records and evidence of emerging changes. Record the source and date of important information. Separate established facts from assumptions and expert judgement.
Build plausible scenarios
First, identify the drivers of change. These can include new technology, workforce skills, supplier capacity, demand, regulation and environmental conditions. Consider changes caused by other risks, including a disruption that changes several dependencies at once.
Assess each driver for its importance to the decision and the uncertainty about its future state. Give attention to drivers with high importance and high uncertainty. Predictable changes still belong in the study because they affect all scenarios.
Develop a small set of distinct scenarios. For each one, describe the starting conditions, the important changes and their sequence. Explain why the combination is plausible. Check that the assumptions are consistent. Invite people outside the initial team to challenge the stories and identify missing perspectives.
Test the decision and plan a response
Apply the same assessment criteria to each scenario. Ask which controls still work, which assumptions fail and which dependencies become critical. Consider effects on different stakeholder groups. Identify changes that help across several scenarios and actions that are useful only under particular conditions.
For example, an illustrative maintenance study could examine reliable spare-part supply, long delivery delays and withdrawal of manufacturer support. The team could compare additional stock, an alternative supplier and equipment replacement. These are teaching examples, not findings from an actual project.
For each proposed response, assign an owner and identify the time needed to act. Define indicators that can show a change in conditions, such as delivery trends or an announced end-of-support date. Set review points and decision triggers early enough to allow a response.
Treat probability with care
Distinguish the likelihood of a scenario from the likelihood of an event within that scenario. A conditional failure estimate does not establish how likely the whole future is. Scenario analysis alone does not produce reliable probabilities.
Where probability estimates have a sound basis, document that basis and the associated uncertainty. Where they do not, state that likelihood is unknown. Avoid invented percentages. Exploratory scenarios can still support decisions by showing which options perform acceptably across different futures.
Do not combine scenario results into an expected value unless the probabilities and scenario coverage justify the calculation. Overlapping or incomplete stories cannot automatically be treated as mutually exclusive outcomes that cover every possibility.
Case study Thames Estuary 2100
The Environment Agency uses the Thames Estuary 2100 plan to manage future tidal flood risk. Its adaptive approach combines planned investment with reviews of changing conditions. It monitors indicators and can change the timing or choice of future interventions. [2]
The 2026 monitoring review examined water levels, defence condition, barrier operations and changes in people and property at risk. It supported the planned interventions while identifying areas that needed more frequent or detailed checks. It also noted a newer plausible worst-case sea level scenario for consideration in future work. [3]
This case illustrates the value of connecting future scenarios to observable evidence and decisions. The lesson for a safety project is to define how new information will change the plan. A scenario report has limited value if no one monitors its assumptions or owns the response.
Outputs strengths and limitations
Retain the scenario descriptions, evidence, assumptions and comparison of options. Record uncertainties, recommended actions, owners, indicators and review dates. There may be no single preferred future. The useful output is a clearer basis for choosing and adapting a course of action.
Scenario analysis can expose weak assumptions and support preparation for unfamiliar conditions. Its limits include poor data, narrow participation and convincing stories with little evidence. The selected scenarios will not cover every future. Review them when new information changes their basis, and use further technical analysis where a safety decision needs stronger evidence.
Hashtags
#ScenarioAnalysis, #RiskAssessment, #RiskManagement, #SafetyEngineering, #Resilience, #SystemsSafety, #TheSafetyArtisan
Sources
[1] Government Office for Science. The Futures Toolkit. Scenario building guidance (2024). Read source
[2] Environment Agency. Taking an adaptive approach to Thames Estuary 2100 (2023). Read source
[3] Environment Agency. Thames Estuary 2100 15-year monitoring review (2026). Read source
