Decision analysis structures the choices technical assistance raises: which expert to engage, which engagement type fits, whether a declared conflict can be mitigated, whether to deploy to a field site, and whether to continue, rescope or close. It makes criteria, weights, uncertainty and value judgements explicit so that the holder of the decision can examine them.
GCRI · Established · service level R0
Decision analysis under uncertainty is a structured method for choosing among options, making criteria, weights, uncertainty and the value of more information explicit. In technical assistance it frames which expert to engage, which engagement type fits, whether a conflict can be mitigated, whether to deploy to a field site and whether to continue, rescope or close.
The working screen behind this method: its structure, terms and states. Names are placeholders and values illustrative.
Agency · Decision analysis for technical… · Run R-5933
| Requests received | Beta | requests, stated | expert elicitation |
| Teams formed | Uniform | teams, stated | expert elicitation |
| Experts on roster | Normal | experts, stated | expert elicitation |
| Conflicts declared | Beta | declarations, stated | expert elicitation |
Decision analysis under uncertainty provides structured methods for choosing among options when outcomes are uncertain and objectives conflict. It makes trade-offs and value judgements explicit. In governance and institutions and development finance, decision analysis helps choose actions that perform acceptably across interconnected risks and uncertain futures. It clarifies when to act now and when to keep options open. The canonical description is on the STEM Commons.
Competence, context experience, language, availability and independence are scored against criteria set before profiles are seen. Weights are disclosed, evidence labels are kept beside each score, and sensitivity to the weights is reported.
Before a field assignment, the decision holder asks whether a further security, health or access assessment would change the decision. Assessments that could change it are commissioned; those that could not are not.
Programmes set signposts and triggers in advance, such as counterpart turnover or a failed diagnostic, that prompt a review to continue, rescope or close. Options that keep later choices open are identified early.
Each term as this page uses it. Where a standard-setting body defines the term, the wording follows that body.
| Term | Definition |
|---|---|
| Adaptive pathway | A plan that sets out decision points and the signals that would trigger a change of course. |
| Structured expert elicitation | A formal process for obtaining and combining expert judgements with their uncertainty. |
| Risk register | A record of identified risks with their likelihood, impact, owner and mitigation. |
The jurisdictions that hold authority over this question: countries, their first-level divisions and capitals, shown for Europe as an example geography for this subject.
Data from the source register feeds the methods; calibration and an independent test gate the estimate and its limits; the decision stays with the competent authority.
Fig. 01Nodes carry their type, metrics and state; connectors show what feeds what; the gates are the validation steps; the timeline is one application. Design of the method's use, not a record of a run.
Datasets and records used by decision analysis under uncertainty on this platform, with the role each plays and the label its values carry. Sources are listed in the source register.
| Data | Role | Label |
|---|---|---|
| Competence evidence with labels | Scoring inputs | REPORTED |
| Structured expert elicitation | Probabilities where records are absent | ESTIMATED |
| Risk registers and incident records | Likelihood and consequence inputs | REPORTED |
| Cost estimates and budgets | Consequence of each option | ESTIMATED |
| Evaluation findings | Retrospective evidence on similar choices | OFFICIAL |
Analyses are checked through sensitivity analysis, peer review and stakeholder scrutiny. Retrospective review compares decisions with outcomes.
| Check | How it is done |
|---|---|
| Weight sensitivity | Rankings are recomputed across plausible weight sets; choices that flip are flagged for the decision holder |
| Evidence and preference kept apart | Scores record what the evidence shows; weights record what the decision holder values, and each is documented separately |
| Panel independence | Everyone scoring or eliciting declares interests, and conflicted members recuse from the affected choices |
| Retrospective review | Past decisions are compared with outcomes to test whether criteria and probabilities were sound |
Results depend on elicited probabilities and preferences that may be biased. Deep uncertainty can make expected value comparisons misleading.
Systemic Risk Management stewards decision analysis under uncertainty for the Consortium and reviews how it is applied across platforms.
The current service level is R0: information and enquiries. An enquiry receives a reference and names the next step; applications, payments, compute allocations and registrations open only when their team, terms and review capacity exist.