About SAMO

SAMO (Sensitivity Analysis for Model Output) is an interdisciplinary group of academics and organisations that coordinate and promote research and good practice in sensitivity analysis and related fields. In particular, SAMO organises the International SAMO Conference every three years.

Sensitivity analysis

Sensitivity analysis (SA), in the most general sense, is the study of how the ‘outputs’ of a ‘system’ are related to, and are influenced by, its ‘inputs’. Usually the system is a computational model, used to simulate the functioning of a real-world system of interest (e.g. a hydrological system, modelling the economy, climate models, etc.).

The most common context of sensitivity analysis is to understand how the uncertainties in a model output (its results) are related to uncertainties in the inputs. Models are simply approximations to real systems, and therefore have many associated uncertainties in input parameters, boundary conditions and other assumptions. However, the relationship between input uncertainties and output uncertainties can be complex.

Uncertainty analysis quantifies the uncertainty in model outputs, given the uncertainty in the inputs. Sensitivity analysis, on the other hand, allows to decompose the output uncertainty into chunks which can be attributed to inputs and sets of inputs. This allows us to understand which inputs are really responsible for the output uncertainty, and which have little effect.

Why is SA useful? In short, it addresses several fundamental overarching purposes of systems analysis and modeling:

  • Scientific discovery to explore causalities and how different processes, hypotheses, parameters, scales and their combinations and interactions affect a system;
  • Dimensionality reduction to identify uninfluential factors in a system that may be redundant and fixed or removed in subsequent analyses;
  • Data worth assessment to identify processes, parameters and scales that dominantly control a system, for which new data acquisition reduces targeted uncertainty the most;
  • Decision support to quantify the sensitivity of an expected outcome to different decision options, constraints, assumptions and/or uncertainties.

Since computer models are ever-more used to drive economic, political and business decisions, the importance of properly understanding and accounting for model uncertainty has become only more acute.


Sensitivity analysis has been an active field of research for decades, and efficient methods that can be applied in most contexts are fairly well established among those familiar with the field. However, many applied models are still built without a comprehensive sensitivity analysis, or with a version of sensitivity analysis that is not rigorous or optimal. Among other things, this risks that the decisions taken on the results of such models could be ill-informed.

SAMO aims to promote:

  • The use of best practice sensitivity analysis methods across all forms of computational modelling
  • Further research into sensitivity analysis methods, to improve efficiency and applicability

It aims to do this through the SAMO conference series, as well as by generally creating a cross-displinary community of sensitivity analysis practitioners that can share best practices and discuss common challenges.

SAMO Board

The SAMO board is currently composed of 14 academics representing institutions across Europe, North America and Asia. Find out more here.


The SAMO conference is held every three years, in a location that is voted on by the SAMO board, from any proposals that offered from institutions. Find out more about the SAMO conferences here.