Disaster Risk Gateway is undergoing development in July-August 2024. During this period pages will change and there may be short outages.

MATRIX (New Multi-HAzard and MulTi-RIsK Assessment MethodS for Europe)

From Disaster Risk Gateway
Jump to: navigation, search

Year of publication: 2015

Access: Freely available to download from journal website

Link: https://doi.org/10.1080/17499518.2015.1041989

Organisation(s) / Author(s): Norwegian Geotechnical Institute (NGI); Analysis and Monitoring of Environmental Risk (AMRA); Institute of Geophysics, Swiss Federal Institute of Technology Zurich (ETHZ); Helmholtz Centre Potsdam, German Research Centre for Geosciences (GFZ); DNV GL

Description

The MATRIX framework is a three-level framework for producing a multi-risk assessment from existing multiple single-risk assessments. It considers hazard and risk interactions as well as changes in vulnerability with time. The framework was developed during the EC FP7 Project MATRIX (New Multi-HAzard and MulTi-RIsK Assessment MethodS for Europe, 2010-2013).

The three-levels of the framework are:

  1. qualitative multi-risk analysis
  2. semi-quantitative multi-risk analysis
  3. quantitative multi-risk analysis.


The level 1 analysis comprises of a flow-chart that is intended to hep guide the user decide whether a multi-risk analysis is required and therefore whether to proceed with levels 2 and 3 of the framework.

Level 2 consists of assessing interactions between hazards and dynamic vulnerability using a matrix approach.

The final level of the framework consists of a detailed multi-risk analysis using a quantitative model based on Bayesian networks (BaNMUR). This is used to estimate the probability of hazard interactions and model changes in vulnerability of a system with time. Uncertainties and their interrelationships are represented by probabilities within the model.

Examples demonstrating the application of the framework are available in Liu et al. (2015).

Technical considerations

Knowledge of Bayesian networks.

Keywords

Multi-risk; Cascading hazards; Time-variant vulnerability; dynamic vulnerability; Bayesian network; matrix