Skip to content

Toggle service links

You are here

  1. Home
  2. Research
  3. Research groups
  4. Statistics
  5. Bayesian Statistics
  6. Dynamic Forecasting in Dam Safety

Dynamic Forecasting in Dam Safety

Flooding following a dam collapse

Flooding from a tailings dam burst in Brazil.

This research investigates both Bayesian and classical statistical forecasting methods as part of monitoring control systems for safety assessment of artificial dams. The system combines real time sensorial measurements from various monitoring devices together with past data and expert judgement to quantify patterns and generate forecasts that can be used in risk assessment. Current applications of the methodology developed so far include (a) monitoring and forecasting using data from physical displacements of the dam that supplies the Itaipu hydroelectric plant in southern Brazil, and (b) supporting safety assessment of tailings dams in various mining plants in Brazil, Peru and Chile. As a result of this research, a start-up company was created to develop a system based on the developed methodology. The system is now fully operational at various mining plants following the recent disasters at Mariana and Brumadinho tailing dams in Minas Gerais, southeast Brazil.

Related publications: