Project Description

Fellow: ESR8 - Ambuj Sriwastava Host institution: University of Sheffield Duration: 36 months Planned start date: November 2014
Actual start date: December 2014
Project title: Uncertainty analysis frameworks and long term climate change linked to asset management decisions (WP1, WP4, WP6)
Supervisor name: Dr Alma Schellart & Prof. Simon Tait PhD enrolment: Y


  • Develop advice on the feasibility of application of existing uncertainty analysis frameworks and methods in industrial practice.
  • Develop tools for modellers to assess the benefits and cost-saving potential of carrying out routine uncertainty analysis.
  • Consider the propagation of uncertainty through sub models, this will involve the development of a protocol for modellers and end users, in collaboration with ER3 and ER4.
Tasks and methodology: Testing uncertainty frameworks on integrated water quality modelling case studies at Aquafin and Halcrow. This will also involve the identification of model input and calibration parameter uncertainty using available data or statistical elicitation methods when the availability of data is limited1. Discussions on developing tools to optimise the resources for additional data collection in collaboration with ER3 (TUD) and ER4 (Halcrow) and ESR2 (WU). As the project is focused on decision making for assets, the impact of climate derived long term change will be examined using scenario-based techniques and data from probabilistic climate models.
Results: Report and journal paper on application of different frameworks to existing integrated models (D1.1, D4.2, D6.3). Manual/guidelines and journal paper on the practical applicability of uncertainty analysis frameworks linked to asset management decisions, especially with regard to long term climate change (D6.7, D6.3). Journal publications on the use of elicitation methods in limited data circumstances and the impact of climate change uncertainty on water quality predictions and asset decisions.
Dissemination: Reports, Journal papers, two activities/year as MC Ambassador, annual conference attendance.
Planned secondments: To Aquafin (6 months) and CH2M Hill-Halcrow (3 months), both testing practical applicability of uncertainty analysis frameworks.

1 O’Hagan (1998). Eliciting expert beliefs in substantial practical applications, The Statistician, 47, 1, 21-35.