Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.758376
Title: Uncertainty based decisions to manage combined sewer overflow quality
Author: Sriwastava, Ambuj Kumar
ISNI:       0000 0004 7431 1493
Awarding Body: University of Sheffield
Current Institution: University of Sheffield
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
Environmental regulators stipulate performance and modelling requirements for water utilities managing sewer networks to demonstrate regulatory compliance in order to limit their impact on the environment. Uncertainty in urban drainage modelling presents challenges to decision makers attempting to achieve compliance with regards to Combined Sewer Overflow (CSO) and treatment plant discharges. This study provides methodologies for making decisions to improve the environmental performance of the urban sewer systems while accounting for uncertainty in model predictions of their performance. In doing so, an objective uncertainty quantification process is first described using a case study in Belgium which enables the water utility to evaluate and report the uncertainty in their CSO spill predictions and is transparent enough to satisfy their regulator. Second, six practitioners from a water utility are interviewed to identify their preferences for uncertainty in the performance variable and the risk of non-compliance. Given identical uncertainty levels in model predictions, individuals' preferences are found to have a significant effect on the decisions taken. Subsequently, two uncertainty based decision models are presented which reflect individuals' preferences in making decisions accounting for uncertainty in model predictions. The first decision model includes the concept of Buffered Probability of Exceedance as a risk measure accounting for the magnitude of extremes along with the mean and the skewness of the performance distribution. The second decision model applies Cumulative Distribution Function (CDF) matching which minimises the difference between the CDFs of the performance variable and a target function specified by the decision maker. The decision models presented in this study enable a better-informed decision making by allowing a comprehensive understanding and representation of modelling uncertainty in evaluating decisions instead of only using exceedance probabilities or extreme values. The decision models provide a significant improvement over existing uncertainty based approaches found in literature to manage sewer overflows.
Supervisor: Schellart, Alma ; Tait, Simon Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.758376  DOI: Not available
Share: