Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.798604
Title: Adaptive and exible approaches for water resources planning under uncertainty
Author: Pachos, Kevis
ISNI:       0000 0004 8507 9490
Awarding Body: UCL (University College London)
Current Institution: University College London (University of London)
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Abstract:
Planning for water supply infrastructure includes identifying interventions that cost-effectively secure an acceptably reliable water supply. In investigating a range of feasible interventions, water planners are challenged by two main factors. First, uncertainty is inherent in the predictions of future demands and supplies due for example to hydrological variability and climate change. This makes fixed invest-ment plans brittle as they are likely to fail if future conditions turn out to be different than assumed. Therefore, adaptability to changing future conditions is increasingly viewed as a valuable strategy of water planning. However, there is a lack of approaches that explicitly seek to enhance the adaptivity of water resource system developments. Second, water resource system development typically af¬fects multiple societal groups with at times competing interests. The diversity of objectives in water resource systems mean that considering trade-offs between competing objectives implied by the highest performing interventions is useful. Nonetheless, few multi-objective applications have aimed at adaptive scheduling of interventions in long-term water resource planning. This thesis introduces two novel decision-making approaches that address these two challenges in turn. Both approaches apply principles of real option analysis via two different formulations (1) a multistage stochastic mathematical programme and (2) a multi-objective evolutionary algorithm coupled to a river basin simula¬tion. In both cases, a generalised scenario tree construction algorithm is used to efficiently approximate the probabilistic uncertainty. The tree consists of possible investment paths with multiple decision stages to allow for frequent and regu¬lar modifications to the investment strategies. Novel decision-relevant metrics of adaptivity and flexibility are introduced, evolving their definition in the context of water resources planning. The approaches are applied to London's urban water resources planning problem. Results from this thesis demonstrate that there is value in adopting adaptive and flexible plans suggesting that flexibility in activating, delaying and replacing en-gineering projects should be considered in water supply intervention scheduling. To evaluate the implementation of Real Option Analysis (ROA), the use of two metrics is proposed: the Value of the Stochastic Solution (VSS) and the Expected Value of Perfect Information (EVPI) that quantify the value of adopting adaptive and flexible plans respectively. The investment decisions results are a mixture of 'long-term' and 'contingency schemes' that are optimally chosen considering different futures. The VSS shows that by considering uncertainty, adaptive invest-ment decisions avoid £100 million NPV cost, 15% of the total NPV. The EVPI demonstrates that optimal delay and early decisions have £50 million NPV, 6% of total NPV. Additionally, a comparison study of alternative optimisation approaches to water supply capacity expansion problem demonstrate that there is benefit in waiting to allow for improvements around supply uncertainty in the case of London's urban water resources planning problem. The results from the case study suggest that the proposed adaptive planning approach achieves substantial improvement in performance compared to alternative optimisation approaches with fixed plans saving more than £377 million, reducing NPV cost by 35%. Using a multi-objective multi-stage real-options formulation of the water planning problem, the trade-offs between a long-term water management plan's resilience and its financial costs under supply and demand uncertainty are explored. The set of trade-off solutions consist of different investment plans that are adaptive to demand growth, approximated by a scenario tree, while robust to the effects of climate change supply uncertainty, represented by an ensemble of supply (hydro-logical) scenarios. Results show that, by being adaptive to demand uncertainty, the total NPV of the most resilient plans is lowered by 58.7%. The value in de¬laying investments by waiting for more accurate supply and demand estimates is 28.9% of total NPV. It should be noted that the results from the case study are indicative and should not be considered prescriptively as they are based in a simplified representation of London's water supply system and should be further tested with the more detailed simulation model employed by the water utility which includes the latest proposed option designs, includes requirements to supply neighbouring water utilities, and considers more objectives.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.798604  DOI: Not available
Share: