Advancing Robust Multi-Objective Optimisation Applied to Complex Model-Based Water-Related Problems

1 Introduction 
1.1 Background 
1.2 Motivation 
1.3 Research questions 
1.4 Objectives 
1.5 Innovation, practical value and social relevance 
1.6 Thesis structure 
2 Literature review 
2.1 Evolutionary algorithms 
2.2 Main types of multi-objective optimisation (MOO) algorithms 
2.3 Robust multi-objective optimisation (RMOO) 
2.4 Conclusions 
3 Methodology 
3.1 Methodological framework 
3.2 Identification and characterization of uncertainty sources 
3.3 General formulation of the optimisation problem 
3.4 Mathematical definitions of robustness metrics 
3.5 Selection of MOO algorithm 
3.6 Formulation of the proposed RMOO algorithm 
3.7 Reliability of satisfying constraints 
3.8 Dealing with more than two objective functions 
3.9 Experimental plan 
3.10 Exemplifying ROPAR
3.11 Analysis of computational complexity 
4 Robust optimisation of a simple storm drainage system 
4.1 Problem statement 
4.2 Experimental setup 
4.3 Results and discussion 
4.4 Conclusions 
5 Robust optimisation of two larger storm drainage systems 
5.1 Problem statement 
5.2 Experimental setup 
5.3 Case studies 
5.4 Results and discussion 
5.5 Conclusions 
6 Robust optimisation of a storm drainage system: more objectives and sources of uncertainty 
6.1 Problem statement 
6.2 Experimental setup 
6.3 Case study 
6.4 Results and discussion 
6.5 Conclusions 
7 Robust optimisation of water quality in distribution systems 
7.1 Water age minimisation problem and its deterministic solution 
7.2 Using ROPAR with the new MOO, two objective functions, twenty four sources of uncertainty 
7.3 Conclusions 
8 Conclusions and recommendations 
8.1 Summary 
8.2 Conclusions 
8.3 Limitation of this study 
8.4 Recommendations