Affiliations: OMIS Area, Schulich School of Business, York University, Toronto, ON, Canada. Tel.: +1 416 736 5074; Fax: +1 416 736 5687; E-mail: syeomans@schulich.yorku.ca | [x] Engineering Optimization and Modeling Center, School of Science and Engineering, Reykjavik University, Reykjavik, Iceland | [y] Mathematics and Scientific Computing, National Physical Laboratory, Teddington, UK
Abstract: Public environmental policy formulation can prove complicated when the various system components contain considerable elements of stochastic uncertainty. Invariably, there are unmodelled issues, not captured or apparent at the time a model is constructed, that can greatly impact the acceptability of its solutions. While a mathematically optimal solution may be the best solution to the modelled problem, it is frequently not the best solution for the underlying real problem. Consequently, it is generally preferable to create several good alternatives that provide very different approaches and perspectives to the same problem. This study shows how a computationally efficient simulation-driven optimization (SDO) approach that combines evolutionary optimization with simulation can be used to generate multiple policy alternatives that satisfy required system criteria and are maximally different in decision space. The efficacy of this stochastic modelling-to-generate-alternatives approach is demonstrated on a waste management planning case. Since SDO techniques can be adapted to model a wide variety of problem types in which system components are stochastic, the practicality of this approach can be extended into many other application areas containing significant sources of uncertainty.
Keywords: Modelling to generate alternatives, simulation-driven optimization, environmental decision making under uncertainty, planning and strategy
