Week 12 Reading

Multi-Objective Optimization

Draft Material

This content is under development and subject to change.

Overview

Yang et al. (2023)

Discussion Questions

  1. Yang et al. optimize for both flood reduction benefits and ecosystem co-benefits of nature-based solutions. Why do the authors treat these as separate objectives? What information would a weighted sum lose?

  2. The paper uses NSGA-II to approximate the Pareto front. What role does the choice of algorithm play in the results? How would you assess whether the algorithm has converged to a good approximation of the true front?

  3. Look at the Pareto front figures in the paper. Pick two solutions on opposite ends of the front and describe the trade-off a decision-maker would face in choosing between them. Is there a “knee” — a region where small sacrifices on one objective yield large gains on the other?

  4. The authors apply their framework to Sint Maarten, a small island recovering from Hurricane Irma. How does the local context (limited resources, post-disaster recovery, small geographic scale) shape which objectives matter and how results should be communicated to decision-makers?

  5. Compare this paper’s approach to the benefit-cost analysis framework from Week 6. Under what conditions is BCA sufficient? What does multi-objective optimization provide that BCA does not?

References

Yang, S., Ruangpan, L., Torres, A. S., & Vojinovic, Z. (2023). Multi-objective Optimisation Framework for Assessment of Trade-Offs between Benefits and Co-benefits of Nature-based Solutions. Water Resources Management, 37(6–7), 2325–2345. https://doi.org/10.1007/s11269-023-03470-8