How can I evaluate different NBS decisions depending on multiple conflicting criteria?

Overview

Flood risk reduction, water quality, biodiversity, agricultural income, public expenditure and social acceptance cannot be added together, and a decision that has to serve all of them cannot be reduced to a single number without deciding in advance which of them matters less. The multi-criteria toolkit exists to support that decision without concealing it: it makes the trade-offs explicit and lets those affected by them see and argue about the weighting.

What the toolkit contains

The toolkit brings together the methods used in the project for comparing options against multiple criteria. It covers the structuring of the problem — defining the objective, the criteria and the options in a form that stakeholders recognise and can contest — and the elicitation of preferences, which is where the weights or the rankings come from and which is therefore a social process rather than a technical one.

It covers the treatment of criteria measured in incompatible units, including qualitative criteria that cannot be scored on a cardinal scale without distortion, and the treatment of criteria on which the options are not simply better or worse but differently distributed across groups. It covers sensitivity analysis, which shows how far a recommendation depends on the weights and how far it survives their revision — often the most useful output, because it identifies the options that are defensible under almost any weighting and the ones that are not.

The toolkit is used together with the integrative modelling framework, so that the performance of each option against each criterion comes from the same connected model rather than from separate studies, and with the participatory methods, so that the criteria and the weights are established with the people the decision affects.

How it is used

Applications in the demonstration areas range from comparing interventions within a single site to comparing strategies across sites with different hazards. The outputs are presented as the ranking under stated preferences together with the trade-offs that produced it, so that a decision-maker can see what is being given up and to whom, rather than being presented with a conclusion whose basis is hidden in the weighting.