Measuring the sustainability of a natural system by using multi-criteria distance function methods: Some critical issues

Abstract

There is an important body of literature using multi-criteria distance function methods for the aggregation of a battery of sustainability indicators in order to obtain a composite index. This index is considered to be a proxy of the sustainability goodness of a natural system. Although this approach has been profusely used in the literature, it is not exempt from difficulties and potential pitfalls. Thus, in this paper, a significant number of critical issues have been identified showing different procedures capable of avoiding, or at least of mitigating, the inherent potential pitfalls associated with each one. The recommendations made in the paper could increase the theoretical soundness of the multi-criteria distance function methods when this type of approach is applied in the sustainability field, thus increasing the accuracy and realism of the sustainability measurements obtained.

Accounting for multiple ecosystem services in a simulation of land‐use decisions: Does it reduce tropical deforestation?

“…The quotients on the right‐hand side of Inequalities 4 quantify the individual relative distances between the achieved landscape level indicators, , and the reference point. We normalize these distances between 0% and 100% by dividing the absolute distance by the range of input values in one uncertainty scenario (Diaz‐Balteiro et al, ). During sensitivity analyses to study the influence of weighting on the results, we included higher weights for each single indicator than for all other indicators.…”

“…8)The simulated decisions of the three perspectives aim for a compromise land allocation to improve levels for each single indicator.Land is allocated to the seven LULC types with the objective of balancing the achievement of indicators comprising the relevant decision criteria for each perspective(Figure 1). Our optimization procedure compares the distances between the most desirable level of each indicator achievable by a single LULC type (set as the reference point) and the levels actually achieved by a given landscape composition, and then minimizes the largest distance(Diaz-Balteiro et al, 2018;Romero, 2001;Stewart, 1992;Tamiz, Jones, & Romero, 1998) across all indicators and uncertainty scenarios (see Section 2.3).…”

Sustainability in Forest Management Revisited Using Multi-Criteria Decision-Making Techniques

“…From a sustainability perspective, it has been implicitly assumed up to now that the criteria selected functioned within weak sustainability parameters [31]. Namely, a certain degree of As could be expected, in the comparison between models (Figure 3), the values of the criteria obtained in the LGP model are lower than the ideal ones reached in the pay-off matrix.…”

“…From the MCDM techniques that fulfil the above condition, goal programming has been selected. This methodology adds flexibility to forest management models [37], and it has been used quite frequently to tackle this type of problem [31]. The basics of goal programming can be seen in [38][39][40].…”

“…The preferences introduced into the model will only take into account the opinion of the stakeholders and exclude intergenerational justice components from the analysis [47]. In addition, only criteria for which values can be obtained throughout the entire planning horizon are considered [31]. That is to say, if the evolution over time of a certain criterion cannot be modelled, it is not included in the analysis.…”