Article Preview
Top1. Introduction
In a companion paper (Part 1), the authors advocated the need for using a systems approach, based on cross-impact analysis (CIA) combined with network analysis, to analyze the interaction among the 17 Sustainable Development Goals (SDGs). The analysis was based on the rationale that the SDGs form an anthropocentric network consisting of many interactions at the goal- and/or target-level. As reviewed in the companion paper, although the interactions among the SDGs have been acknowledged by many authors in the sustainable development literature (Allenet al., 2017; Coopmanet et al., 2016; Le Blanc, 2015; Nilsson et al., 2016; Nilsson, 2017; Nilsson et al., 2017; Nilsson et al., 2013; UN Water, 2016; United Nations Economic and Social Council, 2015; United Nations General Assembly, 2015; Vladimirova & Le Blanc, 2016; Weitz et al., 2014), little has been done to model how the SDGs influence or depend on one another.
Cross impact analysis originated in the 1960s (Gordon, 2014; Gordon & Hayward, 1968) to analyze “weak [soft] structured systems” (Weimer-Jehle, 2006, p.336) for which theory-based computational (hard) models do not work well due to system complexity and disciplinary heterogeneity. Since its inception, cross-impact analysis has been used a general method for assessing the “interrelations between the most important influential factors in a system by experts who evaluate [subjectively] pairs of these factors” (Weimer-Jehle, 2006, p.336). Cross -impact analysis is a soft systems approach that uses a systemic process of inquiry.
As discussed in Part 1, combining cross-impact analysis and network analysis has many advantages when analyzing the SDGs. More specifically, the approach enables decision makers to: (1) attain a better qualitative and quantitative understanding of the way the SDGs interact; (2) detect emerging patterns resulting from those interactions; (3) use context-specific information about direct impacts to identify indirect effects; and (4) identify leverage points hidden within the SDGs.
Cross impact analysis also has limitations. As noted by Checkland and Poulter (2006), the limitation of a soft systems approach is that it uses an action-oriented learning process for learning to address problem areas and not the content of the problems. The second limitation of cross-impact network analysis is that it is static; it represents a snapshot of a system in time and cannot handle dynamic (time-dependent) issues.
A third limitation is that cross-impact analysis is entirely based on human input, ideally, expert judgement. The cross-impact matrix, which is at the fundamental core of cross impact analysis, requires judging experts to estimate how variables interact with each other, the degree of their interactions, and the possible results of their impacts (Weimer-Jehle, 2006). Essentially, the quality of the analysis depends on the accuracy and expertise of the people undertaking that analysis. The experts filling out the cross-impact matrix are “expected to possess insights which rather should be the results of an analysis (Weimer-Jehle, 2006, p.337).” This is the classic Catch-22, where input depends on output, rendering the analysis ineffective through circle-logic.
We present below an alternative hard systems approach to analyze the interactions among the SDGs which helps address the limitations mentioned above. The approach uses system dynamics which, as the name suggests, accounts for how the components of systems and their linkages change with time. Furthermore, it takes under consideration the components of a system in addition to following a systemic process of inquiry. Unlike cross-impact analysis that can be carried out using simple tables in Excel, system dynamics analysis requires using more sophisticated software packages. The system dynamics (SD) models presented in this paper use the STELLA (Systems Thinking Experiential Learning Laboratory with Animation) Professional software by ISEE Systems, Inc. v1.2.1.