A computational Intelligence-based Method to ‘Learn’ Causal Loop Diagram-like Structures from Observed Data

Abdelbari, Hassan; Shafi, Kamran

doi:10.1002/sdr.1567

Cited by 17 publications

(8 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A more quantitative approach to develop causal loop diagrams is through the use of statistical and machine learning methods that can synthesize causal loop diagrams by exploring data sets in a more automated way. Abdelbari and Shafi (2017) showed a general application of this using a neural network‐based methodology to learn from available data and design alternative conceptual structures for modeling.…”

Section: Discussionmentioning

confidence: 99%

Eight Archetypes of Sustainable Development Goal (SDG) Synergies and Trade‐Offs

Hosseini¹,

Bertone

Gao

et al. 2022

Earth's Future

View full text Add to dashboard Cite

Achieving the Sustainable Development Goals (SDGs) is contingent on managing complex interactions that create synergies and trade‐offs between different goals. It is, therefore, important to improve our understanding of them, their underlying causal drivers, future behaviors, and policy implications. Prominent methods of interaction analysis that focus on modeling or data‐driven statistical correlation are often insufficient for giving an integrated view of interaction drivers and their complexity. These methods are also usually too technically complex and heavily data‐driven to provide decision‐makers with simple practical tools and easily actionable and understandable results. Here, we introduce a flexible and practical systemic approach, termed archetype analysis, that generalizes a number of recurring interaction patterns among the SDGs with unique drivers, behaviors, and policy implications. We review eight interaction archetypes as thinking aids to analyze some of the important synergies and trade‐offs, supported by several empirical examples related to the SDGs (e.g., poverty, food, well‐being, water, energy, housing, climate, and land use) to demonstrate how they can be operationalized in practice. The interaction archetypes are aimed to help researchers and policymakers as a diagnostic tool to identify fundamental mechanisms of barriers or policy resistance to SDG progress, a comparative tool to enhance knowledge transfer between different cases with similar drivers, and a prospective tool to design synergistic policies for sustainable development.

show abstract

Section: Discussionmentioning

confidence: 99%

Eight Archetypes of Sustainable Development Goal (SDG) Synergies and Trade‐Offs

Hosseini¹,

Bertone

Gao

et al. 2022

Earth's Future

View full text Add to dashboard Cite

show abstract

“…To support our discussion on the challenges faced, Figure 1 captures the dynamics of the effort using a causal loop diagram -a systems dynamics representation (Abdelbari and Shafi, 2017).…”

Section: Analysis Of Case Studymentioning

confidence: 99%

Briefing: Process digital twin: lessons learned from a construction case study

Akinyemi

Erdogan

Bosché

et al. 2022

Proceedings of the Institution of Civil Engineers - Management,

View full text Add to dashboard Cite

This briefing paper reflects on an effort to create a digital twin of the work processes of a construction company. The digital twin was to be used by the business to take advantage of digital techniques like machine learning, artificial intelligence and data analytics in optimising their activities. However, the effort to create the required process maps took longer than expected. In capturing the lessons learned, a system dynamics approach was used to reveal the structural and behavioural issues that prevented the timely completion of the task. The issues identified include problems with planning, process documentation, process standardisation, leadership and stakeholder engagement. From analysing the results, good practices are recommended to ensure timely delivery of similar tasks in the future.

show abstract

“…CLDs are not the final simulation and are not a mandatory part of the system dynamics modelling process [8]. However, the CLDs allow a smoother transition to the final quantitative stock-and-flow diagrams used in simulation [9].…”

Section: Introductionmentioning

confidence: 99%

A Multi-Methodology Approach to Creating a Causal Loop Diagram

Dhirasasna

Sahin

2019

Systems

View full text Add to dashboard Cite

Developing causal loop diagrams (CLDs) involves identifying stakeholders and endogenous variables and formulating variable causal relationships. Traditionally, the CLDs are developed mainly using a qualitative approach such as literature review, observations and interviews with stakeholders. However, modellers may question which stakeholders should be approached, whether the relevant variables are selected, and what to do when stakeholders perceive different variable relationships in the CLDs differently. Applying in a case study, this research proposes a multi-method approach by combining both quantitative and qualitative methods to select stakeholders, identify endogenous/exogenous variables, and develop the CLDs. The proposed quantitative method is expected to provide modellers with a justifiable stakeholder and variable selection process. The method also highlights possible hidden variables and relationships, which were further explored with a traditional qualitative approach.

show abstract

A computational Intelligence-based Method to ‘Learn’ Causal Loop Diagram-like Structures from Observed Data

Cited by 17 publications

References 32 publications

Eight Archetypes of Sustainable Development Goal (SDG) Synergies and Trade‐Offs

Eight Archetypes of Sustainable Development Goal (SDG) Synergies and Trade‐Offs

Briefing: Process digital twin: lessons learned from a construction case study

A Multi-Methodology Approach to Creating a Causal Loop Diagram

Contact Info

Product

Resources

About