Verkehrsplanung und Verkehrstechnik | Institut für Verkehrswissenschaften

Technische Universität Wien

What are Causal-Loop-Diagrams (CLD)?

Introduction

Causal-Loop-Diagrams (CLD) can be interpreted as a simple language to describe, analyse and communicate dynamic systems. CLDs consist of only three components: elements (entities), interrelations incl. polarities (links) and time delays. CLDs are not sequential, i.e. there is no beginning and no end but rather closed circuits (feedback loops). CLDs are a useful tool to communicate mental models. CLDs can be used to determine the structure of a system and thus its basic behaviour. A CLD describes a system by its elements and the interrelationships existing between them. The elements have to be so called non-specified quantities, i.e. it must be possible to state, whether the element increases or decreases. The relationships between the elements are represented by an arrow and an associated polarity. The polarity indicates how an entity behaves when the influencing entity changes. An arrow marked with the symbol “+” represents the polarity “same”. This means that if the influencing entity (cause) increases, the entity (effect) increases above what it would otherwise have been, and if the cause decreases, the effect decreases below what it would otherwise have been. An arrow marked with the symbol “-” represents the polarity “opposite”. This means that if the influencing entity (cause) increases, the entity (effect) decreases below what it would otherwise have been, and if the cause decreases, the effect increases above what it would otherwise have been.

Simple example

Figure 1: Simple Causal-Loop-Diagram Population
Simple Causal-Loop-Diagram Population

Figure 1 shows a simple CLD of the development of a population. The system is defined by the three entities population, number of live births and number of fatalities. All three elements are non-specified quantities. The statements that population, number of live births or number of fatalities increases or decreases are reasonable. The relationship between population and number of live births has the polarity same. If the population increases, the number of births increases. But there exists also a relationship of the polarity same between number of live births and population. If the number of births increases, the population increases. These two links form a reinforcing feedback loop. A reinforcing feedback loop is characterised by zero or an even number of links with the polarity opposite. Population increases with each completed cycle through the feedback loop.

A relationship of the polarity same exists also between population and number of fatalities. If the population increases, the number of fatalities increases. On the other hand there exists a relationship of the polarity opposite between number of fatalities and population. If the number of fatalities increases, the population decreases. These two links form a balancing feedback loop. A balancing feedback loop is characterised by an uneven number of links with the polarity opposite. Population oscillates with each completed cycle through the feedback loop. A system can only reach a dynamic equilibrium if it includes at least one balancing feedback loop. If this is not the case, exponential growth and self-destruction of the system in the long term are inevitable.

Note: In the context of CLDs the symbol “+” has nothing to do with good or positive! Additionally the symbol “-“ has nothing to do with bad or negative.!

www.fvv.tuwien.ac.at