Problems with causal-Loop diagrams
George P. Richardson
A Summary and Evaluation by Riadh Dhaoui
This paper describes the various problems that occur more frequently in causal loop dia-grams, as well as the development of the diagrams and the explication of the behavior of them. Particularly, it is about the difficulties in the traditional definitions of positive and negative links in causal-loop diagrams. The fundamental difficulties arise that the causal loop diagrams mask the structure of the stocks and the flows of the systems. Sometimes we focus so much on the role of the feedback structure in the production of behavior that the crucial role of accumulation processes is lost. Even experienced modelers are easily misled by causal loops diagrams. Since causal-loop diagrams has become very popular, the researchers are trying to make system dynamics accessible to a wider range of people therefore this problem is important to Students (all different layers of society specially engineer). In recognition of these difficulties, the research makes an important contribution which is the flow and stock diagrams have been used as means of communication of the structure (Morecroft 1982).
The researchers propose the following experiment: take the causal loop diagram for the family feud described in Richardson’s essay and ask a random sample of system dynam-ics modelers or students how it will behave. In my experience, not only will you receive a wide range of answers, but most will be wrong. Repeat the experiment with the stock-and-flow diagram (obviously with another group of people). While the answers will vary, the number of correct answers should increase. Recognizing these difficulties, in-ventory charts have been revived as a means of communication structure (Morecroft 1982). Richardson’s contribution, however, should not be used as an argument for aban-doning causal loop diagrams or qualitative system dynamics. But it serves as a precau-tionary measure for the easy use of a slightly abused technique. Despite their problems, causal loop diagrams are likely to remain important tools for communication of the feed-back structure. The Weaknesses in using causal-loop diagrams are, In the one hand, the simplicity of causal loop diagrams, however, hides a subtlety that poses problems that have not been sufficiently recognized. In the other hand, the crux of the problem with causal loop diagrams, is that they make no distinction between information links and level-to-level links. This simplification is generally considered to be one of the ad-vantages of causal loop diagrams, but it has a rather dramatic drawback: in cases involv-ing level flow links, standard characterizations of positive and negative polarities in causal loop diagrams are wrong.
The results obtained by implementing are, Firstly, the traditional definition of positive and negative links fail in a wide variety of cases. Secondly, the traditional definition of positive and negative links in causal-loops fail for at least one link in most causal-loop diagrams system dynamicists might draw. Thirdly, the traditional definition work for links that represent proportional relationships, but fail in every case representing accu-mulations of a rate of flow. Lastly, causal-loop diagram might be improved if two differ-ent symbols were used, acknowledging the two kinds of links. The conclusion of the research is very significant.for example traditional definition of the polarities of causal links and loops are insufficient. Moreover, with slight modifications taking account of the accumulating nature of rate-to-level links, the traditional definitions of causal links can be corrected, with the recognition of conserved flows enhancing the legibility of a causal-loop and the urge to define the polarities of causal loops in terms of behaviour over time, however, must be resisted.
This work impacts my research in two directions. The first is the methods which are used in this research are very old and there are better methods to improve a diagrams or product such as “Scrum”. The second is the coherent structure of this work.
