A common scene in detective films: In order to find clues, the police put up a lot of exhibits (photos, records, text, etc.) on the wall, and then draw related arrows between these exhibits, and use the arrows to find out the relationship of the physical evidence. In many accident analysis or causality analysis, analysis models are used to help find the cause of the problem. The more complex accidents and events, the more related elements and the more complex the relationship between the elements. This requires a method to help establish the relationship between the elements and support quick understanding of questions and answers. Here is a commonly used analysis model: association graph.
■Association diagram : A diagram that associates cause and result elements according to the interaction relationship. Use the correlation line to help find the cause of the result.
1. Concept and interpretation
1.
In reality, many research objects of model concept contain complex elements. These elements are mutually causal and coupled together in a complex form. It is difficult to separate and express clearly in a structured form, as shown in Figure 2 (a ) Shows that the elements separated from the object include: causes, results, problems, opinions, goals, methods and other different types. These elements are not in a one-to-one relationship. Obviously, such objects cannot be structured. However, it is easier to express with the "relation diagram", as shown in Figure 2(b), the relation diagram can be used to connect the elements and find the law and causality in the complex relationship.
Figure 2 Association of complex elements
Because the form of this correlation diagram is not restricted, it can freely correlate any type of elements, the analysis efficiency is relatively high, and it is easy to quickly find the cause and effect relationship and the solution from the complex object. The main purpose and function of the association diagram is to analyze the relationship between the elements.
2. Model interpretation
According to the characteristics of the correlation diagram in Figure 3, it can be interpreted from the following aspects:
Figure 3 The meaning of the direction of the associated arrow
1) Direction: Since most nodes are causal to each other, they lack a clear direction and can be initiated by 1 to n nodes;
2) Relationship: There is only a certain association between nodes, but there may not necessarily be a strict logical relationship;
3) Nodes : It can be seen that the characteristics of each node and the direction of the arrow set on the node are different.
□ Node 1: The arrow only goes out and does not enter, indicating that it is active and the main cause of the problem;
□ Node 2, 3, 5: Yes In and out, it means that they are the intermediate cause of the problem;
□ Node 4: Only in and out, it means that it is passive, and it is the place where the problem is concentrated;
4) Structure: There is no definite structural relationship;
5) Scope: There is no clear starting point and end point, and the scope cannot be determined;
6) Convergence: The content of the collected elements does not converge to a certain point;
2. Painting method and scene
1. Model drawing method
The drawing method of the association diagram is very simple, only a circle (or box) and an arrow are needed, see Figure 4, the drawing method is as follows:
□Determine the theme and collect all the elements related to the theme;
□List the elements in a circle , The order is not important;
□Mark the name of the element in the circle;
□Mark the arrow according to the principle of "cause"→"result", "means"→"purpose";
□use color to mark the main element 1 (arrow All outwards);
□Use color to mark the main problem element 4 (arrows all inward);
Figure 4 How to draw an association graph
Although the association diagram advocates free thinking, it is best not to be too random when drawing. The arrangement elements that are too random will not be easy to identify, and it will take a long time to find the causal relationship. If you arrange the elements, you should make a rough classification. , And then arrange the classified elements according to a certain law, which is helpful to quickly find the law of the analysis object.
□As shown in Figure 5a, it is more random and it is very difficult to find the final conclusion.
□As shown in Figure 5b, when arranging the positions of the elements, simply group the elements of different purposes into the four sides, so that you can see clearly. If you find that the position is not right, it is easy to adjust. You can make the "→" association while checking, and you can see the problem. Use an unstructured graphical representation to make as much as possible during the association process. They show a certain regularity, which greatly improves the efficiency of analysis.
Figure 5 Layout of related elements
2. Applicable scene
association diagram is mainly applicable when there is no clear logical relationship between the elements, and it is not sure whether there is a strict association relationship, etc., through the association between the elements, the cause and effect between the elements are gradually found Association, law, logic... etc. Be prepared for the subsequent architectural expressions that can be used with architectural diagrams.
■The next blog post in this series: How to use the analysis model — 2. Fishbone diagram to clearly express causality
For a more detailed description of the analysis model, please refer to the book "Dahua Software Engineering-Requirements Analysis and Software Design".