Theory

Theory

Disjunctive fusion is the method to be used to fuse sources in conflict, and consists in the union of fuzzy sets.

Disjunctive fusion is used when one of the two sources, and only one, is taken as being reliable, and the other wrong. This method is appropriate to fusion sources providing either low or non overlapping sets. Disjunctive fusion with the operator ``maximum'' comes down to the following operation:

$\begin{displaymath}\forall \omega \in \Omega, \quad\pi_{\text{disj}}(\omega) = \max ( \pi_1(\omega), \pi_2(\omega) )\end{displaymath}$

(54)

**Figure 26:** Disjunctive fusion.
$\begin{figure}\begin{center}\fbox{\begin{tabular}{cc}\subfigure[Total confli......[No conflict]{\epsfbox{c3-disj4.eps}} \\\end{tabular}}\end{center}\end{figure}$

Figure 26 shows disjunctive fusion behaviour using the operator ``maximum''.

(u)When the sources are in total conflict, we might not know which source is reliable and gives the right answer. Disjunctive fusion does not make any assumptions on the reliability of the sources and stores all the information available. The result is obviously very unclear.
In all other cases, (v), (w) and (x), disjunctive fusion shows similar behaviour: all the information available is stored.

The use of disjunctive fusion, when the sources are in total conflict, always makes it possible to obtain a result. It is not the case with conjunctive fusion. Unfortunately, disjunctive fusion generates a very unclear result as all the information is preserved, even if some part is useless. In fact, as we do not know which source is right, no decision is taken and the information is stored just as it is.

IRIT-UPS