diff --git a/doc/report/annex.tex b/doc/report/annex.tex
new file mode 100644
index 0000000000000000000000000000000000000000..5a13ff80d63589716ede3cd8d594351147117439
--- /dev/null
+++ b/doc/report/annex.tex
@@ -0,0 +1,7 @@
+\section{Annex}
+
+\begin{figure}
+    \center{\includegraphics[width=\textwidth]{figures/fsm/full_mgm2_agent.png}}
+    \caption{Complete MGM2 agent automaton}
+    \label{full_agent}
+\end{figure}
\ No newline at end of file
diff --git a/doc/report/fsm.tex b/doc/report/fsm.tex
index 1a57e3d364fa6a37cec10723d4d9d7dc3941bf22..7fd49a5e965bc14f33cc2caa07a6ea939abdec55 100644
--- a/doc/report/fsm.tex
+++ b/doc/report/fsm.tex
@@ -6,7 +6,7 @@ Agents systematically base their decision on their local knowledge.
 
 The process is triggered by a supervisor agent which invites other agents to begin their rounds of interactions.
 Each agent $a_i$ is sent a \beginAlgo{} message by the supervisor and initiates its variable's value at random in the variable's domain.
-The behaviour of each agent is defined by a finite state automata (see Figure~\ref{full_agent} for the big picture and Figures~\ref{beginning}, \ref{offerer}, \ref{receiver}, \ref{committed} and \ref{uncommitted} for focused extracts ).
+The behaviour of each agent is defined by a finite state automata (see Figure~\ref{full_agent} in annex for the big picture and Figures~\ref{beginning}, \ref{offerer}, \ref{receiver}, \ref{committed} and \ref{uncommitted} for focused extracts ).
 
 Communication of potential moves and their corresponding potential gains helps agents determine who should act. A single move is an action wherevy an agent changes the current value of its variable to another possible value belonging to the variable's domain.
 A coordinated move is similar but involves two agents changing the values of their variables. Note that a move can also be static if the change to \textit{another} value of the domain happens to be the current value of the variable.
@@ -36,11 +36,7 @@ A coordinated move is similar but involves two agents changing the values of the
     \item \final{}: the last state an agent can be into, after having received the \stopAlgo{} message from the supervisor.
     \end{itemize}
 
-    \begin{figure}
-        \center{\includegraphics[width=\textwidth]{figures/fsm/full_mgm2_agent.png}}
-        \caption{Complete MGM2 agent automaton}
-        \label{full_agent}
-    \end{figure}
+
 
 
 
diff --git a/doc/report/main.tex b/doc/report/main.tex
index 066b815458b0dc72b3be773f43e8fbb767c1b256..9c1f0fdc7fa42610718d433c87c59a340d043837 100644
--- a/doc/report/main.tex
+++ b/doc/report/main.tex
@@ -239,6 +239,11 @@ We proceed to the description of the MGM-2 automaton and finally move on to two
 We detailed MGM and more particularly MGM-2 algorithms and presented examples of execution on toy DCOPs.
 We likewise detailed the automaton representing MGM-2 and proposed two examples of execution on toy problems.
 
+
+\newpage
+
+\include{annex}
+
 \bibliographystyle{plainnat}
 \bibliography{main}
 \end{document}