The NETDRAW Procedure

The Layout data set produced by PROC NETDRAW contains all the information needed to redraw the network diagram for the given network data. In other words, the Layout data set contains the precedence information, the ID variables that are used in the current invocation of the procedure, and variables that contain the coordinate information for all the nodes and arcs in the network.

The precedence information used by the procedure is defined by two new variables named _FROM_ and _TO_, which replicate the ACTIVITY and SUCCESSOR variables from the Network data set. Note that the Layout data set has only one _TO_ variable even if the Network data set has multiple SUCCESSOR variables; if a given observation in the Network data set defines multiple successors for a given activity, the Layout data set defines a new observation for each of the successors. In fact, for each (node, successor) pair, a sequence of observations, defining the turning points of the arc, is saved in the Layout data set; the number of observations corresponding to each pair is equal to one plus the number of turns in the arc connecting the node to its successor. Suppose that a node 'C' has two successors, 'D' and 'E,' and the arcs connecting 'C' and 'D' and 'C' and 'E' are routed as shown in Figure 9.7. Then, Table 9.4 illustrates the format of the observations corresponding to the two (_FROM_, _TO_) pairs of nodes, ('C', 'D') and ('C', 'E').

For every (node, successor) pair, the first observation (_SEQ_ = '0') gives the coordinates of the predecessor node; the succeeding observations contain the coordinates of the turning points of the arc connecting the predecessor node to the successor. The data set also contains a variable called _PATTERN, which contains the pattern number that is used for coloring the node identified by the _FROM_ variable. The value of this variable is missing for observations with _SEQ_ > 0.