Followings are certain principles that have to be implemented while constructing the network diagram. These are :
1. For constructing a network diagram, the decision maker should prepare a list of activities required to complete a project and the respective time relationship. In some cases, it may not be possible to list all the activities initially. However, additional activities will come to light as the network progresses.
ADVERTISEMENTS:
Each activity should be represented by only one arrow in the network, i.e., no single activity can be represented twice in the network. Length of an arrow has no significance.
2. Determine the logical order of the activities, i.e., how does an activity fit in with other activities.
In order to ensure the correct logical sequence and inter-relationships, the following three questions, which must be answered for each activity in the project?
(a) What activities precede this activity? That is, what other activities must be completed before this activity can be started?
ADVERTISEMENTS:
(b) What activities follow this activity? That is what activities cannot be started until this activity is completed?
(c) What activities can take place concurrently with this activity? That is, what activities can be worked on at the same time while this one is being performed?
3. The general rule for numbering the events is that no event can be numbered until all preceding events have been numbered. The number at the head of an arrow is always larger than that at its tail i.e., events should be numbered such that for every arrow (i, j), i < j.
In order to conform to this rule, the numbering of the events should be done by a procedure given by Fulkerson:
ADVERTISEMENTS:
(i) Delete all arrows leaving the nodes which have been numbered.
(ii) Number the initial node with 1.
(iii) Continue the numbering by identifying all nodes with no incoming arrows and by assigning them consecutive numbers in any order.
(iv) Repeat (ii) and (iii) until the terminal node is numbered.
4. Draw an arrow diagram to show how the activities are inter-related in time. The starting and ending points of activities are circled. These are called events (or nodes) and are numbered serially from start to finish of a project.
5. A network may have only one initial node (the starting event of the project).