The Critical Path Method (CPM) is a valuable tool for calculating the duration of a project from start to finish. So this tool is helpful to define the schedule of a project.


The first step is defining all the activities needed and determining the duration and eventually the predecessor (an activity that needs to happen first to another action). So the starting point is a table like this:

ActivityPredecessorDuration
AStart5 Days
BA3 Days
CA2 Days
EB1 Day
FC1 Day
Table1 – Activity Duration and Predecessor

From the table1 we start by depicting all the paths in a graph like image1.

CPM step1
Image1 – CPM Step 1

In the second step we add the time that every activity need and we also add:

  • Early Start: The early start of one activity, that is at the end of his predecessor;
  • Early Finish: The early start + the duration of the activity;

In the image2, we have the result where we can quickly look at path A=>B=>C is the longest, and it’s called critical path because if it takes more than nine days, also the project needs more than nine days. Instead, path A=>C=>F needs eight days; if we use nine days, we don’t have a delay on the entire project.

CPM step 2
Image2- CPM Step 2

In a project, you can have more than one critical path, which means more risk. On the graph you can also add this other value:

  • Late start: the latest day that an activity can start without affecting the end date;
  • Late finish: the latest day that an activity can start without affect the end date;

The activity on the critical path has the same late start/early start and the same late finish/early finish. For the other, you can calculate it simply starting from the graph’s end and assuming that the activity ends precisely at nine days (the critical path method). Next, you continue to subtract the duration of each exercise. In image3, you can look at an example.

CPM Step 3
Image3- CPM Step 3

Looking at image 3, you can easily say that, for example, activity C can start at day 6, instead of day 5, without affecting the nine-day schedule.

In all of these examples, we have to look at the finish-to-start arrow: one activity starts when the predecessor ends. Other kind of connection are:

  • Finish to Finish;
  • Start to Start;
  • Start to Finish;

They are less used but, in particular, can be helpful.


Meanwhile, the CPM gives you the minimal hypothetic duration; for a complete schedule, you need to pass to actual days on a calendar.
Gantt chart helps pass from a hypothetical nine-day schedule to the final plan. How can you do that?

You start from the information in the table1, and for each activity, you find the natural person or team that will do the job. You ask need to ask the actual start and end day of the activity considering, in addition of the predecessor, also that:

  • Some activities, even if not correlated, can’t be made in parallel. For example activity C and B, even if they isn’t connected, require the expertise that only one person has;
  • Some activity can be done in less time, maybe using more than one person;
  • You need to take into account holidays, people that doesn’t work;

Finally, you put all the activity, with the actual day estimation, on a bar chart, and you got something like the image4:

Image4 - Gantt example
Image4 – Gantt example

As you can see in the image4, you have the activity on a bar chart, but you also have the actual day on the top. This can be a final schedule if you estimate each activity from a real person or team (and in this case, you also have a table with the assignment of the activity to each person or team). If you don’t have the assignment, you can only do a macro-schedule that will need further analysis.

Share on: