How to create a critical path diagram

Updated April 17, 2017

Critical Path Analysis involves the creation of a network diagram of tasks required for a project. Some tasks depend on the completion of others and several tasks can be carried out concurrently. Each task will have a start and end date and some tasks cannot start until others complete. You have two options for the start and end date of the entire project. You can either specify a date upon which the project must start and work forward from that, or set the date the project must end and work backwards.

Create a table with the following headings: “ID,” “Task,” “Earliest Start,” “Duration,” “Type,” “Dependent on.” Create the table in either a word processor or spreadsheet, to make it easier to alter the values in the table as analysis progresses.

List all the tasks involved in the project. You do not have list them in chronological sequence, although it is natural to remember tasks in time order. For each task, give in a letter as an ID, starting with A. Mark the duration of each task in weeks. On the first pass leave the type and dependency fields blank.

Read through all the tasks and select those that could start right now -- that is, those tasks that do not have any prerequisites. Write “week 0” in the earliest start date column for those tasks. Put a dash in the “Dependent on” column.

Read over the other tasks and see which one cannot start today, because they require one of those “week 0” tasks to complete first. Do not consider those tasks that have to follow the week 0 tasks plus other tasks. Only look at those tasks that can immediately follow a week 0 task. For each of this group, write the ID of the prior task in the “Dependent on” column. Check the number of weeks the prior task will last to get the earliest start date for this task and write that in the “Earliest start” column. Continue until all the tasks have a “Dependent on” value and a week number in the “Earliest start” column.

Go back over the list and see if any of the tasks that are dependent on a prior task are only dependent on part of that task. For example if the prior task can provide all another task needs when it is only partially completed. If this is the case, split out the prior task into two parts and change its ID. Change the “Dependent on” value in the task that can follow this newly created prior task.

Write out the linked tasks that follow on from all your week 0 tasks. You will probably end up with three or four groups. Something like: A, C, F, J; B, E, H; A, C, G; A, D, I.

Mark weeks on graph paper with a pencil. Draw lines down at equal intervals and write the week numbers at the bottom between the lines. Draw a circle half way up the first line and write “Start inside it. Then draw a horizontal line across the number of weeks your first task will take and then draw a circle on the line it reaches. If there is another week 0 task draw a diagonal line up from the start circle to the line representing the week of its end date. This will create a branch for a new path, if there is another week 0 task branch this below the central path to create another task. Branch off either diagonally up or down to create a path for each week 0 path.

Follow each task, marking lines on the path and putting a circle at its end date. When the longest line ends, write “end” in the circle that terminates it. The shorter lines will run out sooner, but join them to the “end” circle with a straight line.

Things You'll Need

  • Graph paper
  • Pencil
  • Word processor or spreadsheet
Cite this Article A tool to create a citation to reference this article Cite this Article

About the Author

Stephen Byron Cooper began writing professionally in 2010. He holds a Bachelor of Science in computing from the University of Plymouth and a Master of Science in manufacturing systems from Kingston University. A career as a programmer gives him experience in technology. Cooper also has experience in hospitality management with knowledge in tourism.