• Home
  • |
  • Blog
  • |
  • Managing Project Timelines CPM or PERT

Article 16

Read Time 5 minutes

How do you manage project timelines for complex projects?

 Most of us never get involved in mega projects, like constructing a chemical plant, building a Boing airplane, Sending people on lunar orbit, or the Manhattan Project, but we can read and learn from them. These megaprojects have mega budgets and they still experience failures. These megaprojects also provide us with opportunity for research and thereby develop better methods. Did you know that Critical Path Method(CPM) was developed by DuPont during 1950s? Program Evaluation and Research Technique was developed by US Navy during 1950s? [ref. 1]

 If you have managed any reasonable size project, chances are you have heard of and perhaps used CPM.

For this short article I will assume that and explain the PERT approach.

 However, for those who have not heard of CPM, I will provide a brief primer.

 Let us assume we have an experienced team. We want to build an Alkylation plant. Table 1 shows a simplified task list. 

The team reviews the task list, concludes it is reasonable and adds their time estimates for each task shown in the Duration column.

Team also identifies task order by adding a Predecessors column which shows precedents for each task because some tasks cannot be started before other related tasks are finished.

This figure drawn using Microsoft Project also shows as RED bar the Critical Path Method, which shows the Critical Path or the shortest route to complete the project on time. For our simple project, finding the critical path is easy, but for complex projects with hundreds of tasks with interrelated precedents, this could require a lot of computations, fortunately project management software such as Microsoft Project used here will identify that fairly quickly.

 Many people use CPM as their preferred approach for project planning and time estimation. Most experienced users add as contingency 10-15% to CPM time to protect against unforeseen problems

A common problem with CPM has been that the project managers tend to use aggressive or overly optimistic time estimates for each task and as a consequence grossly under estimate the actual completion time for the project.

 

During the 1950s US Navy developed a more sophisticated approach called PERT, Program Evaluation and Review Technique to develop better estimates of project time and resource costs. In this PERT approach each time for task is estimated three different ways, called the Optimistic, Most Likely and Pessimistic methods. See Table 2.

 PERT realistic time estimates use all three time estimates for each task. The idea is some task will finish early; a few tasks finish late and most will finish close to the most likely time estimates. Since pessimistic time estimates are typically long, the PERT realistic time estimates are typically quite a bit longer than the CPM methods.

From three estimates of task times, PERT estimates “more realistic” time estimates using statistical formulas.

 For each task, more realistic time estimate is "time - realistic".

 time - realistic = (o + 4m + p) ÷ 6

where o, m, p are optimistic, most likely time and pessimistic time estimates

.Standard deviation for each task’s realistic time estimate is computed using 

 std-realistic = (p-o)/6

 Total variability standard deviation is Total Std-realistic = sqrt(sum of squares of individual task std. dev.)

 CPM estimate is developed using the realistic time estimates for each task and confidence interval is calculated using 95% or 99% confidence intervals.

Both CPM and PERT techniques can be used for R&D projects, equipment and pilot plant projects, and full blown construction projects. 

We can also extend the PERT techniques to estimate resource costs.

If we pick scope and time as our two degrees of freedom and keep them fixed, we cannot control the project cost! If we pick scope and cost as our two degrees of freedom and keep them fixed, we cannot control the project time! And so on.

So project triangle conveys we can control only two out of the three. What would you like to manage? Sacrifice?

What can we learn from software projects?

Software projects are notorious [ref 2]. Most software projects are 30-70% late, result in 30-70% higher costs and still result in poor quality! One key reason for that is it is software; we cannot make a good two dimensional blue print. Mental processes that we try to codify cannot be thought through or documented with sufficiently high accuracy. Also people keep changing their mind, what they need in software, so, specifications keep changing.

What can you do? Be paranoid. Track project timeline; track project buffer time as a key metric. If buffer time is reduced, investigate. Keep a check list. Check often. 

So crank up your Product Development engines... Let us speedup new product development and growth rates. And let the fun begin!

{"email":"Email address invalid","url":"Website address invalid","required":"Required field missing"}

Want to learn more?

Check out these articles below

>