Project planning

by Admin
Updated: July 16, 2018

Basic principles for good project planning that can be applied to even the smallest projects

A project involves many components summed up in a single idea. It is usually defined in terms of its ultimate goal which is great for maintaining focus. However, even with a few components, it is easy to underestimate time & cost overruns. As the number of components increases so does the possibility of something going wrong.

While inadequate planning almost guarantees that even the smallest projects will encounter problems, close attention to detail is still not sufficient to anticipate the unexpected. Good project planning seeks to protect against all eventualities in order to achieve the objective by the deadline without too much stress in the process.

Project planning can be a vast subject. Here is a quick rundown of some key features for the sake of keeping it as simple as possible.

Any collection of tasks can be classified as a project. “Getting organized” can be a project, even though it may not have specified start & finish times and the tasks are not connected.

Normally, however, a project is defined with these parameters:

  1. Objective
  2. Finish date
  3. Required tasks
  4. Required resources
  5. Start date

Dependencies

The first thing I like to do is break down the gap between the objective and now into a dependency chain, i.e. what cannot be started until another task has been done.

Arguable the first formal method for doing this was a system called “Program evaluation and review technique” (PERT) developed by the US Navy in the 1950s. PERT gave rise to critical path analysis (developed in the late 1950s by Morgan R Walker of DuPont and James E Kelley Jr of Remington Rand) and can make make use of Gantt charts (invented Henry Gantt in the early 1910s).

Critical paths

Advantages: Critical path analysis is used to identify bottlenecks. For example, it is not enough to know that step A comes before step B if either are severely limited in throughput.

Disadvantages: Strictly speaking, critical path analysis does not necessarily identify overlap. Each node is normally considered to involve parallel processing (where nothing is contributed until the node is complete) in a linear path. Otherwise, parallel pathways are spawned which leads to very complicated-looking charts.

For example, it may take step A a long time to complete all units needed by step B, but B can actually be started as soon as A has produced one unit.

Critical Path Analysis (CPA) a.k.a. Critical Path Method (CPM), along with its precursors and modifications, has formally defined methodology which can become complex when stepping beyond the basic concept of linked nodes.

Gantt charts

Advantages: Gantt charts are horizontal bar charts that with time as the x-axis. They show each task as a function of time and also show the scheduling relationships of the tasks relative to each other. In their simplest form they are very easy to understand.

Disadvantages: Strictly speaking, Gantt charts do not illustrate dependency. Although it’s possible to annotate these charts to provide this information, they can become cluttered if there is a lot of interdependency.

For example, if step B is shown as starting at the same time step A finishes, it is no guarantee that B requires A.

The bar for each task can show completion percentage and the extent to which the demarcations are behind or ahead of the current time marker shows whether or not they on schedule.

Simplified, hybrid

If two or more chains can be separated out (e.g. product and packaging), I define them as separate sub-projects tracked by a very simple master project on a traditional Gantt chart. All senior management want to know is: Are we on track?

I have a hybrid program charting method which works for most of my own projects. It uses paths only, although parallel processes can sometimes be represented by pseudo-nodes.

A takes 5 weeks to complete but begins to produce results that can be processed by B at week #2. B can run up to 4 batches in parallel but C requires all processing to be complete before it can be started. The area including all B processes can be represented by a pseudo node.

Where it all goes wrong

Estimating project costs is usually not too difficult. Estimating time is usually where it all goes wrong. And time overruns inevitably mean cost overruns.

I learned the hard way in my very first business, that there are two principles at work:

Planning Fallacy: We always underestimate the time we need to complete our own tasks, despite past experience to the contrary (proposed by Daniel Kahneman and Amos Tversky in 1979).

This optimism bias doesn’t apply when estimating the time needed by others. So, a good simple solution is to incorporate estimates by others who have relevant experience but who are not personally involved. Failing that, double it!

Hofstadter’s Law: It always takes longer than you expect, even when you take into account Hofstadter’s Law (proposed by Douglas Hofstadter in 1979).

Since this law is recursive - every time you arrive at an estimate, you have to revise it upwards - it basically states that projects can never be completed on time. Thankfully, it’s really a principle, not a law, but it still needs to be taken seriously. Very seriously indeed.

Methods I have used include multiple safety margins, false deadlines and contingency plans (including damage limitation).

More info

“The Program Evaluation and Review Technique (PERT)” at pmknowledgecenter.com.

“Critical Path Analysis” at tutor2u.net.

“The Ultimate Guide to the Critical Path Method” at smartsheet.com.

“The Ultimate Guide to Gantt Charts” at projectmanager.com.

“The Planning Fallacy Explained” at bigthink.com.

“Hofstadter’s Law and Realistic Planning” at psychcentral.com.

Internal links

Micro tasks Winning strategies The focus dilemma ABCD system Go to Articles
Agree? Disagree? Questions? Answers?
Please post a comment...
Log in