Monte Carlo simulation challenges. Activity lead (negative lag) challenge.

Monte Carlo simulation challenges. Activity lead (negative lag) challenge.

Odesa anecdote:

– Tram. Can you tell me when the Deribasovskaya stop will be?

– Follow me and get off two stops earlier!

Lead (or negative lag) is a situation when the successor activity overlaps the predecessor activity.

It is very beneficial to overlap activities as it may help to shorter project delivery, so leads should be encouraged. However, the leads need to be used correctly to get the expected result.

Usually, an activity with a lead is linked to the completion of another activity. Practically though the trigger to start the successor activity depends on the achieved result of the predecessor activity, not the completion date. The forecasted completion date of the predecessor is a PROXY that could be used as a trigger. It could be a good proxy or a bad proxy.

The best practice is to define the required result (via a volume of work) and link the successor to this result.

This simple example demonstrates the challenge.

Task A requires completing four units. Task B takes three days and can commence after three units of Task A are completed. With a planned productivity rate of one unit per day, it takes four days to complete Task A, and Task B can commence on the 4th day. The whole project takes six days.

Such a scenario could be programmed in two ways:

Option 1: Completion date is a PROXY.

Use the predecessor completion date as a proxy.

If Task A does not progress as planned and the actual productivity rate is one unit per two days, it takes eight days to complete Task A. Task B can commence on the 8th day, and it takes ten days to complete project:

Option 2: Volume of work is a PROXY.

The successor is linked to SS + 3 units (not the same as SS+ 3 days). The schedule has the exact activity durations, start and completion dates as option 1.

If Task A does not progress as planned and the actual productivity rate is one unit per two days it takes eight days to complete Task A, but Task B can commence on the 7th day. It takes only nine days to complete the project.

A project planner can adjust lead durations manually, but during the Monte Carlo Simulation analysis, it must be done automatically based on changed project conditions.

A similar problem occurs when the predecessor activity progress is better than expected. In this case, the schedule incorrectly shows that the successor activity can commence when practically the required work volume is not yet achieved.

If the schedule with (FS – X days) leads is used for the Monte Carlo Simulation, the simulation result may be impacted as simulations don’t represent how work could actually be performed.

The forecast completion date of the predecessor may not be the best PROXY for activity leads. Often, it is the only available proxy (due to scheduling tools constraints).

The application of activity lead is beneficial but planners developing a project delivery model need to apply leads correctly.

Alex Lyaschenko

PMO | Portfolio Planning & Delivery | PMP | P3O Practitioner | AgilePM Practitioner | Six Sigma