The foundation of the Monte Carlo Simulation method is a project delivery model that generates possible project realities based on existing uncertainties.

For the model to be reliable, it must accurately reflect all existing constraints, including technological dependencies, resource availability limitations, and project policies. However, many projects either fail to develop such a model or do not recognise its necessity.

While there is a common recognition that base models must be of high quality, consultants often overlook the fact that these models also need to achieve a sufficient level of maturity.

A common issue is that base simulation models often ignore resource constraints or resource levelling is performed manually before running the simulation. However, during the simulation, activity durations change, which alters resource demand profiles and creates resource overload gaps. In reality, some gaps can only be resolved by introducing additional resource dependencies.
Furthermore, supply uncertainties can also affect task sequencing and lead to project delays.

When demand and supply uncertainties are ignored, risk simulation tools generate tens of thousands of non-feasible schedules, producing misleading probabilistic outcomes that appear convincing.

Simply adding an automatic resource levelling feature to a risk simulation tool is not enough. Unlike the Critical Path Method (CPM), resource levelling lacks a standardised algorithm that can be universally implemented across different systems. Each system uses unique heuristics, leading to inconsistent results.

When scheduling and risk simulation tools apply different resource levelling methods, the calculated probabilities in the risk tool may not match real-world outcomes.

Activity uncertainties

The duration of each activity is estimated as a range. Let’s consider 2-point estimations to keep it simple.

Resource uncertainty

Activities A and C require the same skill, so two resources are needed to perform them in parallel. However, whether the second resource will be available is unclear.

For this project to be feasible, project resources must be levelled.
Schedule analysis shows that:

If resources are unlimited, the project duration very between 12d and 14d.

If the skill ‘Resource 1’ is limited:

• An optimal solution is to start from activity C;
• Project duration varies between 12d and 16d.

If the Monte Carlo Simulation is performed, ignoring the resource supply uncertainty the result will be misleading.
Also, if levelling is performed but an algorithm in the applied Monte Carlo Simulation tool prioritises Activity A, the project duration varies between 15d and 19d. In this case, the Monte Carlo Simulation result also be misleading. The shape of the calculated probability curve doesn’t play a significant role anymore, as the range is wrong. 

Detailed schedule analysis is performed in the ‘Project Resource Levelling and Risk Analysis’ post.

Read more: https://saluteenterprises.com.au/project-resource-levelling-and-risk-analysis

Monte Carlo Simulation is a powerful risk analysis method, but it requires the development of a mature dynamic delivery model that generates realistic possibilities based on all existing uncertainties. If the development of such a model is not possible, the method produces unreliable results.

Resource levelling must be performed dynamically during the simulation. Not before.

Monte Carlo Simulation must be fully integrated into the project delivery system, ensuring that scheduling and risk analysis are based on a consistent resource levelling approach. Standalone Monte Carlo Simulation tools can’t produce reliable results if a project has limited resources.