MONTE CARLO SIMULATION CHALLENGES project changes or the volatility of the result. When applied correctly the Monte Carlo Simulation for a large project may take days. As a workaround, some consultants recommend running simulations based on a high-level schedule, also known as a level 3 schedule. Editorial remark: With the SSK we have examined this exact question! Our outcome was that 10.000 iterations are sufficient for a realiable result (given the type of projects). Level of details When each work package in a high-level schedule is represented by an activity, the schedule is missing a level of complexity of each work package. Without this information, the result of the simulation may be dramatically different. Example: A work package has five concurrent activities. Figure 6 - Distribution of High-Level Schedule. 50% instead of 3% is a dramatic difference that may drive wrong project decision. A work package may have a different number of activities, number and types of dependencies, constraints and calendars. All these parameters impact the complexity. Simulating the result of uncertainty rather than its source Usually, duration and cost uncertainties are used as a base for the simulation. However, the duration and cost uncertainties are the results of other types of uncertainties, not a primary type. Figure 4 - Example of a work package. Simulation performed based on a detailed schedule shows that it is only 3.25% (50% 5 ) of probability completing the work package within 5 days.. It requires that each activity be completed to at the most likely duration or earlier. Figure 7 - Sources and results of uncertainties. Simulating the primary source and secondary result may lead to different estimates. Sometimes substantially different. Most risk simulation tools do not work with the source data and only simulate duration and cost uncertainties. Let’s consider a simple project consisting of a single activity: • The volume of work for this activity is 800 pieces; • Activity duration depends on the assigned resource’s productivity with the optimistic estimate of 1.25 pc/hour, the most likely estimate being 1 pc/hour, and the pessimistic estimate being 0.8 pc/hour; • The workday is 8 hours; • Figure 5 - Distribution of Detailed Schedule. A similar analysis performed based on a high-level schedule where the work package is represented by a single activity shows that it is 50% of probability completing the work package on time. Uncertainty Type: Resource Productivity (+25% / -20%). Figure 8 - Three scenario’s. 00:0 7
8 Online Touch Home