Monte Carlo simulations of project schedules have become one of the foundations of quantitative project risk analysis. Monte Carlo method helps to determine the chance that project will be completed on time and on budget, expected project cost and finish time given risks and uncertainties, as well as identify critical risks and crucial tasks. There are a number of ways how Monte Carlo schedule risk analysis can be conducted. “Traditional” Monte Carlo schedule analysis is performed based on statistical distributions of task duration, cost and other input parameters. Event-based quantitative risk analysis incorporates risk events, which can affect project schedules. The chapter discusses a number of important concepts related to Monte Carlo simulations: statistical distribution, sampling process, convergence monitoring, sensitivity analysis, probabilistic and conditional branching and others.
TopIntroduction To Monte Carlo Schedule Risk Analysis
Monte Carlo simulations of project schedules have become one of the foundations of quantitative project risk analysis (Salkeld, 2016, Vanhoucke, 2016, Wanner, 2013). Monte Carlo method is used to approximate the distribution of potential results based on probabilistic inputs. Each simulation is generated by randomly pulling a sample value for each input variable, such as task duration or cost from its probability distribution. These input sample values are then used to calculate the results:
- •
Project duration,
- •
Start and finish times,
- •
Success rate,
- •
Work,
- •
Cost, and others.
This ‘traditional’ Monte Carlo method for schedule risk analysis is based on statistical distributions of task durations, cost and other parameters, and has a number of short comings. Particularly defining distributions is not a trivial process. It is difficult to elicit distribution parameters from subject matter experts. Also project managers perform certain recovery actions when a project slips. These actions in most cases are not taken into account by Monte Carlo (Williams, 2004).
The one of the solutions is to combine risk events with statistical distributions. Project risk analysis with events has been used since the early 2000s (Virine & Trumper, 2007, Virine & Trumper, 2013). This approach is sometimes referred to as “Risk Drivers” (Hulett, 2009, 2011). From the computational perspective, using statistical distributions and risk events are very similar.
Event chain methodology is an extension of “traditional” and event-based quantitative risk analysis. Event chain methodology is an uncertainty modelling and schedule network analysis technique that is focused on identifying and managing events and event chains that affect project schedules.