• Projects are often forced to apply predefined scheduling and risk analysis tools. The companies that run training usually focus on existing features, not on missing functionality. Well too often, planners and risk managers have good knowledge of the corporate project management tools but don’t actually know schedule and risk management. • If companies that promote their risk management or quantitative risk analysis training say that Monte Carlo Simulation requires many hours of effort, no one will buy their tool or service. • Precision of this method creates an illusion of accurate estimation. Some consulting and construction companies use it to justify desired decisions. It is not critical for them to understand how the method actually works. Quality of analysis is defined by colours in reports, not by accuracy in calculations. Terminology What do the terms optimistic, pessimistic, and most likely really mean? Because they are very different for different people, when one estimator says pessimistic, that person is thinking the technology might prove to be a bit more difficult than expected and thus take 10% longer. Another person in the team may assume that ‘pessimistic’ means that the work might be interrupted by an earthquake, tsunami, and nuclear meltdown simultaneously (even if it has not happened in the past). Data collection Popular scheduling tools, like Microsoft Project and Primavera, don’t have native features to capture ‘3 points estimations’, so planners usually don't capture estimations provided by Subject Matter Experts. A risk manager or scheduler applies a range based on a single deterministic estimation instead. This approach is based on two dangerous assumptions: • All provided estimations are ‘Most Likely’ estimations When Subject Matter Experts are forced to provide a single estimation, they have to decide on how much contingency to include in the estimation. Some estimations in schedules have ‘optimism bias’, others, on the opposite, are too conservative. • All activities have the same level of uncertainty There are many reasons why some of the activities may have dramatically different uncertainties: new technology, new equipment, new contractor, etc. Distribution shapes Risk simulation systems offer a large menu (usually up to 30) of user-selected distribution shapes that can be implemented for each and every activity. However, selecting a specific shape for each of 1,000 or more activities (on top of generating three estimates for each item) requires many hours of invested time. The majority of users run the system using one of the default distriFigure 3 - Distribution shapes. The ‘3 points estimation’ known as a PERT methodology is relatively simple to use for cost, but for the schedule, it’s more complicated. The fact that some activities are on the critical path means that the volatility of those specific activities, as opposed to the set of all project’s activities, can make a big difference. The time impact of any risk is dependent on the criticality of impacted activities. If activities are on a critical path, the risk is likely to impact project delivery dates. Changes in critical path mean that time impact is not a stable characteristic of risk. The impact has to be calculated during the simulation, not predefined in a risk register. It could be achieved if an analysed model includes probabilistic branching and conditional scheduling. Volatility of results How many iterations are sufficient for a reliable result? Hundred, thousand, million? Unfortunately, there is no formula to calculate the answer to this question. It depends on the number of activities and other parameters such as the level of complexity, quantity and type of dependencies, risks and uncertainties. The volatility means that result of a simulation that is based on precisely the same inputs could vary, sometimes significantly. Risk simulation tools usually have a default number (usually 1000) of iterations. Users don’t know how many simulations are required to reach the desired level of volatility and leave the default number as is. While 1,000 simulations may sound a lot, in reality, often it is not near to the required level. When after some time the project team runs a new analysis, they have no idea why the result is different. It could be due to butions: the triangular, PERT or a beta-distribution. However, different activity distribution shapes can substantially alter the result of the analysis. The result of the simulation performed with PERT or triangular distributions at the P80 mark is likely to have a noticeable 815% difference. Beta-distribution is often recommended as a good alternative to these two distributions. While it sounds very scientific, each risk management tool uses its own formula of beta-distribution and analysis performed with the same initial data and the project delivery model will vary based on the tool applied. Risk data 00:0 6
7 Online Touch Home