Why is there a difference in my risk assignment view

September 9, 2015 Q: Why is there a difference in my risk assignment view We have several projects in planning and we noticed that in on project, the risk assignment had only one Outcome, but in another it has Low, Outcome, and High fields. What is going on? A: RiskyProject has a very flexible risk assessment capability and one of these is the ability to either assign single outcomes or outcomes as a distribution (low, base, high with a statistical distribution). The latter is generally considered a more advance feature and you can use it when you want to analyse a range of possible impacts on a risk. For example if a risk occurs, you think it could impact your costs, but it could be anywhere from 5 to 20% of your activity budget. In this case, you could use the model this using these ranges along with a distribution type (uniform, triangular, etc.). To enable the statistical distributions for risk outcomes, go to the Project Options> Risks. Click the Enable Statistical distributions for risk outcomes. Likewise, you can disable this view in the same dialog. If you plan to assign outcomes with both single or distributions in your projects, we recommend that you set the default with the statistical distributions for risk outcomes enabled. In the cases where you do not want to have a distribution, you can enter the same value in the Low and High outcomes to get the same result. http://www.intaver.com/IntaverFrm/viewtopic.php?f=3&t=1117

BetaPert statistical distribution for Monte Carlo

September 4, 2015 Q: BetaPert statistical distribution for Monte Carlo We have been looking at alternatives to triangular distributions for modeling uncertainty in our tasks. One of the alternatives that has been mentioned is BetaPert. In what cases should it be used as opposed to triangular? A: Both BetaPert and Triangular both require 3 pt estimates for Low, Base, and High. BetaPert is a variation of the original Pert distribution that was developed in the early 1950s to provide probability distributions for project management purposes. BetaPert is generally used as an alternative to Triangular Distribution in cases where the estimators believe that there is less possibility that an activity will have a duration that occurs at the boundaries. In practical terms, PetaPert generates a distribution with lower standard deviation and that is shown by a more probability density around the calculated mean. Lets take a look at a simple example and compare the Triangular and BetaPert distributions with a 3 pt estimate of 3,5, and 8 days. What does this mean? BetaPert is less sensitive to the boundary estimates (in this case 3 and 8 days) and returns probabilities that closer to the original most likely estimate. In very simple terms, if you have that that suggests the low and high estimates have a very unlikely probability of occurring, BetaPert may be a better alternative than Triangular. A point of interest. Some recent studies that we have seen seems to indicate that while BetaPert is appropriate in specific cases, the boundaries estimates are generally two narrow. http://www.intaver.com/IntaverFrm/viewtopic.php?f=4&t=1107

Number of Monte Carlo simulations

September 6, 2015 Q: Number of Monte Carlo simulations We are just starting our project risk analysis process and would like to set some standards for the number of iterations. Are there any guidelines for this? A: First, you can set the number of iterations for the Monte Carlo simulations in the Project Options>Calculations. There are really no standards for the number of iterations to set, but here are a couple rules of thumb. If you require quick estimates and are looking for results closer to the mean such as standard deviation, you can use the convergence settings as found in the Monte Carlo parameters options. When convergence is on, RiskyProject monitors the simulation and it will stop the iterations when the mean and standard deviation are not changing significantly. If however, you have a more complex project with risks with low probabilities or are looking to measure the results at the boundaries, you will need to run more iterations for Monte Carlo simulations so that the analysis captures the impact of unlikely conditions or events. In this case, you may want to run up to 3000 iterations which is set using the Maximum number of simulations box. As the latest generation of software and hardware is much faster, this is less of a concern than it was even a couple of years ago. http://www.intaver.com/IntaverFrm/viewtopic.php?f=4&t=1108

Cost accrual for risk analysis

September 8, 2015 Q: Cost accrual for risk analysis We have just started a new process and are doing some integrated cost and schedule risk analysis. Some of the results we were seeing for costs in the Cash flow view looked odd until we noticed that the cost allocation was set for the end of the activities. We had to manually go in and reset them all to Prorated which is our standard. Is there any way to set this as the default? A: In RiskyProject you can define when costs accrue in tasks. The costs can be associated with resource usage (prorated cost accrual). You can also place cost realization at the beginning or end of the task. This might be used for alignment when the costs are realized by billing systems and it can help you manage cash flow in your project. Cost accrual is very important for integrated schedule and cost risk analysis. For example, if risks related to the cost occurred at the middle of the task and cost accrual is prorated, the risk will affect only portion of the cost, accrued after the risk occured. RIskyProject has probabilistic moment of risk property of the task, which controls when risk occurs. The standard default settings for cost accrual is prorated, so I am not sure how this may occurred in your case. If you imported it from a third party tool, it may have overwritten the defaults. Regardless, you can set the default accrual methods for all of your projects in the Project Options -> Cost. The accrual method is set using the Default Fixed Cost/Income Accrual drop-down lists. http://www.intaver.com/IntaverFrm/viewtopic.php?f=4&t=1111

Q: Risk properties and risk analysis

September 9, 2015 Q: Risk properties and risk analysis Does the Start and End dates in the Risk Properties view impact the risk analysis? Particularly, in the Risk Properties view, does the risk Start and End Dates affect the results of the Monte Carlo risk analysis? A: Quick answer is no, the Start and End dates of the risk (sometimes referred to as sunrise and sunset) are part of the meta data used as part of the risk management process, but is not used as a input parameter in the Monte Carlo schedule risk analysis. Risk does have a time dimension in addition to probability and impacts; however, when you assign risks to activities or resources, this time dimension of the risk is automatically accounted for by the time phased aspect of the tasks within the project schedule and resources assigned to them. http://www.intaver.com/IntaverFrm/viewtopic.php?f=4&t=1114

What is the best distribution to use for schedule duration?

August 31, 2015 Q: What is the best distribution to use for schedule duration? When you add low and high durations to an activity, what type of distribution is the best? thanks A: In reality, there is no “Best” distribution to use for schedule or cost risk analysis. The statistical distribution for any parameter (cost, duration etc) should be the one that best fits your data. If this data does exist for particular type of activities, then you can select the distribution which best fits your data or you can create a custom distribution. However, in reality, most of our users do not have this data available for the type of activity. In these cases, we suggest using Triangular statistical distribution as the default. Triangular distributions only require that you provide Best Case, Most Likely, and Worst Case values without requiring any additional parameters. In our experience, triangular distributions are used in 90% of schedule risk analysis and to a lesser degree in cost risk analysis (as the this data is more readily available.) http://www.intaver.com/IntaverFrm/viewtopic.php?f=2&t=1092

Statistical distribution for duration

September 4, 2015 Q: Statistical distribution for duration As part of our Monte Carlo schedule risk analysis, as a first draft we are not yet requiring our team members to provide estimates for low and high duration. We want to import our schedule from Microsoft Project and quickly add uncertainties to all of the activities. I seem to remember that you can do this in RiskyProject, but I can’t find it. A: In RiskyProject you can quickly add uncertainty to your tasks. Uncertainties are expressed as statistical distribution and will be based as an input in Monte Carlo simulation process. These statistical distributions will be used for task duration and task costs. We recommend as a first step develop levels of uncertainty each with coefficients used to calculate Low and High estimates. For example, you can If you have activities that are very routine and have very little uncertainty, they can be characterized as Very Low risk with coefficients for Low and High . 95 and 1.1 respectively. You can do this for all your activities in Microsoft Project and when you import it into RiskyProject, this will allow you to quickly select specific risk groups to apply the uncertainty. To apply the uncertainties to groups of activities. Select the tasks, right-click on the Task ID of one of the selected tasks and click Set Low High Duration from the shortcut menu. You can enter the coefficients you have defined as well as assign a statistical distribution type to the group. http://www.intaver.com/IntaverFrm/viewtopic.php?f=2&t=1103