Hello, my name is Ian Nicholson, VP of Solutions at Emerald Associates.
Oracle Primavera Risk Analysis (OPRA) has long been one of the industry’s most popular risk management tools.
I myself have relied on it for the best part of two decades. However, technology has come on leaps and bounds since the turn of the new millennium. It’s progressed to such an extent in fact that legacy systems like OPRA are, quite simply, no longer capable of meeting the needs of the modern project manager.
My advice to anyone who finds themselves in this situation would be to upgrade to Safran Risk (SR). And, over the next few weeks, I’ll be publishing a series of articles explaining why.
The fifth article in my ongoing series, the topic of today’s discussion is risk and uncertainty modeling: how the process is both quicker and more accurate in Safran Risk.
The Early Days of Risk Modeling
When I began using Pertmaster (as OPRA was known before Oracle acquired it in 2006), all risks were modeled as uncertainties on activities. Usually as three-point estimates.
Activities would be added to the model to simulate and gauge the impact of risk events on the schedule. Task existence would then be used to model the likelihood of occurrence. While this approach worked, it was somewhat tedious; particularly when performing what-if analysis. Specifically, it was impossible to trace the impact of risks that affected multiple activities, since the tornado graph could only show the effect of each activity, not the identity of the risk event that caused triggered the activities in the first place.
To try and improve the process, the Pertmaster team developed a Risk Register. This would allow risk events to be modeled separately from the schedule, after which an impacted risk plan could be generated by creating sub-tasks for each event.
The addition of a Risk Register worked well, as far as what-if analysis was concerned. Changes could be made, and a revised model generated quickly. Similarly, changes to the software’s tornado charts enabled users to observe the impact of risks that applied to multiple activities. But what about uncertainties?
Modeling Uncertainties in OPRA
As things stood, we were still required to model activity uncertainties manually in OPRA. If I had an uncertainty that impacted multiple activities, for example, I would have to either input the uncertainty onto multiple activities and calculate the impact durations by hand or use the QuickRisk function to generate the relevant impacts. Though, as anyone who has experience using QuickRisk will know, it was all too easy to inadvertently overwrite existing values without realizing it.
In an effort to solve the problem, the OPRA team developed a module called Risk Factors. Operating on similar principles to the Risk Register, the idea was that users would be able to add multiple uncertainties to multiple activities.
But, Risk Factors was never really completed to a sufficient standard for it to be of any practical use.
The difficulties experienced by OPRA users in regard to modeling risk events and uncertainties helped inform Safran’s approach to the process when work began on SR — just as it had with analysis by exclusion.
They had a clear understanding of the challenges project managers faced and what they required from their risk management software. Ultimately, this led the team to combine the two processes into a purpose-built module called Project Risks.
At the highest level, SR allows users to categorize all “risks” as either risk events or uncertainties. Known collectively as Standard Risks, they can be turned on and off as required and applied to as many activities as the situation demands. Their impact can then be traced back directly to the specific item in the risk module.
By approaching it in this way, Safran was able to make building what-if models considerably easier. However, to reflect the differences between risk events and uncertainties and preserve the integrity of the outputs, the way they’re modeled in SR differs slightly.
Risk and Uncertainty: How it Works
To model a risk event in SR, the user sets probability of occurrence at less than 100% and (in most cases) applies an absolute impact value to the event. By contrast, probability is set to 100% exactly when modeling uncertainties, while the impact is typically given a percentage rating relative to the original duration.
Crucially, users can mix and match probability and impact types to build a more precise model for their individual needs. They can also set pre- and post-mitigated positions for all risk events and uncertainties, which allows for great flexibility when conducting what-if analysis of mitigation strategies.
Additionally, Safran Risk allows users to:
- Configure estimate uncertainties and calendar risks
- Model cost components to individual risks (this kind of integrated cost/schedule risk analysis being central to SR)
- Map risks to appropriate activities using the tool’s Risk Mapping module, filtering and grouping activities for easy mapping using Activity Layouts
Arguably the greatest feature of Safran Risk is that the total impact of risk events and uncertainties on any activity, prior to running the risk analysis, is available in a fraction of the time. This is vital to finding errors where a risk event or an uncertainty falls outside the range of realistic values. Whereas, in OPRA, it was all too easy to assign risk durations that were out of all proportion to the original task duration using OPRA. Indeed, because of the way ranging worked in the Risk Register model, it was difficult to get accurate durations without a great deal of fiddling.
An Integrated Risk Model
Ultimately, if you want to develop risk models that incorporate both uncertainties and risk events (and most models do), Safran Risk is the obvious choice — especially if you also need to develop integrated cost and schedule risk models.
I’ll be back in a few days’ time with part six. Be sure to join me then.