Introduction to Safran Risk, the leading solution for integrated schedule and cost risk analysis.
The new standard in schedule risk analysis has arrived in the form of Safran 7.1. With its intuitive process-led user-interface and improved planning quality, your organisation can now successfully deploy quantitative project risk analysis across your projects. To help you make your project risk analysis a more transparent process, watch our ‘Introduction to Safran Risk Webinar’ now, or read the transcript below.
Lars: Hello everyone welcome to today's webinar on the new standard in schedule risk analysis Safran Risk 7.1. We are going to be presenting our product here today. But before we do we'd like to just go through some quick introductions and housekeeping. First, my name is Lars Eliassen and I'm the Chief Executive of Safran and Glenn who are you?
Glenn: I am Glenn Jarrad the Chief Developer of Safran Risk.
Lars: Excellent. That's the two presenters today. So, this Webinar is in listen only mode now. So, any questions please enter them into the question panel that you should have on your screen and we’ll address those at the end.
Today we have an agenda that includes a brief overview of Safran company and then a few slides on Safran Risk to set up the demonstration and we’ll then end with some Q&As. So, let me jump right in. Safran is a project management software company. It's been around for 20 years. We have users averaging our tools all over the world in complex projects in asset intensive industries. And you see some of those verticals here on the screen including oil and gas, chemicals, infrastructure project aerospace, defence, and utilities. And we have two main product lines, Safran project: which has a full suite of project controls tools for planning scheduling and executing large projects, complex projects and then we have Safran Risk: which is a quantitative project risk analysis tool and the topic of today.
Just to kind of give you an idea of who or who is using this software. You will see a variety of companies in different industries here. Quite a few in oil and gas. There are owner operators, there are EPCs, there are suppliers of various sorts, engineering companies, EMC companies, basically companies that are planning and executing large complex projects. So, with that introduction I'm going to hand over to Glenn who's going to give you a little bit of slides to set up his demonstration of Safran Risk. So over to you Glenn.
Glenn: Thank Lars, as Lars said just a couple of slides to start and then we'll get straight in to into the product.
So, the focus with Safran Risk was to create a tool that's easy to use and although like other tools on the market, even more powerful, and with this simplicity sort of built in if you like. So, to that end, the user interface that we've designed is hopefully very intuitive as you'll see shortly. And it adopts a process driven approach.
We've also included lots of powerful features to make the simulation as efficient as possible and to allow us to do some very advanced kind of scenario planning and dynamic, innovative reporting that we've added into the product over the last couple of years. Also within Safran Risk is a fully featured scheduling engine.
I mentioned there has a complete scheduling environment built in. I think we all know how difficult it can be to get a schedule all ready for Risk Analysis. It's probably the biggest challenge that it’s ever going to face to have a suitable quality schedule. So, by having an embedded scheduling environment we really sort of significantly reduce all that upfront time by removing the need to constantly round trip between say p6 or Microsoft Project and our quantitative tool because we can tweak the schedule and make any changes necessary directly within Safran Risk. Not only that but we can easily create multiple versions of the schedule which helps with what if and scenario analysis makes it a very simple process.
So, moving on to project risk intelligence. It's something that I guess we have held dear and fundamental since we started the development of Safran Risk. As a risk professional, I'm sure you'll be very aware of the standard risk distribution histograms and tornado charts generally available, and sure enough we have obviously provided these standard reports that you would expect to see in a tool. But additionally, we've tried to sort of innovate, and bring in new reports, to add a lot of value to risk analysis. So, we've got things like an advanced distribution Esker of comparison to help with those scenarios we discussed earlier. We've also got automatic sensitivity analysis that allows you to calculate the risk impact in actual days and dollars, and some other reports that I'm going to mention later.
So, the first commercial version of Safran Risk was available around about 18 months ago, and the latest version that we have on the market now is 7.1 and is our sixth major release. So, as you can tell from that we're very keen on having a fast turnaround of products and being very reactive and responsive to any requirements that we see in the market.
Some of the features that have been added to this latest release include mitigation. So, we can easily model both pre and post mitigating scenarios. And on the reporting side we've added a new scatterplot report to help schedule cost and year risk trending reports.
You can track how the project is performing over time because we now have a database behind us so we can track progress over periods of time in different stage Gates rather than just being a simple sort of file base system. This gives us a lot of advantages in terms of tracking progress over time. We've also made enhancements to sensitivity analysis and added some prioritization features, and in terms of the Gantt chart, we've allowed you to push back lots of information into the Gantt. So, you can see where various percentiles of bars are against the deterministic and the ability to bring in sort of frequent estimates that you've added into your host scheduling tool, whether that's p6 or Microsoft Project.
Right. So, this is Safran Risk and as I mentioned on one of the first slides a key objective really was for us to provide a simple and straightforward process for you to follow. The last thing you want to do is hunt around for four different parts of the risk process. So, to that end, we have a very much left-to-right process, all with tabs that we have across the top, so we build the model up starting from left-hand side with the schedule and then we move on to all the other inputs into the model until we hit analyse. Then after we've done the analysis we move on to look at all the various outputs from the analysis.
So, starting on the far left with the home screen that we are on now you can see the key thing we need for a quantitative risk analysis, a schedule, that schedule can come from our project tool, or can come from P6, or can come from Microsoft Project, or from a variety of different sources. And on the right-hand side you can see some global data so we can run Safran Risk in either a sort of a standalone environment or also in a in a server environment.
So, if we if we run in a server we can create global risks. And they can import them into our individual projects. So, it's information you can share across all the projects. We also have global risk calendars and the way you create calendars in Safran is like the way you create calendars in other tools but we've also extended the capabilities and made it far more advanced than other tools. So, whilst you can create where the calendars are used using similar template environments and tools you can also use the real-time series data that's provided by various organisations, and import that straight into the product to create your winner calendars from that data.
So, that's some global information. But let's move on. So, we're going to start by looking at schedule, and this schedule is one of our samples. This is just the simple prototype, so it has several phases: preliminaries, design, fabrication etc. and what you're looking at now is our schedules environment that I mentioned. So, as you would expect you can insert or add new activities and link them. It has full resource capabilities full calendars. It is as I said a full environment that you can tweak however you need to.
I'm not going to spend too long on the schedules because I really want to focus on the risks side. But obviously if you have any questions feel free to ask those now, for us to answer later. The most important thing really from a risks perspective is that that the quality of the schedule is critical to be in how to run a good risk analysis. To that end, Safran Risk is constantly checking the scale for you and whenever you leave the schedule tab it is automatically checking it for you for several different things that could affect the schedule. Obviously, this is things like constraints, particularly hard constraints, but also lags, and you can configure this to your own corporate standards if you like and they all get reported in this schedule warnings area.
So, on this example you can see that we have a couple of open ends. So, I can just look at just the open items of which we have three. We have a couple of lags which are not going to worry about too much and we'll see the constrained activity with a must finish on constraint. So, that's not going to be very good from a risk point of view. So, I'm going to click straight on that. And it takes me directly to this agreed contract task where I can simply remove the constraint in this case. Then you'll notice here there's a switch back, It's reanalysed that project. Now I only have the two warnings. So, I should fix that one so it's very handy because it's automatically and it's very interactive and allows you to sort of get through those potential problems quite quick.
Right. So now at this stage we've got the schedule we've validated it and we know the quality of it is good. So, can move on to start talking about risks.
So, this is our project risks tab and this contains all the risks that might pertain to this project. And by all the risks it might also be a calendar risk, a poor flying conditions risk, that could affect our testing of us of our drone prototype. How much downtime each of these Calendar risks represent.
We also have standard risks for the kind of risk events that you have that I'm sure you're all very familiar with. This example is a specification design change and has a probability of occurring of 30 percent. And if it should occur on the right-hand side we can define very easily what the impacts will be to that risk so those impacts can be either on the schedule. And/or on the cost as well. So, this example design specification change has a schedule impact that was defined as absolute and by that we mean it's going to impact in a specified or between a range of actual values will that's days or indeed dollars.
This one is going to have an impact according to this triangular distribution. Of course, we support many different distribution shapes to cater for any needs you might have. And we're saying it's going to impact between 6 and 10. Could be is as far as 12. So, when we come to map this design specific action change to activities in the schedule during the simulation that's going to add between six and 12 days according to this distribution shape to those activities.
It also actually has a cost impact as well and a cost impact in this case is defined is relative. So, relative basically means it's not a hard number. It's a percentage value. And in this case, it's going to obviously affect the cost by percentage of the activity we assign it to. So, if we matched this to an activity that should make the maths easy you know a thousand-dollar cost on it then that cost is going to vary sort of between 1000 thousand in 1950 during the simulation. We also have both these things up. These are pre-mitigated positions. We also can define our post mitigating positions as well. So here we've said in our post mitigation situation we're going to reduce the impact of that risk down to 10 percent. We're also going to reduce the impact down to a maximum of five days. And, the cost impact has been reduced as well. So very easy to define those as well.
The final type of risk we have is estimate uncertainty. So, estimate uncertainty would be what is traditionally thought of as three point estimates and Safran Risk takes a risk factors approach to assigning that kind of estimate uncertainty. So here we've defined three different levels of estimating uncertainty; low, medium, and a high. And then when we would come on to mappings you'll see we can easily assign that to certain levels of the schedule.
So, move on in and look at that now. So, risk mapping is probably the most important areas within in the product and possibly where people will be spending the most amount of time. We have this schedule on the left-hand side in the same format it is in now. You saw it here on the schedule. So, there are various phases preliminary design fabrication etc. And across the top we have our different risks. So, to the left here with the radio buttons we have our estimate uncertainty categories that we've created. And again, could create as many of these as you like I've just created 3 for this sample. And then we also have our kind of more traditional risk events listed further on the right-hand side. And then in the centre it can say that we have kind of a matrix of tick boxes and radio buttons that we can use to easily assign the risks to various levels of the schedule. So how do we do that?
Well very simply if we look at an activity called design additional guidance systems and we can see already that this risk has a low risk uncertainty assigned and a construction workforce productivity risk assigned to it. We could also see that here on the right-hand side. And at the bottom what we get is something that is unique to Safran is we get our real-time analyser is giving us data that allows us to calculate precisely what the impact is of these two risks on this one activity. It does that by simulating every time we select in here it's running a mini Monte Carlo it's running 1000 simulations immediately, we check the box to see what the impact will be, that's a unique feature that we have, and it allows you to see the impacts of design additional guidance systems – here it deterministically is 30 days. As we go through and make more assignments you can see that it is immediately reassessing the impact to see where the deterministic sits.
In other tools, you should run a full risk analysis to get me kind of data but in Safran Risk we get it immediately, and the assignments allows us to see if we're sort of the double counting anywhere or anything like that. I think the final thing to say about risk mappings is that if you're dealing with particularly large schedules then it's easy to just create a filter in the schedule to filter down to for example 1 WBS. So, we might want to fill just the construction tasks and any filters we create in here, of which we can basically use any data that's associated with the schedule to create a filter. If you apply the tools to the set it filters the whole product which is a powerful feature and allows us to sort of just perhaps focus on that WPS. But for the whole of the risk product.
So now at this stage we've loaded our schedule. We know it's of sound quality. We've created all our risks and come back to correlation's if we have some time but we could create simple correlations between all our risks very easily. We've mapped them to the relevant elements of the schedule. And now we're ready to run our risk analysis.
So, in the analyse tab we can set up several different capabilities of running the analysis. We can select we can select which kind of risks we want to include. So, if we wanted to just run a scenario where we're kind of mimicking having totally mitigated a few risks we could do that easily. We can also look at a pre-mitigated scenario by default, or we can look at post, for all our risks or again uniquely in Safran so we can specify a combination thereof, so we can look at the pre-mitigated for the majority of our risks but maybe look at what happens if we mitigate just these three risks – It gives us a lot more power in terms of looking at different possible scenarios we might wish to model.
In the centre, there's several options for the analysis. Things like ‘number of iterations’ you can obviously change, you can specify a seed, include correlations, and if you are using resource levelling in your schedule then we will also honour that resource levelling. And if you create your algorithm for resource levelling in a schedule then we can honour that during the actual risk analysis as well.
We can step through to look at what the model would look like during each iteration and on the right-hand side we can specify focus activities. Now these are very important when we want to look at the reports and the output because rather than having to sort of look at thousands of different activities, if we have large schedules, we can set up our focus activities here then we can very easily switch any of the output reports to just look at only those focus activities. So, they're really the key milestones that we that we care about and that we want to track.
So, let me go ahead and run the analysis. So, we have our standard distribution graph and we're currently looking at the likelihood of hitting our deterministic finish date for the overall project. We can also look of course at that cost as well, and duration, we can look at all these reports that I'm going to show today for the cost as well as finished date that you can see here. I’d just run out of time to show you everything.
So, what this is telling us, if we look at the information on the right-hand side, is that following from that risk analysis and all the risk we had mapped to our schedule deterministically, we were due to finish around the end of September 15 is currently 12 percent likely. Based on those risks that we have, if we're tracking an 80 percent level, then we're looking more like sort of 17th of December.
We can say here we can draw curtains between any of these different percentiles that we set up and you can set up multiple percentiles. Whichever ones you're interested in you can very easily set up and create these kinds of curtains that show you very graphically the difference between say the deterministic in this case and the P80 or whatever it might be.
You can obviously look at the data for any activity immediately or any WBS or I mentioned the focus activities so it's very easy to just switch across and then we can remove all the noise and just look at our key milestones to see how likely we are to hit any of our forthcoming milestones.
It's very useful information, we really want to know how likely we are to hit either our key milestones or our overall project finish date. But the next thing you really need to know is what can I do about it?
So, for that Let's move on and look at our drivers. A tornado view is the way we present this information in Safran Risk – on the left of the screen you can see those risks that are driving the finish date of the project, and on the right-hand side you can see the activities driving the finish of the project. So, in this example, we can see that ‘test fails’ is our biggest risk, followed by ‘governance’, followed by ‘merger acquisition’ etc. On the right-hand side unsurprisingly given that ‘test fails’ is our biggest risk you can see there's multiple testing activities that are most likely that see or see the section on the WBS of our projects that are most likely to cause a delay from an activity perspective.
If you want to look at individual key milestones using those focus activities, then it's very easy to instantly just switch and say which risks or activities are driving me to my next mile stone? So, to fabricate engine for example, we can say that governance is our biggest risk to that milestone, and agreed contract is the biggest activity that's driving to that focus activity. Hopefully you'll agree that that's very useful and powerful information. But at the same time. It doesn't show anything we can easily report on. So, you know ‘test fails’ is our biggest risk and it has a correlation coefficient of 0.70 and for governance its 0.46 but it’s not great information to actually report.
So, for better reporting data we can run our sensitivity analysis and what this is doing, as you can see, is drawing all these curves. Each time it draws a curve it's rerunning the entire analysis, through all thousand iterations, and each time it pops out one of the risks. So, the first time maybe it popped out the ‘test fails’ risk, then it reran the entire analysis then popped out ‘governance’, rerun entire analysis for ‘flying conditions’ risk etc.
So, by doing this it can easily calculate the exact impact that these risks are having. In terms of the number of days’ impact they're having. And, the cost impact as well. So, sure enough ‘test fails’ on our project is our biggest risk. And it is followed by ‘governance’ and then ‘poor flying conditions’, but we can report that ‘test fails’ is going to cost us 20 days on our schedule, ‘governance’ 13, ‘poor flying conditions’ 10. So, we can include these weather risks as well. And we can look at what kind of an impact our weather risks are having on the structure as well.
To fully mitigate ‘test fails’ we would say we’ll sell 20 days so it gives us the ability to perform a cost benefit analysis on that. And of course, our focus activities are here as well. So, we can again instantly change to look at our next milestone which risks we need to really worry about and mitigate to the next important milestone that we might be facing on our project.
And, after I've set up three percentiles to look at here P20 P50 and P80 as I said can set those up to whatever you like. But we can just instantly switch and look at different percentile and how much these risks are driving at that given percentile value.
So, moving on to one of the new reports in in the latest release the scatterplot. This is showing us, as I briefly mentioned earlier, the joint confidence level so you can see it's split into four quadrants here. The lower left quadrant, with the green dots in, is showing us where we've been during those thousand iterations and where we finished, and how many times we finished within both cost and within time. Unfortunately, in this case the red dots in the upper right quadrant are showing we came outside of both cost and schedule. And is we look at the percentages in the corner, 79 percent of the time, we were outside both, and only 8 percent of the time we're inside both.
Now we can look at this data again, and use it to focus activities or any individual activity we like and we can also play around with the numbers a little bit so rather than looking at the deterministic value we can specify our own target. So, we know it's only 8 percent to a deterministic but how about if we were to target term target Christmas Day? If we invest this additional cash of 1 million, we're now actually 82 percent likely to hit that date and only four percent likely to be outside both. So, it's easy to just play around to see the impact to both. So, that's one of the new reports that we added in this release.
The other new report is the risk trending. So, our distribution comparison is effectively a report controls area so we can easily compare individual distributions, or, we can look at trending over time of projects. So here is one I've obviously prepared earlier where we have in this case you have six different stage gates on this project. This is our first review cycle where we obviously weren't particularly likely to hit our deterministic finish and we can see that the blue is tracking on the P80 the grey is on the P50 and the black is on the P20. So, as we move through each review cycle hopefully we've managed to do something about the risks that are impacting the project. And we can see particularly as we get to our letter stage gates these move very heavily in a downwards direction giving us this kind of curve that we can see here. Now this distribution comparison can not only be used for comparing sort of the way risks are trending over time but it can also be used to compare individual scenarios, any scenario you like, so if I go to reports area and it does create a brand-new report. I can go to my current distribution graph which analysed the whole project and I can send this curve over to that report. And let's call this time it's called this all risks.
And then it can be easy just to go back to analyse a different scenario without those risks being analysed. So, I’m totally mitigating those particular risks. And I send that across to comparison. And we can immediately see the difference between the ‘all-risks’ curve which is in blue and where we've managed to mitigate a few of our risks we’ve saved twenty-three days on our schedule. And if there was a cost impact we would also see that in here too. So, that's kind of the sensitivity analysis gives you an automatic way of looking at all of this and using distribution comparison we can either view different phases of a project and how it’s progressing over time or we can compare lots of different scenarios, as many as we like and can even compare distributions from different projects if we want to is nothing to stop this doing that.
And finally, the critical path map, this gives us a whole different view on the critical path. So, most products have a criticality column. So, you can see how critical each activity is which is quite common in various risk tools. With Safran Risk though you can also view all the paths through the schedule and how critical each path is. So here we can say we only have two paths through the schedule and path one is critical 92 percent of the time. Path two only 8 percent of the time. And the graphic to the right is showing us where on the curve those iterations on that path are completed. So, as you'd expect with a path that’s ninety-two percent critical that it's occurring pretty much everywhere across the distribution. Whereas if we just look at path two, you can see of course there are few iterations that are showing on this path is only 8 percent of them. And quite a lot of them are down here as well in the bottom left hand corner. So, any kind of paths that are focussed down the bottom left we don’t need to worry about them too much.
So, I know that's a very quick demo but that's what I wanted to get through today. Obviously if any questions please feel free to mention them. I think here are a few slides now to sort of close on.
Excellent. Thank you, Glenn. If you’ll give me just one second here on the screen. Back to our PowerPoint. So as Glenn said please if you have any questions. Go into the question panel or the console you have on the right side of your screen. We'll get to those later. So, here's one more slide to talk about the future direction.
There is indeed so. Thank you. We are very keen on developing this this project well into the future and so we have a very healthy roadmap planned, and we want to innovate in this space. So, one area where we see a real opportunity to do exactly that, is in using analytics and optimizations to help answer the real questions behind why people often run a quantitative analysis. Things like what's the best mitigation strategy given a spend of X? I’ve got this pot of money with 10 million dollars, what areas should I focus on in terms of my mitigation strategy? So, we want to buy some advanced techniques to allow you to look very easily at that. And, something we're often asked about really is that he's the qualitative side so we are also very keen on taking the approach we've taken to the quantitative and sort of really simplifying quantitative analysis and extending that also into the qualitative side to make it much more straightforward for people to carry out of a qualitative analysis as well. And the final point on that really is that organizations do have a variety of tools at their disposal to create their own kind of analytics dashboards. So, we want to really help share all the risk data that we have within it all and make it easily accessible to these various analytics products that people have.
Lars: Thank you very much Glenn. So, as you've seen here today quite an elaborate product with a lot of functionality and hopefully you will agree that it's very user friendly and intuitive. We really worked to make sure that this was this was a powerful but user-friendly product. We have more things in store for the future but we’re very proud of some of the things that we've done here on the reporting side, and hopefully you will agree to have a chance to trial the product itself.
The built-in scheduling engine gives you the iterative route you analyse capability that is much needed to refine is a process to do this analysis and cut down on time that's spent there going in and out of import/export those tools too quickly to accomplish this today in other environments. And then you have the schedule checker that is key to making sure that you have the highest quality plan that's required to perform this type of risk in-house. And one thing we didn't really go into a lot of detail today is the risk calendar – the approach that we've taken there is to allow you to model weather using real time series data, and it’s an interesting technique. You can go to our YouTube channel and have a look at the video there.
Finally, for some of you that are in transportation we can do risk analysis to the minute which in some industries is important, especially when you have a very short duration, such as a compressed project or shutdown of a train station or something or a weekend. Where you really need to get down to that level of detail.
So, that's a summary of some of the things we talked about today. At this point we’re ready to take questions. First one: How easy is it to upgrade from Permaster licenses to Safran licenses?
Glenn: There are two sides to that, there is a commercial and a technical side and commercially I know we’ll make it very easy to help to upgrade from existing Permaster licences to Safran Risk licences. We also have several people who are doing exactly that and running comparisons against the two and have either found that you know we’re getting the same results but are delivering them faster. That’s the main point.
Lars: Another commercial question here which we did mention but it's on our website now I believe. We just launched a new website. The question is what is the cost?
Well the list price is five thousand U.S. dollars per person end user and has had a 20 percent annual maintenance. That's the easy answer for that. I guess that's it for now. Those are big questions we have received so far. Thank you very much for your attendance. I look forward to seeing you next time. Have a great day.
For more information about Safran Risk 7.1 you can visit the brand new Safran website here.