When we plan a project, we need to understand that construction in contractor's schedule is only a portion of the overall schedule. They have to do with construction productivity, labor agreements, means, and methods and so forth. But on top of that, there are certain assumptions in their contractual requirements that need to be met. For instance, one thing is to build a facility. And you can have it built in 12 months. But depending on the acceptance requirements, the acceptance of that facility to transfer from the contractor's responsibility to the owner's ownership may take additional time, 30, 60, 90 days, sometimes more. So that time has to be part of the overall project schedule, and typically is not in the contractor's schedule baseline. In addition as you think of this project from the owner's perspective, you have design activities, you have procurement activities, you have permitting activities, you have financing activities that build up at the beginning or at the end of the construction that definitely increase the duration of the project. And on top of that, if you think about a project by itself, that baseline does not have any risk. On top of that, we need to understand, and we need to have an agreement of how we treat elements that may or may not happen, and plan for them. And those are what we call contingency float. A very simple bar chart schedule I wanted to show of a facility plan that has equipment to be delivered, configuration of rooms, installation of equipment, roughing, wiring, terminations. At the end of the day, there is a red bar that says, third party testing. And that third party testing has a risk to it. It has a potential of being late. It's scheduled for 20 days. Third-party testing could ideally last 20 days. Now, if the equipments are tested and do not perform, they have to be replaced, and they may have to be modified. And they may have to have work above and beyond and be re-tested. So this third-party testing that was planned to be 20 days now become anywhere from 20, ideal scenario, no risk, or 50 days. How do we plan for that? Because if it lasts one day more than 20, it delays my substantial completion. So understanding that these risks exist and are applicable to activities during the schedule, it's the first part. But knowing how to plan for them, it's designs that will be discussed and will be the criteria of the owner or the project manager in this case. Now in the past, we had limitations on technology. So the use of expected value was very common. We identify risks. We said there is 30% chance of exceeding the 20 day testing that we talk about before. And the most likely scenario is that we will exceed it by ten days cause we've done testing before in many different facilities similar to this one. And typically something fails, we don't know what, something also fails and it fails and takes us ten additional days. But it only happens 30% of the time. So on that practice, what we use to do, was take 30% chance of happening, multiply that by ten days, and then add a contingency to testing of three additional days. What was the problem of that? 30% of the chance this risk happened and when it happened ten days were delayed. We budget for three, so in 30% of the cases, we were short for seven days. And we were not able to cover the additional testing day, almost never. So that practice fade away and got replaced to more sophisticated practices. [COUGH] In the current practice, in the current times, computer packages have become more and more accessible, cheaper, and faster. So we use risk modeling for this and we create MonteCarlo simulations that allows us to model the different scenarios and pick a confidence level. Let's work through what that concept means. This is a project that had a substantial completion date of March of 2020. And had a number of risks, including that testing and commission risk. Now, in a simulation model, you run the schedule one time and say, out of 40 risks 10 happen. And this 10 risks have an effect here, here, here in your schedule. So now this one time, this project did not finish in March, it finished in July and we make note of that. Then we simulate the project the second time. And now instead of 10 risks happening, 15 risks happening. And when those risks happen they happen in different places, different than the ones we did in the first pass. And now the project did not finish in March, it finished in October. So we make note of that. When we do that process 3,000, 5,000, 10,000 times, we have 10,000 points in this curve and we start developing this histogram. And if you accumulate them, we get a similar cumulative curve like the one shown in the graph. Now, that cumulative curve has a minimum, because the project will not finish soon than a certain day. And has a maximum, right, will not finish later than a certain day. The places in between are associated with probability of a curious. So, 80% of the records that we noted, 80% of the 10,000 cases, 8,000 cases fell before January of 2021. 70% of the cases that we plotted finished before December of 2020. 10% of all the cases we plotted finished before April 2020. So, we can say what is the percentage of all the different scenarios that I want to budget for? And that's what we call a confidence level. So in this case, if I want to be sure that my schedule, the date that I pull this to the revenue service will cover 80% of all the potential scenarios when we put all these risks in the model. That date is January 2021, not March 2020. And that's where we as planners, have to make a decision of what is the day that we want to advertise. And if we take a contingency bar in our schedule that runs from March 2020 and January 2021, and say that's my contingency. And every time we get a risk, that gets us delay, it consumes that contingency and that's how we control it. Because we have a model now, we can sort which are the elements, which are the risks that have the most impact on this mode,l that push our schedule to be further delay. And we can plot that in something we call the tornado graph. The tornado graph will give you relative impact of the top scheduled risks. What that's going to do is that it will allow you to know what you could mitigate further, in order to avoid setting such a long contingency in your plan. So let's go to one example of how this works. The risk here is that key materials are not available in turn, to support the project. And that risk will affect the pile work, the foundation work. 60% chance of that happening and if it does happen it could delay the project anywhere from 60 days to 120 days. Now when we run this scenario in the project schedule, we notice that the pile work could be delayed at the beginning. Not only during the construction but the start of it. So the best case in this scenario had the pile work starting in January 2018 and finishing in October 2018 for nine month duration. And after we put the risks on it, 80% chance or or 80% of all my scenarios, did not exceed. I start in February 2018 and finish by December 2018 and give us a total delay of one month. So now, we can see this one bar that could be an activity or could be a project, that has a range of start and a range of finish. And if we define a confidence level, like in this case. The 80% confidence level is that the project will not finish in October like the no-risk fund tells us but is more likely to finish in December. We need to make a decision. We need to make a decision for how do we advertise this project? Do we call December 2018 our completion date and have a contingency part at the end. That's where planning actually takes place, but it gets in form by this exercise, this risk assessment exercise. And that's what explains how risk assessment supports the planning process. Well I hope these examples can give you a more tangible application of risk management. I want to finish with a few remarks. [COUGH] Risk management is here to support decisions, inform how to make decisions with limited information and when the information is uncertain or lacking. It's based on a systematic approach, as you have seen. And it's traceable, and it's transparent. And in this stage in time, transparency is a very big deal. It's very important, especially when you're talking about public monies. And the use of, public assets to create a engineer or a construction project. Risk management is more than a good management practice. It's a real tool that enhances your chances of success in a project. In the sense that we define it. It has to follow a formal process. The alternative is just failing us in the past. So it has to follow a formal process to be systematic, disciplined and holistic. It's a rational process. It's an approach that you can follow with a framework. But you can follow rationally because it follows the life cycle of the project. And it allows you to manage your exposures on the objectives. Whether those are costs, schedule, reputation, functionalities and so forth. And it assists project managers and project owners on decision making. Lastly, I wanted to leave you with two resources for you to investigate and get more familiar about the frameworks of risk management. One is PMI. The practice standard for project risk management. That's very widely known and used. And the second one is what we follow. It's ISO 31000. It's a robust framework and it's application is very easy. And it could be applied for all engineering and construction projects that you may find in your career. So, good luck, thank for watching, and good day.
