ClickCease
Lorem ipsum

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua.

Beat Volatility With Stronger Contingency Plans

 

1/26/2022

59 Minutes

Request A Demo

Thanks for contacting us. A member of our team will follow up with you shortly.

Transcript

Lance Stephenson:

Well, hello, everyone. As we let some of the participants start trickling in, I just want to welcome you all to another edition of ACCE’s Webinar Series. Today, we have a discussion on beating volatility and, of course, why strong contingency plans today mean more project wins tomorrow. This is excellent topic that we can see today regards to some of the experiences we’re having around the world, specifically around COVID, supply chain management, things like that.

Lance Stephenson:

My name is Lance Stephenson. I’ll be the host today, and we’ve invited InEight to have discussions around this specific topic. Today, we’ll be talking with Rick Deans and Nate St. John, and I’ll turn it over to you guys to introduce the topic as well as introduce yourselves. Prior to that, though, guys, if everybody could move their questions or have their questions in the Q&A box. Please don’t use the chatbox for your questions. Just move them over. That’d be great, and I’ll be looking at the, see if I can introduce them nicely with the topic, but if not, we’ll be answering questions at the end. With that said, I believe we talked about all the housekeeping rules. Yeah, let’s just… let’s begin.

Rick Deans:

Well, that’s great, Lance. Thanks for the introduction. We’re going to take a minute to really introduce ourselves, and as we do so we’ve got a poll that we’d like our participants to respond to. We really want to get a feel for who’s in the room, so if we could bring up those first two poll questions, that would be great at this time. We’re really interested in knowing what best describes the type of organization you work for and best describes your role within that organization.

Rick Deans:

My name is Rick Deans. I’ve been working with the InEight set of tools I used to round up for credibility to 20 years. Now, I round down for other reasons to 20 years, but I’ve had the opportunity to work with dozens, if not hundreds, of our customers in real life as they go through some of these challenges across a number of industries. I’ve listed several of them there. There’s probably one or two I forgot about.

Rick Deans:

I’ve also worked, and I don’t want to say this too loudly, but I think it’s kind of humorous. I’ve also worked with our customers on every continent except for Antarctica, not that I’m keen to go there, but have been around a little bit and, like I say, over those 20-some-odd years had the opportunity to talk to a lot of customers. We’re going to brink up some of those anecdotal interactions in our presentation today. I’m going to pause for a minute and let my colleague, Nathan St. John, introduce himself as well.

Nate St. John:

Well, hi, Rick. Good to see you. Lance, thanks for having me again. My name’s Nate St. John. I’m currently the Product Director for Planning, Scheduling, and Risk at an InEight. You know, I started as a general laborer. Spent several years doing the work firsthand in building from the ground up. I was then on the owner’s side as a construction manager for a few years, but it’s really been the last 10, 12 years that I’ve had a heavy focus on planning, scheduling, and risk.

Nate St. John:

I spent several years working with teams across North America as a scheduling risk advisor, and now my role at InEight, coming up on just two years, is to really evaluate the market and steer the direction of product development in order to better the role of planning, scheduling, and further CPM mechanics in the industry. There’s a lot to improve on and it’s an exciting time to be at InEight in innovating with some of these gaps that we’re filling,

Rick Deans:

Looking at the poll results. It looks like we’ve got a good mix across contractor engineer owners. That’s exactly the markets we serve. We’ve got cost engineers, estimators, project managers, schedulers well represented as well as business leadership and management, so that’s great. We welcome you all. We hope that this is going to be as exciting for you as it is for us. Let’s sort of get on with it.

Rick Deans:

We’re going to introduce sort of a bit of an agenda. We’re going to talk a about the alignment that typically occurs in the pre-construction phase. Maybe we’re an owner and we’re thinking about creating a budget for a particular project and we’re looking at it through a particular lens. Maybe we’re a contractor and we’ve been invited to bid on a particular project. Maybe we’re an engineer we’re working for a client of ours that is interested in maybe a feasibility study or a good understanding of what a budget might be.

Rick Deans:

We’re going to talk about alignment between the ability for the cost and the schedule groups to share data. We’re going to talk about risk adjusting a cost estimate. We’re going to explore really, what sort of risks are we looking at? Why do we have those risks? How to basically cover those risks through either mitigation strategies or having a contingency bucket in our estimate. Then, what we want to do is we want to take it a step further and validate those risk outcomes against our own historic data.

Rick Deans:

One of the things that we’ve certainly seen in the industry is, especially now as Lance had mentioned, going back about two years, things got turned upside down. Currently, about 93% of general contractors tell us that they face at least at one major material shortage per project, so every project is affected by material shortages. We’ve seen inflation. Certainly recently, we’re looking at 7% inflation, record inflationary pressures and supply chain issues are going to affect both cost and schedule.

Rick Deans:

One of the things that Nate and I discussed as we were preparing for this, we were both sharing our collective experience working with our customers and working with organizations in their offices. One of the things that we’ve uncovered is that it’s very difficult to model and compare various risk mitigation strategies in real time. The collaboration, sometimes it’s there, sometimes it’s not. Sometimes people working on the costs are over here mitigating cost risk. People over here in the schedule group are mitigating schedule risk and rarely do they cross paths.

Rick Deans:

Then, sort of the piece on this is that because of time constraints and the way we organize and collect data and the way we report on that data sometimes hinders this process, especially for items like risk and contingency. Usually when we’re thinking about benchmarking, we’re thinking about direct costs. How many workforce hours did it take for us to put in the form work for a certain type of concrete pour? Things like contingency and risk tend to be a little bit more amorphous, and we’re going to discuss that as well. Nate, any comments you’d like to make on these four pain points that we’ve brought up?

Nate St. John:

Well, yeah, maybe on the third, fourth one, Rick. Starting with what you just finished with, as kind of corny as it sounds, a lot of data is generated on these mega construction projects. I mean, every decision made, everything that’s brought into a technology platform is a data point. We have to be cognizant worldwide that there is a large sector of the labor market that are actively retiring or set to retire soon, and this is just an example. With those talented folks that have been building work for 40, 45 years, they have a lot of experience and they have a lot of history.

Nate St. John:

It’s important for us, especially at InEight, to provide a repository where we can capture all of that feedback and expertise, and I think like you said, Rick, we’re probably at an industry better at capturing cost and schedule items, but furthermore, realized risks, risks that continue to hit us across various programs consistently. More importantly, realized and vetted mitigation strategies that have been able to cover some of that exposure. All of that still is simply data points in the world of technology that we have an opportunity to capture, and once we capture, the tools that are available become extremely powerful in mining that data and exposing that data and then essentially guiding teams down a stronger realism, for lack of a better word, a stronger realism score the next time they go and build a subsequent project.

Rick Deans:

I think you nailed it, Nate. There’s no shortage of data that’s being collected. It’s just the question of, how is that data being collected? Is it being organized? Is it being normalized? Is it being combined with other data where people can make decisions? Or is this data in silos? That’s sort of the next thing we’re going to talk about. What are some of the obstacles that we see to effective contingency planning? One of the things that I certainly see is that in some cases, the cost group and the schedule group, they might as well be on different planets. They’re siloed. Maybe they do share information at a very high level, but let’s face it, that triangle of project management, cost, scope, and time, you can’t affect one without affecting the other two. That siloed work process that we see that exists between cost and schedule groups.

Rick Deans:

Another thing that I’ve certainly seen, and Nate, before I go onto the third slide, I’d be interested in your thoughts on this, but I’ve sat in a lot of meetings around the conference table where it seems like many organizations have little consistency in terms of how they define things like risk and contingency. As the project starts being executed, what am I able to use that contingency bucket for? Many organizations don’t have a really well-defined list of dos and don’ts. I’m curious, Nate, from your perspective what you’ve seen there.

Nate St. John:

Yeah. It’s an interesting evolution here for arguably a topic like risk or contingency, maybe less familiar, or maybe if it is familiar with project team members, maybe it’s more of an uncomfortable topic. We’re seeing an evolution now from teams executing consistent risk and contingency per project basis, but what the ask is across the globe and what all of our clients are now asking for is, “How do I aggregate those up to some of these mega programs that span either a corporation or a sector of a corporation?”

Nate St. John:

It’s taking it to the next level, and by rolling up, capturing this contingency definition, the risk of experience across various projects, you’ve naturally developed what we consider as a global or a program knowledge base risk register where you can go and you can pick. Five years ago, I was in the Middle East operating on an infrastructure project in this type of contract model. The software can point us and hone in on, “Hey, did you realize with that set of criteria that you’re explaining to me right now, these were the things that have popped up in the past?” The evolution is there. We’re starting to see it and I’m just happy that today we have technology that can answer that demand.

Lance Stephenson:

Nate, Nate, sorry, just to build off of that, too, AACE has a number of recommended practices that support the definitions around risk and contingency estimating systems, also historical database development, and so everything you’re talking about, the foundation is provided also by AACE in the recommended practices. That will help support any endeavors that companies might have as well as what you’re doing, so I just wanted to let people know that we have that option as well.

Nate St. John:

Yeah. Great plug, Lance. I have AACE folder up on my computer with all of the recommended practices. I’ve been an accredited member for several years and, yes, it’s [crosstalk 00:13:02] a fantastic –

Lance Stephenson:

I would be [crosstalk 00:13:03]-

Nate St. John:

… way to start.

Lance Stephenson:

Yeah. I would be amiss if I didn’t give a plug to AACE, so [crosstalk 00:13:07]

Nate St. John:

Yeah.

Rick Deans:

And maybe [crosstalk 00:13:08]-

Nate St. John:

Well, we could do some sort of a [crosstalk 00:13:09]-

Rick Deans:

… that’s the answer when we’re sitting in these boardrooms, right? Have you guys-

Lance Stephenson:

Yeah.

Rick Deans:

… have you guys preview review the AACE guidelines on these things.

Lance Stephenson:

Yeah, so anyway, sorry. Continue on.

Rick Deans:

No, I appreciate it. That’s good, that’s good. You know, we hear this term “optimism bias.” Sometimes it’s called cognitive bias, so maybe we are an owner and we’ve got a project we really want to see go through its stage gates and see the light of day. Maybe we’re a contractor and we’re putting together an estimate for a project we’d really, really like to build. Maybe we’re an engineering firm and we’d really like to see this owner go ahead with this particular project. Maybe subconsciously we’re building in some optimism bias into our original budgets.

Rick Deans:

What do I mean by optimism bias or cognitive bias? It’s that idea that, you know, “Hey, we’re looking at structural steel and historically it’s taken us somewhere between 40 and 55 workforce hours per ton to handle and erect structural steel. Well, this project’s going to be different. We think we can really get away with 35. We really think we’re going to beat that, that industry average, our own organizational average.”

Rick Deans:

Sometimes we have to take a step back and just ask ourselves, “How much of this optimism bias are we really putting into these assumptions?” At the end of the day, a budget is just that, it’s a series of assumptions. How much of that is based on our desire to see this project go through versus the reality and what you’re really seeing in the real world. Nate, I know this is a subject near and dear to your heart. You want to add anything there?

Nate St. John:

No. I think you said it very well, Rick. You know, cognitive bias, it’s a real thing. I think that if you were on a job site and you were having a critical conversation with a coworker, they probably call it rose-colored glasses and you’re going to be bullish and force your way through it. You can lump that all under optimism bias or cognitive bias. Sometimes it’s not even a strength of attitude. It’s just a subconscious thing that we-

Rick Deans:

Yes.

Nate St. John:

… we trend towards, so the idea here is that versus realism, realism is a set of algorithms that sit on well-founded statistical principles that you can at least vet yourself against.

Nate St. John:

Now, yes, there’s going to be people out there that says, “Well, an algorithm’s biased because a human had to create it.” Well, I’m not going to go there. I’m telling you that if you’re in the heat of the moment, if it’s last minute, bid submittal, or a final investment decision, you’re going to have naturally some kind of bias, and so the ability to plug that plan into something that is data and calculated-based, it’s only going to assist you in terms of keeping you in the lanes for whether you’re marching towards realism.

Rick Deans:

Well said, well said. Another thing that we’ve just seen organically, maybe more anecdotally, but it seems like this whole feedback loop. I’ve estimated something’s being built in the field. How long did it take as well? You know, guess what? At the end of the project, project teams scatter like the wind. They’re on to the next big one and sometimes we don’t get that feedback loop. Maybe the tools we’re using are just too cumbersome to support that. Maybe I’m tracking things at a very, very low level of detail in the field using my array of Excel spreadsheets and maybe an access database, but that doesn’t make its way back into the corporate ERP system.

Rick Deans:

That whole feedback loop can sometimes be cumbersome with tools that might be disconnected and, again, this goes back to this sort of ties into the lack of available validation via real data, that realism piece we talked about earlier. In our world, we call that benchmarks. Do we have good benchmarks for this?

Nate St. John:

Rick, it might be a good time, and I think this was probably just overlooked, but if we go back to the disconnected tools comment on this bullet, just to briefly outline who InEight is, we are a full end-to-end project management platform for capital expenditure projects. We have eight main technology stacks and underneath each of those stacks fit business processes. I want to do estimating. Well, you can get in and use InEight Estimate or a quantity takeoff.

Nate St. John:

I want to do scheduling. Well, underneath this scheduling stack, you can do preplanning, CPM, short interval planning. It’s eight products tied together that sit on a unified platform and share data so that it’s almost a one-stop shop and a user can come in, land into the InEight suite, determine what business practice they’re set to do for that day, and InEight provides the solutions to do so.

Rick Deans:

Excellent. What we’re going to do now, we’re going to ask you folks to answer another poll question. We’re interested in understanding what sorts of risk tools you’re currently doing, using, and what sort of risk analysis is currently being used in your organization.

Rick Deans:

As those results are coming in, let’s talk a little bit about the role of risk assessment. Again, pre-construction, we’re putting together budgets, we’re putting together schedules. We want to know how much of what we’re putting together could be impacted by risk. Some of the questions we want to ask ourselves, when do we perform a risk assessment? Nate, when’s the best time to perform a risk assessment?

Nate St. John:

Oh, today [crosstalk 00:19:04].

Rick Deans:

Is it-

Nate St. John:

Today.

Rick Deans:

… an event? Or is it a process? Maybe another way of stating that, right?

Nate St. John:

Yeah, it’s today. You can do it at any time, the earlier the better, and then once you’ve established or have moved on to execution, it’s built today as a process that can be democratized across all team members and at the end of the day be a very iterative, quick assessment. The answer is, when to perform it? You perform it today.

Rick Deans:

Yeah, absolutely. What types of risks should we be considering?

Nate St. John:

Well, yeah, so this is a good question and it’s a big question. Most certainly at the project level, you’re looking for really two inputs. You’re looking for what we call uncertainty or a range of plus/minus 10% on this cost item or this scheduled duration. The second part is discrete risks, things that are external or out of our control, things that are typically maintained in a risk register. Those two types of risk set the foundation for a risk assessment.

Nate St. John:

There are also things that I just myself categorize as commercial risks, so perhaps reputation hits, environmental safety things. Things that might not necessarily be able to be discretely mapped to a cost item or a schedule item. Those things still need to have some kind of qualitative analysis and be incorporated in the entire sphere of risk as you move throughout the life cycle of the project.

Rick Deans:

Oh, that’s great, and we’re kind of looking at the poll results here and it looks like most of you were doing a risk analysis that includes both cost and schedule. It looks like, surprise, surprise, Excel is the market leader in terms of risk analysis tools being used out there with this particular audience. Well, let’s take a look at some of these other questions we want to ask ourselves as we’re performing a risk assessment. How much contingency should I add to an estimate to cover risk of exposure? Of course, the answer to that’s going to vary. It depends. Is this an early-stage, rough order of magnitude estimate? Is this a pre-execution estimate? Hopefully we’ve driven that contingency down by getting more details about the project.

Rick Deans:

This is actually something that Nate and I are going to use the tools to demonstrate. We’re going to come up with an idea of what the contingency should be. We’re going to do some risk mitigation techniques. We’re going to do some modeling. We’re going to see how that result compares with what we had come up with earlier. I think one of the things that makes InEight unique is the ability to, then, validate that against our own organizational history.

Rick Deans:

We’re going to jump in and do a little product demonstration here. This is a screen shot I’m going to drive a little bit in one of the applications called InEight Estimate. This is just really a screen shot of how we’ve come up with some contingency based on applying some percentages and how that fits into our overall project cost. I’m going to flip my screen over here and I’m going to bring up our application InEight Estimate.

Rick Deans:

This is a project that I’m putting a budget together, I’m putting an estimate together for, and I just wanted to show a couple of things off. The tool itself is fairly malleable. You can create a hierarchy of items in your estimate that roll up and roll up and roll up to an ultimate parent that ultimately rolls up to overall project costs.

Rick Deans:

We’ve also got data fields that our customers can use to help them categorize and define things, so out of the box these might be labeled tag one, tag two. I’ve created some data fields here for work types, the estimator responsible for that, and you can see I’ve even as I’m putting my initial estimate together. I’ve got a data field here for the level of risk that I feel is associated with particular line items in my estimate. Just like any other row and column-formatted tool, I could filter this down to see my high-risk items. I could scroll through this and see which ones pop out at me. I could even group this where I’ve got it grouped by… I can see dollars subtotals based on my perceived risk level.

Rick Deans:

The way I’ve come up with this contingency at first blush is really this is what we might call a derived cost or a dependent cost. What I’ve said is I want to account for some contingency in my estimate and the way I’ve done that is by looking at certain costs that I know I’m going to incur in the field. I’ve got some labor costs, I’ve got some equipment costs, some subcontract costs. What I’ve done is I’ve applied percentages against those cost categories. In this case, the calculations taking place, and I’ve got about $91,000 of risk that I’ve accounted for in my contingency bucket.

Rick Deans:

Now, what I want to do is I want to take this a step further and I want to run some scenarios and perform some risk mitigation techniques. This is where I’ve handed it over to my colleague, Nate, who’s going to take this information and he’s going to interpret it in InEight Schedule.

Nate St. John:

Thanks, Rick. Are you seeing my screen?

Rick Deans:

Came across fine.

Nate St. John:

Perfect, so what everyone’s viewing here is the cost estimate from InEight Estimate, now in InEight Schedule’s cost risk application. You can see right away, this is Rick’s deterministic contingency. There’s his $91,000. The first thing I’m going to do is exclude that because I don’t want it to be picked up by the simulation. The quantitative risk calculation will provide a required contingency value in which then will compare back to this deterministic nine 91,000. Rick had mentioned that he had flagged a few items as high-risk, and so we’re going to go down in this notes field here and just identify X and backfill 3.1.2. This was an item that he in the estimate process said, “Hey, this is a high-risk.” I’m just going to go up and I’m going to sign it a distribution of very aggressive.

Nate St. John:

What that means is is you have a range here from very conservative to realistic to very aggressive. When I assign it as just a first pass, because this is very much iterative, very aggressive, it assigns this triangular distribution with a minimum of a hundred, a likely of a hundred, and a maximum of 150%. We’re going the first of two things. We’re assigning uncertainty and let’s go down and say, “All right, oh, here’s another high-risk. Let’s go ahead and tag that as very aggressive.” I think there is one more.

Nate St. John:

This here says, “Very aggressive.” Now, Rick also pointed out that there’s a discrete risk to consider here. We also at the same time have an opportunity to add an event, and the event in this case is a risk because it can be multitude of things, but we’re going to add a risk. The activity, or excuse me, the cost item is X can hold a site, so perhaps we’ve got some unknown utilities and the description is just underground. We’ve added this risk. Now, we need to populate the data associated very commonly to a risk. We can come in and say, “Listen, there’s probably a high chance of this risk occurring. If this risk does hit, it has a medium impact on schedule, but it has a very high impact on cost.”

Nate St. John:

We exit out and now one of two things have happened. We’ve now mapped, we’ve created a net new risk, discrete risk. We’ve mapped it to that cost item, and then at the same time, it has populated automatically into our project risk register. If I take us to our risk register, you can see we had two existing risks. Now, this under known utility with its values is pre-populated into the register and that is by design. We want to make sure that we are blending the process of standard risk matrix maintenance siloed with a schedule, siloed with a cost. We developed this so that you can create and gin up risks while you go through your planning and evaluation process.

Nate St. John:

Through our experience, we’ve realized that the content and the quality of those discrete risks are much more intimate and applicable to those items when the process is done together. We’ve got three bits of uncertainty. We have one risk mapped to an item. What we’re going to do now is we’re going to report out let’s just say at a P75 and run a Monte Carlo simulation. We’ll run a thousand simulations. It’ll take a couple of seconds and we’ll be able to report out first pass, “What is our risk exposure histogram?” Here’s your standard histogram. Unfortunately, Rick, we have a 0% chance in this model of hitting our deterministic base cost. What I mean by base cost is that $10.25 million, it’s the total cost just minus that contingency.

Nate St. John:

More important, let’s focus on our target of P75. If we want to be 75% certain that we’re going to hit our target goals or beat them, we need to target that P75 value. What we would have to do to achieve that level of confidence is cover $186,000 in contingency. The simulation in the first pass with our risks decorated into the cost estimate is kicking out 186. We were originally at 91. I pushed this over to Rick and I say, “Rick, how do you feel about adding a hundred percent on top of your existing contingency at the moment?”

Rick Deans:

I think that’s going to be tough to do. Business leadership is interested in getting this project off the ground, but I don’t think there’s an appetite for basically doubling what we’ve already set aside for contingency, so maybe we need to explore some mitigation strategies. There were some questions that popped up and I would [crosstalk 00:29:30] be amiss if I didn’t say that I didn’t … One of the questions was, “What is the tool being demonstrated?” This is InEight Schedule.

Lance Stephenson:

How did you guys manage them prior to doing this? We talk about systemic risks. That’s usually underestimated a lot of times. We talk and. you know, that could be handled in peak inspection. Do we scope enough out? The other issues is allowances. The other one is escalation. You know what we’re dealing with around volatility, most of the time with volatility comes down with escalation. How does the system manage that? How does the cost estimator look at allowances versus contingency? Then, also, how does the estimator deal with escalation [inaudible 00:30:14]?

Rick Deans:

Yeah, and I think I’ll start with that last one first because we do have an integration with schedule applications, including our own InEight Estimate or InEight Schedule, rather. We do have the ability of pulling dates back into the estimate, so we have date fields associated with our line items in the estimate. Now, we have a visibility into not only what is the work going to cost us and what sort of resources are we going to use, but now we have that time-phased view of that information as well. Multiyear project, we could actually identify which line items in the estimate are scheduled to occur after Jan. 1, 2023, where we expect materials rates to go up by X percent, or maybe we’re doing something with a particular labor union and we know it on April 15th, those boilermaker rates are going to go up by 17%.

Rick Deans:

We can actually have that visibility in the estimate as to when those costs are going to hit. Because it’s integrated as we’re moving things around on the schedule, as we introduce new constraints, as we’re pushing out activities, those schedule bars are just sliding into new escalation ranges, which are doing to help us understand what sort of escalation we should be looking at. Then, the other question was differentiating between contingency and allowance, those are just separate what we call cost categories here in the estimating system. It’s just a question of how a human being wants to put money in one of those buckets versus another one of those buckets. In a perfect world, Lance, people would be adhering to the AACE guidelines to differentiate between risk and allowance and they’d be parsing those dollars out accordingly.

Lance Stephenson:

Yeah. Okay. Thank you both.

Nate St. John:

Yeah, great questions, good interaction. I’ll pick it back up, Rick, unless there was as another-

Rick Deans:

Absolutely.

Nate St. John:

… one you wanted to. Okay, so-

Rick Deans:

As long as I don’t have to double my contingency amount, I guess, but-

Nate St. John:

That’s real world life. This is an iterative thing. Usually it’s popping between a couple different groups, but when I facilitate risk workshops, I try to get as many of the critical players on both sides as possible because it’s just the right thing to do. Rick said, “No, we have to mitigate,” and so the system will tell you very quickly, “Well, what areas are top contributors that are exposing this cost risk exposure?” You can run this. It’ll rank order and you can see a couple of things here. We’ve got three line items in the cost estimate. There’s a clear elephant in the room here, and it’s this exit haul to site. You can see about $3,000 of that risk impact at the P75 is about 3200 bucks. That’s categorized as uncertainty, but the bulk of it, this hundred thousand dollars hit as a risk.

Nate St. John:

What risk is it? We can obviously if we had numerous risks com ein, but no surprise to us, it’s risk one, our underground utility risk. We want to make a decision to begin to analyze cutting that back, so we go down this cost benefit analysis approach to find the ideal tipping point or the medium point in which we can pump enough money in, but we want to make sure we’re getting enough savings on the back end. How we do that is we can come in and we can go find that risk. Our risk is right here. On the risk tile, we’ve got this option, add mitigation, and it pops up this thing and you can add in, let’s say, add potholing.

Nate St. John:

What we’re going to do on this activity is we have all of these underground utilities. We don’t know where they are. Let’s add in some additional potholing to kind of see and map out where existing utilities might exist. By doing that, we’ve lowered substantially the probability of us hitting any kind of utility. Now, it’s important to know that even though it’s a low probability, if we do hit a utility, it maintains the same impact in this example as a medium to our schedule duration and still a very high-cost item. We can plug in here quickly, “Hey, it’s going to be $2500 real rough to perform this mitigation. I’m going to give it to Rick as the owner and I’m going to say, ‘Rick, there’s a due date on your mitigation strategy by tomorrow.'” Just some extra tracking mechanisms here. The most important thing here is that now you can see the title of that mitigation against that risk is ad potholing.

Nate St. John:

There’s one more thing we have to do before we can tell the system we’re ready to run a mitigated scenario analysis, and that’s to come into this badge and we can set one of three statuses in the risk world. It’s either unmitigated, it’s mitigated, or it’s closed. We’ll go ahead and assign this specific risk, a status of mitigated, and it assumes the values that are associated with that mitigation strategy, which are naturally going to be lower. This is important because those are the values that the Monte Carlo system will pick up when we rerun the analysis.

Nate St. John:

If we come up back to the top with mitigation turned on, we can run our analysis. We can report out the same way and we can see, “Okay, our P75 contingency has dropped from 180-something down to 107.” We’ve covered, I don’t know, $70,000 of drawdown against that exposure by just adding $2500 worth of potholing. The important thing here to communicate back to Rick is that the deterministic original contingency was 91, so now is it more digestible to add a few thousand dollars to get up to 107? By doing so, you’ve added a third dimension to this risk analysis, and we call that confidence. You are 75% confident statistically that if you add a few more thousand dollars to get to the 107 required contingency that you’ll hit your target numbers.

Nate St. John:

At that point, you can see this is very iterative, back and forth. People are going to run pre, post-mitigation strategies, all of that. This is a very simple example. I’ll stop sharing and kick it back over to Rick so you can see, all right, now we’ve done a quantitative risk assessment. We’ve gotten a P75 contingency value. How does that stack up in terms of our historical benchmarks?

Lance Stephenson:

Does this total number provide you with escalation numbers as well?

Nate St. John:

This provides you with whatever you’ve decorated into the cost estimate, so what we [crosstalk 00:36:57] usually see with escalation is as a first pass, it’s some kind of percentage, any kind of growth factor, any kind of productivity factor, anything that’s a percentage-based is usually best modeled as uncertainty. You’re going to drive that three-point triangle based on whatever way you want to weight it. In terms of escalation, you can weight it how you want so that it naturally will ramp up. That’s how you handle escalation in the risk world, but you’re obviously going to go into much more detail with this number and more suitable in InEight Estimate after this [crosstalk 00:37:33] quantitative has taken place. Yeah.

Lance Stephenson:

I just wanted to make… I’m trying to connect the project to volatility in regards to the marketplace and escalation is usually is the key around that because it’s derivative of the CPI, and there’s artificial issues that plague inflation that we’re experiencing right now. I just wanted to understand a little bit about the escalation. Thank you for that.

Nate St. John:

Yeah, absolutely, and we can tie it back to volatility again. If there’s a discrete event, if there’s some kind of risk of a major piece of equipment that’s supposed to be on a barge coming from overseas, if that doesn’t get delivered, that’s not a ramp-up impact. That is a a discrete, “We’re going to stop work for X number of months until that piece of can get on a barge successfully and brought into your country of origin.” There’s multiple ways to model that. Volatility, I guess is what I’m saying is it can be both modeled with uncertainty and/or discrete risks.

Lance Stephenson:

The other question, though, around that is when I look at this around the escalation clause and how we deal with that, what about, for instance, you know, we look at labor rates and the increased labor rates, but maybe we can’t get the [inaudible 00:38:56] on the job. I’ve been in projects where we actually couldn’t get, say for instance, a welder who can weld. Think of that. You got a 16-inch weld. Also, you got to bring him in [inaudible 00:39:07] you got to bring him in from the States. How do we… I mean, how do we manage the lack of resources? Not necessarily the increase of prices inside the door.

Nate St. John:

Yeah, so-

Rick Deans:

That’s… Go ahead, Nate.

Nate St. John:

Well, I was just going to say from the schedule perspective, you would model it just like what we discussed, but then there’s an opportunity to take a model and analyze your resource histogram. You can take, let’s say, the P75 version of this schedule and look at the histogram and you can easily say, “Am I over or underallocated the target number of personnel that I need in order to hit the productivity that’s planned?”

Nate St. John:

I know there’s a lot of tabular kind of cost-related things that Rick can do very well on that front, but I would also point people to also managing resource-loaded schedules and comparing those histograms from a baseline, a snapshot, in multiple risk assessment scenarios across the board. That’s one route. You can go multiple directions with this, but from the schedule front, at least, that’s how I would tackle that.

Rick Deans:

Yeah, and I was going to say we would have that visibility. When are those specialized Inconel welders supposed to be on-site? What can we do to to make sure that happens? That’s a very specialized skill. Obviously, if we’re down in Southeast Texas and there’s one on every street corner, that’s a different scenario than if we’re up in an isolated province in Canada where maybe there’s not so many of those folks up there. Those are things that we could have the visibility to know, and let’s take a look. Now, Nate has told me that this 91,000 is probably insufficient, that we’re going to have to bump that up to a buck oh seven. Very easy to do. We can start fiddling with these percentages. We can even target a value, so if I wanted my contingency to come in at $107,000, I can just hardcode that in.

Rick Deans:

That’s going to then come back to me and say, “Okay, that means I’m going to be looking at a certain percentage of what that means relative to these costs. Another thing I wanted to demonstrate here, I’ve actually got some line items in the estimate for escalation, and maybe we’ll just take a look and see how these are calculated. I’ve got $209,000 set aside for escalation in this particular project. What I’ve done is I’ve said, “I want that escalation to kick in on a certain date and I want it to run it through a certain date. Then, I want another layer of escalation to kick in on Jan. 1, 2022, and have it run all the way through the end of 2022.

Rick Deans:

These percentages can vary by cost category, so if I wanted to escalate my materials, even if I wanted to take it a step further and define which line items in the estimate I wanted to be affected by this, because we have an understanding of when the work is going to be performed, it’s very easy for us to slice and dice and add escalation data ranges to particular costs that we’re expecting to incur.

Rick Deans:

Then, ultimately what we want to do is we want to come in and we want to take a look at our history and we want to see historically, what has our contingency spend been as it relates to a percentage of certain costs in the estimate? These visualizations can help us. We can see right now I’m sitting at about 15% contingency. My average might be a little different. I can actually see historic data points that show me what we’ve estimated and where contingency has actually come in. This data is available in graphic format. It’s also available in tabular format. Even if I wanted to go through and include or exclude specific data points, I could do that, and then the graph is going to update for me dynamically based on which project’s I’m including in my analysis.

Rick Deans:

What that’s going to let us do, then, because this is tied with the schedule, we’re going to be able to run the risk analysis, and then we’re going to be able to understand not only those benchmarks and how our current situation relates to our history, but we’re able to see what that spend may look like. If we have a normal or a Gaussian distribution of that contingency, we can see when we’re really expecting to drawdown the most of that contingency.

Rick Deans:

Now, I’ve got other customers that say, “Well, we don’t expect to draw down contingency in a normal fashion. We expect we’re going to hit… Maybe we’re going to do a turnaround at a refinery or at a plant, and we expect most of that contingency is going to be drawn down earlier rather than later. We can even model and track how much contingency we’re planning on drawing down over time. In this case, we’ve sort of frontloaded that, and then we can actually compare that with the actual drawdown of contingency as we’re experiencing that. This allows project team members to really see, “Are we drawing down a lot more contingency than we had planned? How does that affect our forecast going forward?

Lance Stephenson:

Around that, then, if the schedule is tied to the estimate and, for instance, we do have a turnaround where we might assume more risk. Should that not be tied back to the schedule that would give you the distribution which would actually not allow you to condition the loading of the contingency? [crosstalk 00:45:00]-

Rick Deans:

Yeah, that’s [crosstalk 00:45:01] exactly [crosstalk 00:45:01]-

Lance Stephenson:

… anyhow, right?

Rick Deans:

… came up with.

Lance Stephenson:

Okay.

Rick Deans:

That’s exactly how we came up with these metrics, right?

Lance Stephenson:

Okay.

Rick Deans:

We defined a cost curve. We said how much of that contingency we’re expecting to spend early versus later, and based on when those activities were going to be performed based on the way they were scheduled, that’s what gave us this district.

Lance Stephenson:

You can go back and look and say, “Listen, I don’t have contingency in my turnaround. I need to move some of that over, or I need to… Well, I need to reassess my contingency [crosstalk 00:45:24]-

Rick Deans:

Absolutely.

Lance Stephenson:

… and my contingency [crosstalk 00:45:26]-

Rick Deans:

Yep.

Lance Stephenson:

… on the turnaround itself. Okay.

Rick Deans:

Absolutely.

Lance Stephenson:

That’s good because it allows you where you have gap analysis, too, in regards to the level of Hartford and the priority and the criticality of some of the work efforts, so…

Rick Deans:

If you’ll notice on the vertical, on the horizontal axis, this is daily. We’re expecting a certain amount of contingency to be drawn down daily. If you were relying on ERP or Excel spreadsheets, you might not be able to get feedback as urgently as we might need it, right?

Lance Stephenson:

How does this tie back to change management and earned value? What I mean by that is, say for instance, the project. It’s supposed to be 50% complete. It’s only 40% complete. It’s earning CPI at .9, so therefore what you’ve done is you’re not getting the work done as much as you can, therefore… but for instance, you’ve got changes in place that have drawn down more at that particular time.

Lance Stephenson:

How do you reevaluate contingency as you move forward based on your earned value? How much you’ve got completed versus all the changes. I’ve seen projects where, based on the change management, they spend 60% of the contingency when they’re only 40% done with the work.

Rick Deans:

Exactly.

Lance Stephenson:

Right? How does the system look at that?

Rick Deans:

Yeah. No, great question, Lance, and we handle that a number of ways. During execution, you’re absolutely right. Project team members can see, “Hey, we’re performing at a .9. Our earned over burned ratio is not where we want it to be.” That lends into the forecasting discussion. How do we want to intelligently forecast the remainder of the work? Maybe there’s a specific one-time situation that occurred. We couldn’t get that piece of equipment off the barge in Nate’s example, but we’ve mitigated that. We’ve brought in something else and now expect the rest of the work to be performed exactly the way it was planned. Great. Maybe it’s an ongoing thing. Maybe productivity rates… Maybe we’re hitting we’re hitting weather delays. Maybe we want to model what the remainder of that work looks like.

Rick Deans:

In our tools, we really allow you to track against three separate budgets. What was the original budget that we put together? What was the approved budget or the current budget that we’re actually working against? Then, any sort of pending changes. If I want to go off to the side and model, “Hey, what if we moved,”, and this happens all the time on turnaround projects, “What if we brought some crews over from these overperforming activities and put them on some of these underperforming activities?” Let’s model that in real time and see where we might end up. It might be worth… In that case, the juice might be worth the squeeze to make those chambers because the tool’s going to tell us where our forecasts are going to be as a result of doing that stuff.

Lance Stephenson:

Well, let’s go back to your optimism bias. I just did an assessment on a project where they had $200 million in contingency and when I just did a synopsis of it and worked through the numbers, I predicted that they need 500 million for contingency. They were really off the mark in regards to that, just based on peer projections of math.

Rick Deans:

Mm-hmm (affirmative).

Lance Stephenson:

Now, you’ve got that optimist who probably says, “No, we won’t spend that much,” but how do you guys deal with that in regards to the volatility of change in relation back to your contingency?

Rick Deans:

Yeah, and you know, at the end of the day, the human being is going to have to make some decisions, right?

Lance Stephenson:

Yeah.

Rick Deans:

The tools are there to help us guide, help us make some decisions, but we need some human beings with some, to your point, Lance, got some experience that says, “This is underrepresented here.” Nate, do you have any [crosstalk 00:48:59]-

Nate St. John:

Yeah [crosstalk 00:49:00] well, I think one of the things that we’ve purpose built in our risk processes and directly in the tool to combat the optimism bias or the cognitive bias is we’ve introduced this capability called markup. What that means is that I can have… Let’s just use the schedule as an example.q I have a schedule structure and it’s set for review. I want to go invite experts across the organization, across the globe, anywhere, it’s completely remote, to come in and provide feedback on whether or not they think those durations are achievable.

Nate St. John:

The easiest example is I have a 10-day duration. I can go in and invite everyone on this call and they can mark up and provide their input on their own copy of it. We pull it into this consensus pool. We’re no longer go to individuals and saying, “Give me your min, most likely max.” We’re saying, “Give one number,” and it can be a range. I think it’s 25% aggressive. Oh, I think it’s 16.5 days aggressive. That consensus pool, the natural varying of opinions are then used to model the triangular distribution.

Nate St. John:

The three-point estimate is a direct correlation from the aggregation of varying opinions. Right there, you’ve suppressed the impact of cognitive bias just be flipping around the philosophy and the way we capture data. That’s one thing I wanted to point out. It’s a very powerful tool and probably a separate discussion in the future sometime.

Lance Stephenson:

Okay. Just to let you know, some people are struggling in hearing me. I know you guys hear me well. Some people are not hearing me, so if I engage with you guys, can you please repeat kind of what I’m saying? That way everybody else can [inaudible 00:50:51]-

Rick Deans:

10-4.

Lance Stephenson:

… sorry about that to the audience.

Rick Deans:

Lance was just telling us that some people are having trouble hearing him and that as we engage, we should repeat his questions that are coming up. Nate mentioned this earlier, but InEight does provide an integrated project controls platform. Some of the benefits of that is that we do feel we provide best-in-class functionality, not as a point solution, but all the way through a project’s life cycle from pre-construction estimating to post-execution, commissioning and turnover.

Rick Deans:

It is a cloud-based solution, so it does allow for that collaboration that Nate was mentioning. We don’t have to be physically connected to the same server. We don’t have to be in the same office. You know, we have real-world examples of people collaborating on our platform all around the world. We’ve been around for a while. As I’ve mentioned earlier, I’ve been working with some of these tools for over 20 years and we feel that we’ve got a proven track record in the market.

Rick Deans:

Well, it looks like we might have a few minutes left for some additional questions, if there are any. We want to, again, thank everybody for taking time out of your busy days to join us. We know that especially today with many people working remotely and some folks going back into the office, it is tough to still for a full hour and absorb some of this information, but we certainly wanted to give you a big thank you for taking time out of your day to join us today.

Lance Stephenson:

Okay. Yeah. No, that’s great. We do have a couple of questions. We’ve answered some of them. I’m just going to go through here and… One person asked a feedback loop. A lot of people that are dealing from the engineering perspective don’t get the contractor’s information that allows them to get the historical data in order to assess and validate construction estimates or the future [inaudible 00:52:57] cost. How do you guys manage that?

Rick Deans:

The question was, you know, feedback loop. Sometimes the data doesn’t come back into the office in a way that we might expect it. Maybe we’re an owner. We’re not getting that information back from contractor. Maybe we’re the contractor and we’re just not getting the good data back from either our subs or our own people in the field. How does the InEight set of tools manage that? Well, again, by having an integrated platform, we have a concept of doing daily field collection or weekly field collection, data collection from the field where that stuff, once it’s gathered at the source of work being progressed, it’s distributed across the various modules of our platform so in individual people don’t have to continue to handle the same piece of data over and over.

Rick Deans:

I always think back about early in my career, I was working with a customer in California and they were contractor and they did work all over their county, all over their region. At the end of the day, the folks would actually come back into the office. The foreman would come back into the office and they would stack paper time sheets on a payroll clerk’s desk, and so that information was collected at the field. It was manually driven in a truck back to the office. It was stacked on someone’s desk and that person at the end of the day manually entered it into a payroll system, manually took that data and entered it into a work tracking system. I think because we have that integrated platform, Lance, we do eliminate a lot of that duplicate touching of the data.

Rick Deans:

Now, there is the argument about getting data at different levels. Maybe we were expecting to get data at a very detailed level and it came in at a summary level, and that’s just something that our customers at first iterate through and then they find out that right combination of, when is the juice worth the squeeze? Right? At what point do we run into the law of diminishing returns in terms of the level at which we’re collecting the data?

Lance Stephenson:

Yes. Thank you for that. That’s exactly it. You know, there’s a lot of things that affect what we do. I just read a paper on behavioral analysis and you talked about optimism bias and the funny thing is that they say that people who are optimistic have undeveloped frontal cortal, and so their executive decision-making is somewhat inept. You got to look at most organizations are hiring optimistic people, so you got to wonder what that statement is.

Lance Stephenson:

Just a tongue-in-cheek kind of funny situation you look at when it comes to optimism bias. I think with that said, there are no other questions, really. There is maybe a couple more. Let me just see here. I just want to be able to see if… People are talking about… You know, they’re single questions I think you can answer offline. One person said, “It is not the likelihood of a risk event that matters. The likelihood of consequences of a risk recurring.” They’re referencing back to ISO 31-010. You have to model the joint likelihood in the event. Does your system do that in regards to this?

Nate St. John:

I’m having a hard time with you, Lance. I’m sorry. Rick, did you catch any of that?

Rick Deans:

I heard parts of it.

Nate St. John:

I know it had something to do with risk.

Rick Deans:

I heard parts of it, but I wouldn’t be able to repeat it.

Nate St. John:

Okay. Maybe I can go find it in the chat or something.

Lance Stephenson:

Yeah. Maybe look through the chat.

Nate St. John:

Yeah. I’m not seeing what one it could be, Lance. Does anyone else have another thing? I can keep on trying to search for it, but the latest-

Rick Deans:

We’re happy to [crosstalk 00:57:05] offline [crosstalk 00:57:06]. We’ve we’ve got our contact information up on screen, so if folks wanted to reach out to us independently offline, we’re happy to answer those as they come up.

Lance Stephenson:

Okay. Can you guys hear me now?

Rick Deans:

About the same/

Lance Stephenson:

About the same? Why don’t I let you guys close off, then, and I’ll just thank everybody, and then I’ll let you guys have the final words? That way I’m not breaking up.

Rick Deans:

Very good. Well, it’s been our pleasure to address this audience again. We thank everybody for taking time out of your busy days to join us. Nate, do you have any last-minute thoughts for our audience today?

Nate St. John:

No, I appreciate it. You know, there were some comments about maybe showing more real data. There’s a couple comments about, “Hey, this is standard, Monte Carlo, maybe a little more depth.” I think just to reemphasize the point here, the demonstration of sharing data across the platform is probably the most impactful thing here and unifying disparate systems or methodologies under under one umbrella. From my point of view and where I sit, Rick’s estimate can do a lot of a very robust education, if you will, in terms of what it provides users and contingency.

Nate St. John:

What I add is, is the additional insight of a quantitative analysis and capture of feedback from risk into the system. Appreciate all comments and, and hopefully that our messages got across clear. That’s all I [crosstalk 00:58:57] got Rick. It was a pleasure, Lance. Thanks to AACE and, again, our contact info’s up on the screen, so if there’s something we didn’t get to and people are eager, invite them to contact us and we’ll answer the bell.

Lance Stephenson:

Perfect. All right. Thanks, everyone.

Rick Deans:

Thank you.

Nate St. John:

Thanks.