Overcoming the “If It Ain't Broke, Don't Fix It” Mentality in Plant Operations

SHALE ROBISON: I was going to give you a quiz, a geography quiz. Mike and I are from Idaho. Not Iowa. And not Ohio.

MICHAEL JOHNSON: And, no, we're not potato farmers.

SHALE ROBISON: So I was going to let you vote for what state you believe Idaho is. The north borders with Canada. And we're next to Washington, Oregon. And we are famous for potatoes. And that will come up later on in the presentation.

Go ahead. Oh, I can do the pointer now. Oh, you got it.

All the words by my name basically mean I am old. I finally look the age I've looked for 25 years. Thank you. If I had gifts, I'd throw one to you. All right, go ahead.

I've been doing this a long time. In Idaho, in kind of the Northwest, there's not really enough industry to focus on one thing. So my background is multidisciplinary projects. Mike is our rock star in more ways than one. And that's literal, too. Anything you want to add?

MICHAEL JOHNSON: I'm not old.

SHALE ROBISON: Touche.

MICHAEL JOHNSON: So I don't have as much experience as Shale in the market, but I am a little bit younger.

SHALE ROBISON: We all have our cross to bear.

MICHAEL JOHNSON: Still have a dozen years of experience in the market. That doesn't include experience that I had growing up. I grew up in a household of drafter and engineers and people in construction. So I've been around it my whole life.

SHALE ROBISON: OK, so to begin with our topic is resistance to change or overcoming the attitude. I actually was at an Autodesk University. And one of the keynote speakers, a former CEO of Autodesk told a story about five rhesus monkeys. Has anyone heard this story just by that description? OK. Turns out it's a little bit of embellishment, but it works for our-- what I was told works what we're doing.

So the idea here is you see-- that's not a rhesus monkeys, but it's a wet monkey. And that's important. Five monkeys were placed in a room. Bananas were hung from the ceiling and a ladder to the bananas. One adventurous monkey climbed the ladder to go get the bananas. That's what monkeys do.

And the researchers praised monkey with cold water and sprays the other four monkeys with cold water. They all scatter and shiver. And another monkey gets curious and climbs the ladder, starts to climb the ladder, gets sprayed the same way. Shivering cold monkeys.

I don't know if-- this was done in '67. I don't think we could get away with that today.

But the third monkey goes to climb the ladder, a different monkey. And the other four monkeys attack that monkey to keep him from climbing the ladder, because they don't want to get sprayed.

They take out one of the monkeys and introduce a new monkey. The first thing he wants to do is get a banana. He goes to go get a banana. Four monkeys attack him and pull him off the ladder.

Pretty soon they replace all the monkeys that were in that original group that got wet. So no monkeys have ever been sprayed. If they introduce a new monkey in, they all attack the monkey and pull him down.

That story really informed my life and kind of made me think about why I do the things I do in my job. And so that was probably about 10 years ago when I heard that. And it's culminated in this presentation.

For full disclosure, the monkeys don't actually pull each other off the ladders apparently. I did some research. They just growled menacingly at the other monkeys. They didn't really overtly act like that.

MICHAEL JOHNSON: So when we decided to put this course together, I'm sure everybody here is familiar with the saying, if it ain't broke, don't fix it. So I got curious and decided to look up where the phrase came from, who coined it first. And it was actually a lot sooner time span than I was actually guessing that it would be. I figured to be somewhere in the '30s, maybe even in the '40s. But in 1977, the office budget advisor in Jimmy Carter's office, his cabinet, was actually the first person to a phrase through the media, mass media anyways. That was the first time that everybody in the United States became familiar with it.

I do have a question for everybody in the audience. Can anybody give me a guess as to who first wrote it or when actually they first wrote it? Just give me a year. Anybody have any guesses?

AUDIENCE: '78.

MICHAEL JOHNSON: '76. That was the first time it was written. I guess you can guess where it was written from. It does sound like a Southern slang.

AUDIENCE: [INAUDIBLE]

MICHAEL JOHNSON: Texas. The ironic thing is is the person who appended to or coined it into mass media, Jimmy Carter's, one of his advisors, he ended up getting caught in a banking scandal. So I guess, if it ain't broke, don't fix it.

SHALE ROBISON: It's a lesson for everyone here. It's broke, so you got to fix it. So I attribute this quote to a very well known person now. We are from Idaho. There's a lot of trees and lumber industry, forests. And I've never seen anybody chop a tree down with a stone ax, but maybe-- you developed the analogy. And why not just use a chainsaw? And if that's the case, split it to, split the word and use technology. So famous Mike Johnson.

MICHAEL JOHNSON: Too often we get stuck in the routine of using the same tools when better tools are available to us on a regular basis.

SHALE ROBISON: All right, last kind of jokey thing we're going to do is this. So Mike's been doing production and using software his whole career and understands-- you'll find out some of the things we've done with the Autodesk Plant software in particular to help minimize clicks and picks, so we can minimize errors and duplication and become more efficient.

MICHAEL JOHNSON: So by show of hands, how many people here today are familiar with what P&IDs are? Lots of people. How many of you have used Autodesk P&ID? How many of you still use Vanilla CAD for P&IDs? A few of you. So few of the things--

SHALE ROBISON: No judgment.

MICHAEL JOHNSON: Yeah, no judgment.

SHALE ROBISON: Know that's mandated.

MICHAEL JOHNSON: A few of the things that I have noticed over the time that I've used P&ID anyways is the amount of time spent creating P&IDs, whether that's drafting up a new P&ID or making changes to existing P&IDs.

And the more that you use P&IDs and everything, the more comfortable that you get, the better you're going to get at it and everything. But as far as using Vanilla AutoCAD, the benefit to using the software that's in hand, the better software that's in hand, is minimizing the one thing as much as we can that affects the bottom line the most probably in market, which is human error.

You know, if it takes you 12 clicks to do something that I can do in 1 in another piece of software, you're 12 times more likely to have an error than I am. And you multiply that over thousands of P&IDs over years of time and it becomes exponential. So that's where I came up with the equation, the less steps taken, again fewer clicks that you make building SOPs so that everybody's on the same page, that we're all making the same maneuvers inside the software, which equals less human error, because it's less time spent in the software, which ends up saving time, or giving ROI, return on investment.

SHALE ROBISON: OK, so now just to talk about our history, background, industry we work in, the company is called Environmental Intellect. That is a lot of syllables. And we just shorten it to Ei. It started out in 2008 doing environmental air quality compliance at oil and gas and chemical facilities. And for those familiar with that industry you know how concentrated all the assets and people and equipment are and how highly regulated the industry is.

We focus completely on oil and gas chemical now. When we submitted for this class, we were doing some other things kind of in the AC world. And the potato thing comes back again, full circle.

From Idaho, we do a lot of water and wastewater treatment, food and beverage, French fries, milk, cheese, fertilizer and pesticides, a lot of ag. And so a couple of the case studies we're going to talk about come from that world. Very similar to an oil and gas installation, but not nearly as dense of assets.

OK, so my experience is I have owned an Autodesk reseller and a CAD consulting firm. Our team was on the beta testing team for the P&ID product back in 2009. And then the next year Plant 3D came out. I was really excited. I think I was nine back then. I have established that I am old.

I was excited in my career to have an intelligent, data-driven P&ID solution. It didn't really change a lot of things that much. But I started to evangelize the product, because it really kind of opened my eyes to the possibilities. And it just kind of invigorated me.

And we used to sell our services or sell the software. And we would lead with one or the other. And then follow up. Usually, if we sold the software, the client needed someone to implement it and perform the work. So then we had a team that would do that. Or if we led with our services, then we would use the software that was more efficient and effective. And then we would come and be able to sell it.

But what I did realize, in fact, not to belabor this too much more, Ei was actually a client of my previous company. And we would convert P&IDs from PDFs, from TIFF files, from mega station files, from CADworx files from chicken scratches to Intel's P&IDs.

But when we got into the Plant environment, we were trying to sell this better tool. And I don't know if I've heard it before, but I just struggled against this negative momentum, like resistance to change, that even though it was better and most of times the client realized it was better, they just do not want to change the way they're doing it. It's the way they've always done it. The monkeys have always not eaten the bananas because they were conditioned not to. And so that's, yeah, basically that.

MICHAEL JOHNSON: So how many people in here are familiar with the phrase PSM? It's Process Safety Management. Basically, just monitoring of activities by OSHA and EPA.

So when I was a teenager, I worked as a welder and an iron worker. How many people in here have worked in an industrial plant anywhere that's got OSHA standards that you have to adhere to on a regular basis? And regardless of the fact that OSHA does have standards and you are supposed to follow them in industry and in the plant, in the field, as you can see from the picture, that doesn't always happen. I have many stories of me almost getting pulled off of sky rises and everything as an iron worker because I was told by my manager not to tie off because it took too much time.

So what's the forest evolution of PSM? Well, obviously we want to decrease the amount of time you spend in the field. And in order to do that, we have to evolve the way that we use software. The safest job you can probably have is sitting behind Autodesk, unless you've got carpal tunnel.

SHALE ROBISON: Let's jump into some case studies that really represent a big change that a company had to make. OK, so the first one is a French fry plant. We all love them. But no one wants to admit it.

OK, so in this case study, kind of to set the context, this was a large French fry manufacturer. You can choose between five main manufacturers. They had just completed a sweet potato fried plant in Louisiana. And they had done it using SolidWorks. So they modeled everything in the Plant in SolidWorks.

Who here thinks that's a bad idea? And why? Why is that a bad idea?

AUDIENCE: [INAUDIBLE]

SHALE ROBISON: It was, to use a term my kids use, ginormous. Yeah. They had to spec out the highest performing computers you can imagine just to load parts of the model. So it defeats the whole purpose of having an integrated collaboration model, because you have to turn half the things off just to get it to load.

That's prior to this change. They were looking at ways to change and wanted to model. The one positive thing that came out of this was they saw enough benefits of simply just having a model of their entire facility, that they wanted a better way.

So at the time, I was selling software and services. I actually didn't sell any software to this client. They be bought from a corporate entity in the Midwest and through a different reseller. But what I did do is I went in and kind of showed him some things about how the Autodesk products worked together.

And I remember talking to the project manager. And he just straight out asked me, is this going to work, because the SolidWorks guy told me it would work? And finally after they've exhausted all the excuses of it's the hardware, and then the owners, operators, bought the best hardware you could buy, they just said, OK, I know what it is. And just walked away.

And so he asked me straight up, is this going to work? My answer to him was I don't know. Actually, it I wasn't I don't know. It was I've never worked on a project of this size with all of these components. So there was Revit, Navisworks, Plant 3D, P&ID-- at the time, they were separate products.

What else was in that? Oh, there was some SolidWorks models. Inventor, so some of the manufacturer's equipment was given to us as models. But I said, I've worked on different projects with most of these packages working together. So I am very confident that it will work. And apparently I have an honest demeanor. So they went ahead and did the project.

And the only benefit out of the first project was it had some really bad workstations. They probably still use them today as far as I know.

A couple of things about this is the general contractor-- it was a new product line. So $200 million project, 193,000 square feet. General contractor was using Revit. So we had to coordinate a Revit model. The biggest Revit model I've seen to date, it was a gigabyte. We had to make the plant 3D models between 10 and 20 megabytes in size and then used Navisworks to coordinate everything.

And it worked out really well. Everything was coordinated. There wasn't a change order in the project, except for a conduit.

MICHAEL JOHNSON: So, yeah, there were zero RFIs. They saved over $0.5 million in RFIs alone.

SHALE ROBISON: Yeah, I said change orders. RFIs. There were some change orders.

MICHAEL JOHNSON: Something to take away from this project as well was we kind of did discover things and learn things as we were going through this project, because, you know, it goes in there with you saying you don't know if it's going to work or not. We had a couple of people, myself included, on-site working with this company on a regular basis.

And I think that's part of change that's necessary was, well, you can get new software. You can get the new tool. You can upgrade to the new tool. It doesn't mean you're going to use it or use it right.

They were doing something as simple as not understanding that you can hit a Refresh button instead of Navisworks. They were putting the whole model together inside of Plant 3D and then pulling that into Navisworks. And then just re-updating it by saving and then copying and pasting it back into Navisworks every single time.

So just little things like that you recognize. You can always get online. Autodesk, obviously, they've got forums out there where people are constantly communicating back and forth. Even if I have a question, I utilize Autodesk forums on a regular basis or Google on a regular basis even. If there's something that you don't know, odds are the answer is out there. But you just have to dig deep and search for it.

SHALE ROBISON: Yeah, that reminds me, this particular project had 17 different subs across seven states. And it was really beautiful to watch it happen. And it was pretty stressful as you would think. But not as stressful as it could have been, because the things worked together the way they were meant to.

And even the lessons learned-- what they got out of this, so the project manager told me it was the best project they ever had done to that point. So essentially in one cycle they changed the way they did everything. Actually, they did it in two cycles. They did a SolidWorks try. And then they did the kind of coordinated Autodesk solution.

And then the other thing they got out of that-- because the SolidWorks didn't provide them ISOs or P&IDs. They could extract some Orthos, but they weren't connected to the model. The bill of materials and their equipment list, they used the heck out of the equipment lists. Engineers were putting in all the data. And then we were sucking it back into P&ID. And there was one other thing I was going to say about that--

MICHAEL JOHNSON: How many people here have been inside a French fry plant?

AUDIENCE: I have.

MICHAEL JOHNSON: Just food processing facility at all? How messy where they?

AUDIENCE: They were actually clean.

MICHAEL JOHNSON: They were clean? So I've been in several over the years and everything. This particular facility is probably one of the cleanest facilities I've ever been in. And, you know, even going into some of the other clean facilities, any of that's on the floor, fries do get on the floor, oil gets on the floor, and everything becomes slick, which becomes a slip hazard, which again, safety is a huge concern.

So adapting to the software and everything, the way that they built this facility, like I said, one of the cleanest facilities that I ever walked through. There was no slipping issues. I didn't see any safety hazards at all basically going through that plant. And they do have two plants right next to each other. So going back and forth between one and the other is night and day.

SHALE ROBISON: Yeah, a couple other interesting things about it. The openings between the floors, there was no issues. Because it was modeled, everything fit perfectly-- floor drains, openings between pipe stuff.

I did remember what I was thinking of. So even anyone that hasn't been in kind of a French fry facility, they actually just, FYI, they actually run the potatoes through pipes. So pumps, exchangers, boilers, I mean it's just like most other facilities you're thinking of. The impellers on the pumps are just flexible. So the potatoes go right through them.

And one way that they create fries is they shoot it out-- it's called a Lamb gun. It's like a cannon. And they have knives, a set of knives that are kind of at the end of the pipe. And they just shoot the potato through it. And it's French fries. I love telling that story, because I was amazed before I went into one.

Are there any questions about that particular instance? OK. I'm taking that as a compliment that we've explained it so well.

All right, so now Pharmer Engineering. OK, so the context for this change is they're a consulting engineering firm, primarily doing water and wastewater treatment in food and beverage. So all kinds, I mean, Frito Lay, Kraft, Lighthouse Foods, French fries.

And the CAD manager was actually a partner in the company. And he left. And I had known him-- I actually tried to sell him and try and get services from him and he never would budge. He never bought from me. But he did recommend me to his partner when he left.

So I go brand new into this company. They're using Trimble Pipe Designer 3D. Let me pause on that too. Just so you know, my attitude about our job as consultants is that any CAD package is a tool for us. So we would use whatever the client was using. So we've worked in MegaStation, CADworx, Pipe Designer 3D, PDMS, OpenPlant, that sort of thing.

But in this case, it was kind of an exotic solution. And so QuickPen, if any piping designers in here that use QuickPen, Trimble bought it, made it Pipe Designer 3D. So we had to come up to speed pretty quick on that.

They did-- to their credit, they modeled everything. All the plants that they did were modeled in 3D. They were what I call dumb models. I mean, they were kind of AutoCAD objects. There was some intelligence in PD 3D. They did extract Orthos, but they had to re-extract them and kind of recreate them. They had kind of cool sectioning tool. And then the P&IDs that they did were just manual iP&IDs was intelligent P&IDs.

So that's what we walk into. There's a vacuum. Their brain trust just walked out. He had really had documented everything really well about how they did things.

But given the opportunity, I sat down with the remaining owner and said, OK, are ISO metrics important to you? And he's like, I'd love to have ISO metrics. He didn't have the capability, so we never consequently produced them. And I said, P&IDs will talk to your models and inform your ISO metrics. And so he was totally on board with it. So we had to implement this in their work practice.

And again, as far as giving some words of advice for doing this, we again had to embed people there. You can't just come into someone's business, even if they do the same thing you do and hit the ground running. There's always enough nuance to how they manage files, file management, CAD standards. So we embedded people there. And that was kind of a win-win scenario.

So with that, like by show of hands, how many people here work for an owner-operator? OK, then what about like an EPC or a consulting business? OK, so in those scenarios, it would be my recommendation to bring somebody in-house. I mean, there's a lot of firewall things about plugging your computer into their computer. But bring them in-house and do it for a long enough time that they understand your business. And then they become an asset for you.

So even to this day, we still get calls saying, hey, can you come help us on a project? And we like to have a lot of clients, so we're not dependent on one client for all of our business. But they'll call us when they need us. And they'd like that flexibility too. So from a consultant or EPC standpoint, that's a different approach I think than most. You want to get that contract and be long term and go forever. And sometimes owners, sometimes they appreciate it, but sometimes they like the flexibility.

So the other thing was using PD3D, you had to have a plug-in for that. You had to be on the right version of the plug-in if you were sharing files with other consultants. So the predecessor there, he didn't care. He's like that's their problem. We create this model. And we use this program. And if they need to access it, they got to figure it out.

But in talking with the owner, I said, wouldn't it be easier if you could work with all your subs and communicate and collaborate and coordinate? Which has always been a big tenet of what I have strived to do over my career. So using Autodesk Plant products, we were able to speak the same language as all the other subs.

And then as the result for that, their P&IDs were standardized. When they were using just Vanilla AutoCAD to generate P&IDs they weren't even using the PD3D P&ID creator. Lots of different sizes, different layers, and all those things that AutoCAD P&ID Plant 3D can standardize.

So we had to standardize P&IDs. They had Orthos and ISOs that were connected to the models. And the modeling actually was much faster. And so just this year, they stopped their subscription to PD3D. I actually suggested to them-- they reuse a lot of the models from that program. I suggested to them to keep that program for a year to see if they had-- they would always go borrow from previous projects. So keep that in case you need to bring a model forward from a previous project. And so in the last year, they haven't had to do that.

When there was downtime, we had their guy, their CAD lead, convert all these models into the Plant 3D format, where they would take the blocks and just assign connection points to some of the pumps or whatever they had already developed. So it wasn't a total loss on their previous development of content.

And then I like to say better, faster, stronger. OK.

AUDIENCE: I have a question.

SHALE ROBISON: Yes, go ahead.

AUDIENCE: From the previous case study, so you said [INAUDIBLE] AutoCAD or is that PD3D model--

SHALE ROBISON: PD3D.

AUDIENCE: OK. When you convert it into Plant 3D, did you actually tie the P&IDs to the Plant 3D?

SHALE ROBISON: Yes.

AUDIENCE: Was it a one-way link or was there a two-way link? [INAUDIBLE]

SHALE ROBISON: So it's not a two-way link, but you run a verification. And it says, but on this line, there are three valves, two gate valves, and a check valve or something in the model. But there's only one gate valve and one check valve on the P&ID. So it identifies through the check the validation. It identifies those things. Then you update the P&ID in that case and then rerun the validation, and that will go away.

So you use the P&IDs line list to draw in Plant 3D. You take the line list and you just on that line will have all the components on the line. You just pick it from the line list and place it-- the spec and everything comes from the P&ID. And you place it in your model.

AUDIENCE: They issue that I had with that or what [INAUDIBLE] it allowed me to certain multiple of the same exact object. It actually allowed me to do that. It gave me a warning if I [INAUDIBLE] verification. But it actually allowed me [INAUDIBLE]

SHALE ROBISON: With the same tag?

AUDIENCE: Yes.

MICHAEL JOHNSON: Did it out of question mark to the end of the tag at least?

AUDIENCE: Yes.

SHALE ROBISON: OK. So that's another way you can know it's not right. It's not exactly the same, because it won't put the same tag number. It shouldn't anyway without knowing seeing the setup. But, yeah, I can see that as a-- you would find it in a validation.

It's not that good-- the very first thing, one Plant 3D came out, we were actually told as one of the beta testers and reseller partner that there was a two-way thing and I've been waiting for it ever since. It wasn't there and still isn't.

AUDIENCE: Is it coming in the future?

SHALE ROBISON: I have no idea. Yeah, I've dealt with that long enough. I've stopped thinking about it, because I don't want to be disappointed. Yeah, so I don't know. I don't know if it's in development or not. Yeah, I've just worked around it.

And really, that's something I've learned in my career is you just figure out a workaround, because it was a sad day in my career, because I was really a big Autodesk and you know CAD technology proponent. And I started to tell clients, yeah, it's just software. I mean, it doesn't do everything.

But for what it does, I know if I have to have workarounds, I can make those workarounds. I'm not a programmer, so I can't reprogram it. So I just have resigned myself to the fact that there's some things I'm going to have to use workarounds on.

Or like Mike said, we're really big-- this is another point that I would make. Create standard operating procedures for everything. Anything you do more than once, create an SOP. So, A, you document that and standardize that, because if it's not part of the program, you have to create an SOP to get everybody to do it the same.

That's what's one thing a lot of clients have said is I can tell who did that P&ID and I can tell who did that one. And they were different because they look different. And so by standardizing on SOPs, then we kind of remove that. But that--

AUDIENCE: So where [INAUDIBLE] require also create the line number [INAUDIBLE] in 3D models so as to match them?

SHALE ROBISON: You're asking if you're required to or--

AUDIENCE: Yeah, were you required to create a line by custom line number to match?

SHALE ROBISON: You got to make the tags match. The tag formats have to match between the two.

MICHAEL JOHNSON: Just add to that, Shale, so the way that I've done that in the past and everything, the way that I would recommend that that's done to any client is that you finish your P&IDs first. While that does seem a little steep and everything, it's hard to convince people that sometimes you have to make it very apparent to them that if you don't do it that way you're opening yourself up for issues down the road. I mean, you're literally kicking the can down the road. That's all you're doing.

So while it may seem like it's slower and everything, if it's slower and you do it once, it's still 10 times faster than doing it, making a bunch mistakes down the road. So I would recommend completing your P&IDs first. That way your line numbers, valve numbers, anything like that, it's not changing once you drag and drop that into the Plant 3D model.

SHALE ROBISON: Yeah, the best way to describe this is I realize that that's not always possible to complete your P&IDs before you do your models. So I would at least say get to an 85% level on your P&IDs in the P&ID aspect of the project before you start modeling, because you're just going to end up redoing things and errors will happen. But, yeah, I mean I don't know if that answers your question.

AUDIENCE: It does.

SHALE ROBISON: Yeah.

AUDIENCE: And if I may add, sometimes, once a week we got to the point where we found an error in our P&IDs and we couldn't fix it. So the solution that we do it is to detach the drawings from the project and then copy them to the project. We thought it was an easy fix. But that lost all the line numbers. And that basically meant we had to reassemble these line numbers to be matched in 3D. And that was just devastating.

MICHAEL JOHNSON: That would be difficult.

SHALE ROBISON: What industry do you work in? What is the--

AUDIENCE: Just assembly-- the [? drinking ?] plant [INAUDIBLE]

SHALE ROBISON: OK, so consulting or for the owner-operator--

AUDIENCE: Actually, I work [INAUDIBLE] and I'm modeling it as it. So I'm making a 3D [INAUDIBLE] of it.

SHALE ROBISON: OK. Yeah. Yeah, a lot of my experience was in water, wastewater. And when Plant 3D first came out, nobody within Autodesk-- our territory manager wasn't talking about it. And I said I know this will work for water and wastewater. So I'm going to keep pushing it till it's used. And it's got quite a bit of adoption at this point.

But it just sounds like there's maybe some consulting opportunity to help tighten that up there. I don't think you should be running into that problem as much is as you described. You should be able to bring it back in, and it will keep the line number is the first thing that I think of.

So I think we hit that one. OK, so now the last one is in an integrated refinery and chemical facility. The context for this is so we have field teams that are actually going around and tagging components for FUGEM emissions for the LDAR group within the refinery. OK, pop quiz, who knows what LDAR is? Someone start saying it? Oh, somebody was having a dream. Yeah, good ahead.

AUDIENCE: Leak detection remediation.

SHALE ROBISON: Yeah, leak detection remediation and leak detection repair. And my business partner cringes every time I bring that up in class.

So we are on-site there. And we have field guys that are hanging a tag. It's different than the asset equipment tag. It's for the FUGEM project for the LDAR department.

We use P&IDs for this. Everybody in the industry to do this particular job takes an 11x17 paper copy of a P&ID into the field and highlights as they walk down the plant. So I've heard people kind of compare things. Who remembers the time before smartphones or before the iPhone? This is way before the iPhone came out. And the industry is still using the same practices.

So we knew there was a better way using technology. So we're there, and we actually convert the P&IDs. And in this case, we did 2,500 P&IDs. We converted them. We convert them for about $100 a P&ID from AutoCAD P&IDs intelligent P&IDs. So they use AutoCAD. They just don't use the intelligence.

And what we do is we pre-highlight them based on data. And data is really the way to do this. Because if you have to worry about layers, manipulating layers and layer managers and layer states, it just becomes really difficult to manage and really kind of untenable.

And so some of the things we had to monitor are difficult to monitor. DTM is what they call it. And they're usually high. So we have to build a scaffold to access them. Or they're under a platform. And so I've got to build a scaffold for that too. And those take a majority of the time to monitor a minority of the valves or the instruments that we need to monitor.

So we introduced laser scanning as a way to operate more safely within our job. It was hard to tell the client that we needed to laser scan in the facility. And you have to get a low energy permit or sometimes hot work permit, depending on the facility, to do that. There's a lot of things kind of resisting using this.

But actually because of the safety incident, we worked with the client to help them understand why we needed a laser scan, so we wouldn't have to climb up and get to those things. We could just shoot it from the ground and tag it electronically. And so that's kind of the context of this.

It's basically a two-year long project. We're going to walk down and re-validate every unit on the site. And then earlier this year, the client hosted a safety symposium basically in one of the firehouses at the facility. And we were able to demonstrate and show some videos that we created showing the benefits of these.

And from that, we had operations guys coming in and seeing this turnaround planning. The safety guys were already aware of it because of the safety incident. And so we started to give them better ways of communicating with 3D digital data. We created little videos walking through a plant.

So if you have somebody coming in to replace a piece of equipment, if they're a contractor, they may not be familiar with your site. So the way they plan their work packages now is they take a bunch of pictures. And they markup drawings and P&IDs. And they have very tabular data saying step 1, step 2, step 3. And so that's not a very good way of communicating. And so-- yeah.

AUDIENCE: Did they recognize before you guys got engaged that something was broken, or did you have a challenge from in convincing them that it was broken?

SHALE ROBISON: Thanks. That's a good question, because they didn't know anything was broken. That's just the way they had always done it. And that was the best way they knew how to do it with the tools that they had. And the tools were a room like this size, big whiteboards with magnetic pins on them. They would just stick on drawings and take pictures and have meetings. And take everybody out in the field and prepare work package.

I guess we start to work through this with one of the turnaround planners. But when we were able to show to all these people-- there was probably 10,000 people that worked between employees and contractors at that site. And so we had a few thousand people walk by. And they saw some videos that we had made.

Mike and I did a presentation yesterday about the way we did the laser scanning. And now I regret I didn't include any of those videos, because I thought there'd be people from that presentation here. So I didn't want to be redundant. But maybe at the end, a little bit, we can show a couple of videos.

But the videos, just to your point, the videos made their eyes open up. And they said, if we can use these videos to communicate to everybody involved, it's a virtual experience for them. They don't have to be in an unsafe environment. It minimizes the amount of time they need to be in an unsafe environment. They're already familiar when they walk out there and see this. And so it really prompted them to change the way they're doing things.

And so we've been performing some pilot projects for different departments. So that's they other thing. You get departments talking. There are probably 100 databases of the same piece of equipment for different purposes within a refinery. The different silos or the different groups don't necessarily need to talk to each other, so they don't. They just do their jobs.

But if we can bring it together and make a unique address of each asset, then we can tie all the databases that relate to that asset together. And so that's what they're realizing. And so they actually converted one of their turnaround planners to a full-time position, dedicated position, of using the scan data that's coordinated with all the other databases in the plant to plan these things.

And some of the benefits that specifically happen, he used an example to me, he said, if I have a crane working above an area and we also have a work package down at grade, and we didn't catch that in the planning, the coordination didn't happen, now the guys at grade have to get out of the way for four to six hours, while the crane work happens. And had they known that ahead of time, they wouldn't be paying guys for four to six hours and not having the work done in have to fit that work in some other time slot during the turnaround.

And that's a really big company that has really good methods, that has actually pivoted pretty quickly to realize that their system, although it was the best they could do, it was still broke when they realized there was technology that could help them perform that same task. And just by comparison, planning some little things, like if you're just going to change out a valve, they're kind of these mini-- they're not even really turnarounds, they're just work packages, work letters-- it took four hours for a planner to plan a work letter.

They get PDFs of the P&IDs. They get the drawings, and they get the work package letter. And they assemble all that. And then they go out and write it up. And those would take four hours of time. And now, they're able to do 10 a day.

And they do it safer because they're in the plant less. And so now the challenge is to not go too fast and kind of pull the reins in, because they know what helps their job. In fact, I've said to a lot of them to start with, I said, we know how to create this content. We just need you to tell us what you need it to shown.

Are there any questions? I'm going to cue up a couple of the videos?

AUDIENCE: I have a question.

MICHAEL JOHNSON: Go ahead.

AUDIENCE: I guess it might be a question for you. I'm in oil and gas out in Montana. I work basically in this realm. What is the average time basically of a user changing from a software package to AutoCAD 3D?

MICHAEL JOHNSON: So like on a per user basis?

AUDIENCE: Learning curve?

MICHAEL JOHNSON: A lot of that really depends on the user. And a lot of that really depends on how the user is trained. So I used to work with Shale, by the way, when we were an Autodesk reseller. And I would travel and train people on a regular basis.

And I can tell you that while training is good in some light, when you travel and train, you're there for one day, maybe two, maybe three. They're only going to retain maybe 25% of what you say. That's not a lot of knowledge that you're actually passing on.

So the best way to really do that is instead of just training you on the program, I would train you on the project, meaning we get into the project. We get into the drawings. I show you how to start things. And then you just kind of figure it out as you're going through them more and more-- repetition basically.

On average, I would say it can take a user probably up to two months to get super familiar with the software, to an adequate level anyways. And that will help decrease how long you're spending on your drawings and everything. An example with that is a new user converting a P&ID, for instance. You know what that means, by the way, converting a P&ID? That means just taking basically a Vanilla CAD drawing. It's a P&ID and converting it to an iP&ID or smart P&ID.

Typical time for doing that for a beginner, it can go from like a day to day and a half. Any more though, we have SOPs and a QIQC process to keep that time extremely low. We try to get it within the one to three hour mark for converting a P&ID.

And then quality assurance, quality control, which is QIQC, we try to keep that to a minimum as well. I think we have a 3-step process that's pretty rigid to make sure that we don't miss anything. And that helps with the learning curve as well, because what you're seeing is you're seeing how to do it, and then you're seeing what you did that was wrong at the end. So in a sense fixing you own your own issues that you're finding.