Transcript: Collaborative Planning and Future of Robotics

Before I even really got into this whole lean thing, I was trying to get the computer to build a building and give it the inputs, and figure out the optimum solution, and bring that to folks and say, hey, you got a good schedule here, but if you do this, this, this, and this, you're going to have a better schedule. People are like, oh, interesting. Sometimes they did it, but most times they did a few tweaks, because they didn't come up with it. It was a lesson learned as much as a debate internal.

We had experienced the other side. We kind of come full circle and recognize the real, the value that people bring to the process is instrumental. Hello and welcome to the Constructor podcast, the future of the built environment. I'm your host, Brittany Campbell-Turner, and this podcast is dedicated to helping property owners have certainty in their decisions about their construction projects. We talk about fostering trusting relationships, help you to understand how to lower risk, be under budget and on schedule, and most importantly, exceed your end user's desires.

So I've been letting you guys know for the past couple of weeks, I'm looking to add to the Constructor team, looking to have some support in the way of graphic design and report generation. So if you'd like to partner and create reports with us, reach out to me at Brittany at Constructor.com. That's B-R-I-T-T-A-N-I-E at ConstructRR.com. Last week, we kicked off our recap of the 20th annual Lean Construction Institute Congress, and we started it off with Bruce Cousins, Managing Partner at Sward Integrated Building Solutions. We had a casual conversation about his focus on the business of design and lean in design.

This week, we will continue our recap, and today's episode is a fun one that talks through a few additions to TouchPlan, enabling the digitization of the last planner system, and the potential of robotics in construction. Not only that, but we talk about the necessity to keep humans at the center of design and the work. So today, we're talking with Michael Carr. He's the president of Mocha Software Products Division called TouchPlan. He has 20 years of construction management experience, and he currently leads the development of TouchPlan.

So I'm sitting here talking with Michael Carr from TouchPlan. Hey, Brittany. How are you doing today? Doing good. Yeah, these conferences are always a lot of fun.

It's funny, the first one I came to, you don't know anybody, right? And now, on the elevator, you bump into somebody, you know, I haven't seen you in like a year, about the last time we were all together. So you've been part of this community for a while now. Missed last year's conference, but I was here at the other ones, and we've been coming either full on, everybody sees us or kind of hidden. So this is probably the sixth one that we've had some kind of representation from our company here.

We weren't even TouchPlan back then. I think we were still, we didn't have a name yet. Really early days. Really early days, yeah. I was sharing with one person that I remember, we were all in a room, it was like the cocktail hour.

It was a room not much bigger than kind of this foyer here where we got these booths set up. And there were like four corners, obviously, and in each corner, there was a different vendor. It wasn't really necessarily a vendor. It was some kind of a sponsor. And they were just set up for that two hour period, and then they tore it down, and that was it.

Wow. So they've come a long way. Now we've got actually an exhibit hall, and we've got probably two dozen, maybe three dozen different. It's a real thing now. We had you on the podcast.

Yeah, it was a lot of fun. Not too long ago. Some of you may remember Mike's and my interview a couple of months back. If you haven't listened to it, or if you want to catch up on that, check out the episode at constructor.com slash 83. We talked about what the last planner system is, and I'm actually pulling in some of that interview for context so you understand what TouchPlan does.

It essentially enables you to do the last planner system electronically. The really cool thing about the last planner system is that it involves those people that are actually responsible for getting the job done. It brings them into the process. So they put together a plan. It's usually a better plan than what one person can do on their own.

The other thing that happens is, well, it's their plan. So there's some ownership, there's pride of ownership there, and that actually drives up everybody's accountability to one another. It's like, well, yeah, I said I could do it. I'm going to do it. When that happens, a really amazing thing happens.

Things start on time, because work that was scheduled to be done is done on time. So one of the big issues that, this is traditional execution, not last planner system, in construction, and I was talking to a mechanical trade contractor a few months ago, and they were saying, look, about 30% of the time I show up at the job site, and it's not ready for me. And he's the mechanical guy. He's putting in the duct work, which is one of the first things that goes on the floor.

But if 30% of the time he's showing up and he can't start because something's in his way, then think of the poor souls that are later in the process behind him. It's only going to get worse. So the main thing that happens is these trades are making commitments to each other, and the byproduct of that is they're feeling responsibility to one another. They're finishing their bits on time, which allows for the immediate start as planned. That just pulls that whole schedule in and keeps everybody kind of flowing and working together.

Another big thing about the last planner system is there's a big emphasis on looking ahead and kind of clearing out all the constraints or blockers. The term would be like make ready is something you might hear them say. The whole purpose there is to identify issues or the things or blockers out in the future that need to be taken care of now, like getting a submittal approved or a response to a question answered or getting a permit, material delivery scheduled, getting that thing dealt with now so that when the time comes to do the work, it's planned, we're ready to go.

Now that you're reminded of the last planner system and touch plan, I've asked Michael here at the LCI Congress to share some updates about the software. You know, obviously there's changes, there are improvements to the software itself. So I mean, maybe you can give some updates on that. And then I have some really interesting topics I'd like to dig into. Right now, it's kind of not well known because we're rolling out kind of slowly just to make sure everything's tied up nicely.

But we've got a version, we're calling it beta for release number 33. And it's basically a combination of a lot of effort over the last several months to streamline the behind the scenes in the app, improve performance. A lot of the stuff that's getting rolled out, it's getting rolled out a week from Saturday. That's the plan to the full production crew. So there's a handful of people in there playing with it right now, but it gets rolled out in a week from Saturday.

And a lot of it is people aren't going to be able to see a huge difference, but it's a performance thing. So it can handle much larger plans. There's no lag anymore. It's just a lot better. It's actually based off technology that we looked at five years ago and it just wasn't ready.

And so we're like, okay, it's ready now. Let's rewrite some of this stuff so we can take advantage of it. And so we have. But on top of that, we did have to add some new user features. And so there's some cool things in there like when you drag your tasks around, you can actually see them change their shape, change their size, depending on if they're crossing over a weekend or not.

And the current version, I think you just drag it, you drop it, and then you wait like it's not too long, maybe, you know, a split second, and then it changes its shape. So you don't but you don't get the real real-time feedback. So now you're going to start getting that. Another key thing we did, we recognize that there's this mode where we've got all these arrows that we call the dependencies mode. It was a mode and it was kind of hidden.

I mean, you had to push a button and pull it up and then you saw it and you're like, oh, okay. And you couldn't really interact with the plan while you're in this mode. So we've eliminated that. And now while you're in the plan, you can actually see the arrows and you can add arrows and you can remove arrows and you can move your tickets around while they're connected so you can see what happens to your arrows. And so bringing that out to increase the usage of that feature and that's actually our step one.

So that's all kind of getting released in a little over a week. And what the team's working on right now, a couple of really cool things. Today, when you want to see kind of labor totals for your plan, you got to print a report. I mean, that's the way we've got it set up. So you have to go print a report.

You can look at the, okay, I'm going to have 10 plumbers on Monday, 12 plumbers on Tuesday, it goes up to 30 plumbers on Wednesday. That ain't happening. I got to go back and fix my plan, right? But you get that from the report. Well, we're modifying it.

So you're going to be able to see that live real time right in the plan. So while you're planning and you're moving your stickies around, you're going to see those headcount change to reflect whatever your plan is. And so we're going to show it for you within your plan and also across your entire project. And you can look at it as the plumber or I can look at it for the whole project. What's the total headcount on the job today?

That's getting developed right now. There's some other things in terms of being able to organize your work and pick up a swim lane and move it. But when you move it, you're dragging all the stuff that's in it with it. I mean, that's stuff like that. Right now, swim lanes are kind of simple.

They're just drawings, if you will, on the grid. Then you can move them independent of the tickets. We're fixing that. So you can actually move the tickets with them at the same time and just make it more useful. Those are kind of the early things.

A little bit further out, some of the cool stuff we're looking to do is start capturing plan versus actual. So right now in our app today, we do capture plan versus actual end date of the task. But that's it. So when you promise you make a pin, you put a pin in your ticket, we'll capture that and say, OK, that's a promised date. And then when you actually finish, you put an actual date down and we can compare and say, well, that was later.

That was early. What have you. So the next version will have not just finished dates where we capture plan, but in actual, but also all the attributes. So what was the start date? Was the plan start versus the actual start?

What was the plan crew size versus the actual crew size? Was the plan duration versus the actual duration? So just have a little bit more there so that you can start capturing plan versus actual. And then the next step beyond that is augmenting what you can put on a ticket. So today there's a handful of things that we let you put on by this time next year.

You'll have custom or user defined fields. So if you want to capture material quantities, we get this request often from trade partners. Maybe it's a mechanical guy and he's like, I want to just write on the sticky, you know, 3000 pounds of ductwork because I know what it is. I've captured this on a spreadsheet off on the side. I'd rather just not do that.

I'd rather just put it in the app because my labor is already there. Put my materials there. And now when I move the sticky around or whatever, I'll have I'll have a total count of ductwork. They need on site that day or what have you. And I can do the math behind the scenes and figure out productivity.

And not only can I do that on this one job, but if I'm using touch plan on all my jobs and we got some trade partners that are doing that, even though the jobs themselves, the GC is not necessarily using the trade, might be using it just behind the scenes for all of his stuff. Then he's got this window on productivity across the organization. So user defined attributes, user defined fields will be will be here by this time next year. And the other thing that we're working on right right now is is kind of a dashboard.

So the reports are nice, but I want to see them live in real time. I want when you when you actually log into the app and you open up your project before you go in. We'd like you to see the latest statistics. What's the what was the PPC, you know, this week versus last week? What's the what's the current expected end date today versus what it was last week or last month?

What's your percent complete, you know, and then give you the ability to kind of interact with that so you don't have to generate reports. You can just play with the dashboard and do some drill down effects. I mean, it's business intelligence. This is the big thing, right? We're looking to incorporate that.

And so I've got some developers focus on that actually right now. What's the time on those dashboards? I think the first version Q1 of next year, but the idea is put something out there. Get feedback. People say hey, this looks great.

But can you add this or I don't really use that can you know, that's not useful, but this is and so that's the way we've always developed. What I tell the team is they put out the 80% solution. I say I say them if I wasn't doing what I'm doing, you'd be selling this to me because I was going down that path to be ultimately a user of touch plan. But since I'm not and we're doing this, we have a pretty good sense inside. Get the 80% solution.

Put it out to get the feedback and then iterate and improve on it. And that's that's been our model and it works great. We don't want to stop that until we're never just going to have the app and say oh, it's all done and we're done. There will always be developers in the back room. Will you do it with a lean perspective plan?

Do check act right? So you're doing it on all fronts. Eating our own dog food there. Exactly use that software language. I'm personally excited about the dashboards just from the owner side.

I think it's going to be super valuable to be able to respond to their requests about the progress on the projects particularly and what the forecasting is. I'm actually working on a project right now. Okay. Yes, this this would be very useful. Unfortunately, we're going to be using touch plan.

So I'll give you some feedback personally if I have any challenge. Awesome. Awesome. We talked a little bit yesterday about some of your passions and I wanted to ask you about this on record robotics. You have an interest in that and I wanted to ask you why and then like what's the hottest thing that you're interested in?

One of the nice things about being in the kind of the startup community is you're surrounded by other startups and investors and other startups and you kind of get a pulse of what's going on in other industries. And we look to other industries for inspiration. It's not hard to kind of look outside of construction say gee we could take that and apply that to our industry and do something with it. And so one of the things that crossed my attention were these robots that basically this this outfit the same company had developed those Spheroes for those Star Wars people out there, you know BB-8.

He's a Sphero with a little cap on him. I was at a conference in Denver five years ago. It was a startup thing and it was a lot of investors there and it was just new ideas and these guys were demoing their little Sphero balls running around, you know, you can control them. So they had built this little robot kind of cool and whatever, you know, two years later there it is in Star Wars and like I saw those before they were anything. Well, anyways, these guys their experience working with Lucas Films and Disney and what have you.

They really learned about characters and personality and all of these human attributes and they're like we should put that in our robots. We should make them a little more human a little bit more personal and so they've developed this this new robot. They call it Misty. They were selling it. It was kind of a buy it early and it will arrive at Christmas like for half price or something like that.

And I looked at my wife and I said, you know, when I was a kid, I never had the personal computer. I had a handful of friends. They did that we go to their house and we play, you know, it was like a Commodore or a Tandy or whatever. Is he's really old? But I look at those kids that had those devices those, you know, they were the only ones on the block with it.

And what are they doing today? Computer engineer computer science. I mean, they're hot in this technology thing and they just got a passion for when they were like six. Right. So I got a five year old three year old run around like, you know what robots were doing this and what I really loved about this particular robot is the approach that they're taking.

They said, look, think of everything you wish the robots could do and or you were promised we were promised back in, you know, watching the Jetsons or whatever. Right. Exactly. Where are they? They're not here yet.

Their premise is look, the problem has been we haven't had a stable platform to develop from. And so what they've done is they've gone ahead and put together this robot that's got some tremendous capabilities from a sensory perspective. It can go into a room and it does a point cloud scanning. So it knows its position because it needs to navigate. It has facial recognition.

It has voice recognition. The ability to sense the direction of sound. This robot will turn its head actually to look at you if if you're talking to it. So it's got a lot of these sensors for inputs, but it also has some output. I mean, obviously can drive around.

It can find its way from point A to point B. It's got lights. It's got its eyes, if you will do turn into kind of a display and so it can provide output and can also it's got audio so it can speak and the way these guys look at it. It's kind of like your iPhone the very first one, which was this powerful little thing, but there weren't really very many apps for it, right? And then what happened is they said let's have developers build a bunch of apps and suddenly the value of this device that we now all have, you know, some version of a smartphone will go through the roof.

But the problem is developers need a stable platform that they can call upon to get the inputs and to provide the outputs. But once they have it, let them create what they're calling skills skills for Misty. So the whole thing right now, they got this 10 year plan. The whole thing is let's put out this platform. The stable robot that does it's robust and it does stuff, but it's kind of skillless and let's give it to developers and let's give it to elementary school kids and let's give it to whatever and let them start playing with it and give it some skills.

And we're going to do this in the cloud in the community and you can upload your skills slowly. But surely this device is going to start having it's just like the app store. You're going to have all these apps and you can download it and you can suddenly give your robot the skill to I don't know measure a distance from point A to point B with a laser point. I don't know. I mean come up with something and I'm thinking about this.

I'm like, so this is really taking robots the hardware piece and solving that problem for me. So all I have to worry about is software. Okay, I know software. I'm going to do this. Got this thing and it will be an early Christmas present.

I've already started playing with you know some of the coding that they've got there and trying to figure out. Okay, what are we going to do with this thing? And what's the coding for kids? You can drag and drop, you know if statements or whatever in it. So they got it set up for that, but they also have it written works for JavaScript.

So if a JavaScript programmer can you know make this thing sing and they're working on other languages to to use as well. But I love the premise. It's like let's build a stable piece of hardware and let's just let the world figure out what they want to do with it. I got this thing it's coming and I've already started dreaming of things that it could do in the construction site. I'll give you one example.

So today you go on you go after they've poured the concrete slab and you know, we've probably put the fireproofing on the steel. You got this open space. I'm talking about a building here. The next thing that has to happen somebody's got to go in there and do layout. Yep, they're going in there with a total station.

They're snapping chalk lines. They're painting. They're doing their stuff to set this stage for everybody else coming in who's going to pull dimensions off of it. Yeah, you got your MEP layout. You got your layout for your walls specifically.

Yeah, all your framing. Yeah, so right right and great. That's that's awesome. But if you just had this thing that rolls in it already knows its position knows exactly where it is on this floor and it just starts drawing the lines for you based off of what's in the drawing right based off the bin model or whatever base right. It just improves the accuracy eliminates the need to have to do that and actually expands what you could probably do.

So the thing could work overnight and then the next morning you show up and it doesn't just show you grid lines. It shows you where the walls are. It shows you it shows you more lay out the 2D on the floor. You got it right. I'm thinking why hasn't this been invented yet?

Why no why it's because it's it's that you need the marriage of software and hardware and right now both of those things have to be done by the same entity and that's hard. So if somebody is already taken care of the hardware piece and we just have to come up with software for it pretty easy. So there's a company I interviewed called Locus Robotics. They have a robot that helps a person to pick pack and ship cool. It brings the items to them so that they can walk straight there.

They don't have to pick up an item. They don't have to do anything like that. It's novel. I mean what I understand is a person typically in a pick pack shipping location. They would be walking 12 miles every day just to do their job.

This robot reduces the amount of time that they're walking on the site to four months or rather the distance they have to walk to four miles and I was like what a great metric. I mean of the other time so as you're talking I'm like okay think about all the ways you know just simply from walking around reviewing that type of thing. I mean literally somebody is is laying it out. Like you said I mean we have the laser we have the snapping of the people still spray painting you know so much.

That can be improved we can improve it. We can focus on things that rather than spend time doing the kind of the mundane standard stuff we can change the role of the person to do the more complicated more important things value add activities planning ahead looking forward. I mean rather than not having any time and just reacting to things actually taking the time that we get from technology and applying it to things that add like you said add value. One of the one of the key things that I think we as an industry just need to probably agree on is this goal of one day kind of eliminating jobs with a robot.

I think it's foolish. Frankly I often compare what's happening in the health care industry and how they're using technology and no one's talking about you know one day we're not going to have doctors one day. We're not going to have health care professionals. Because you know intrinsically that there's a skill there there's human reasoning and there's things that people bring to the table that we don't want to put in the hands of robots. But there's a lot of stuff that they stick they can still do I mean they can scan inside of me and see oh gee you got a whatever herniated disk or you got a you know a hole in your heart valves or whatever it's amazing what they can do and yet no one's talking about replacing the skilled labor.

We're talking about augmenting them. I think we need to be doing the same thing and thinking about the same thing in the construction industry augmenting what our skilled labor does so they can do more and do it in a safer fashion. That's less risky. But at the end of the day they're still needed and they still bring things to the table. Some folks are in a completely other camp the ones that want you know driverless cars and that sort of thing and I'm like why you know I see both perspectives but I do agree with you in that when you're delivering to people.

Yeah a product right. Something that they're going to live in or work in we spend 90% of our time in built environment right you need to have a human and a human perspective on that delivery the entire time if we transition over to AI to determine how we should be living and moving around in a space. It takes away the value from that human right so it's about human centered design. It's about making sure that human can be the entity where essentially the value is coming from right because that's the only way we're going to continue to have improvement in the experiences that we have in built environments.

You're absolutely right these things that we're living and working in there put together once. I mean there are buildings that are very similar and but they're not identical. There's always little nuances and little things that are different even if they're exactly the same design because the way they get implemented but more often than not they're not even similar in design they're completely different and that difference is we actually relish it. I mean we like spaces that are a little bit different and you know it would be very bland if we just kept walking around and everything's the same and so that's not going away and people skilled labor that's what brings that to us.

I mean they actually are the ones that ultimately implement these designs that are kind of flights of the imagination for certain you know for architects or engineers they kind of come up with these concepts and then but at the end of the day somebody has to figure out how to put it together. Right so I mean so kind of segmenting you know this is how my brain is working human centered design for the built environment for people to obviously live and work and play in but human centered design around the way people are working so the design of the word.

That's that portion you're talking about with the skilled labor so how do we support the humans that are working you know so that they can be enabled to do their best work. One of the things that we had it was a short lived debate that touch plan when we first started and the debate was this look we could probably have the computer do an analysis of your plan after you've made it and tell you. What you need to do to improve it to optimize it right because it's just it's logic it's math it's you give it a set of assumptions and if those assumptions hold true then.

It holds that this solution that it's coming up with is an optimal solution I mean you can you can do a cause and effect on this thing. And we had this internal debate do we do this or not and he said you know what no we're not going to do it and here's why we're going to focus on building a system. That enables people to work together and to collaborate and to come up with these solutions quickly on their own and kind of plan learn from it replan replan replan and within 2 3 weeks maybe a month maybe 2 months if they're following the process the team.

They're going to find the optimal solution that works for them on their own they're just going to continuously improve and get better so let's give them a platform that enables that improves that collaboration kind of a central place to remember what we've all agreed on. And we can apply our lessons learn and we can look at it and and and draw information from it rather than focusing on having the computer tell us what to do because nobody wants to be to tell him what to do nobody and and so it was a short lived debate.

And here we are so you know that's been our focus and we're like we're not looking back and some people ask us I had somebody show up at our booth this the other day and so does this do tech plan I'm like yeah this tech whatever you need to do you can do it and we'll show me. And what they were looking for was push a button and have the system say hey to improve your flow move this here move that there do this do that and I'm like we're not doing that we've got it in set up in an environment where you can go in and you can do that.

Do you tack planning as a team create standards duration tasks have business rules between you in terms of how you're going to move things around and then do it and then see the result and then from there you can tweak it what have you but we're not going to optimize it for you. It's it's it's that buy in it's it's effective team building it's really the process is it's that collaboration right buy in right you might have the best plan but if nobody believes it or wants to do it or doesn't work for them.

Then they're not it's not going to get executed don't believe that it works for them right they have to come to that right that's part of the big process and it's a departure from kind of where mocha you know our home company where we started because we actually we were spun out of MIT and what we were doing is computer simulation of construction. So we actually did before I even really got into this whole lean thing I was trying to get the computer to build a building and give it the inputs and figure out the optimum solution and bring that the folks and say hey you know yeah you got a good schedule here but if you do this this this and this you're going to have a better schedule people like oh interesting and sometimes they did it but most times they just they did a few tweaks when they you know because they didn't come up with right.

So I think it was a you know it was a lesson learned as much as as a as a debate internal we had experienced the other side we can't do that so we kind of come full circle and recognize the real the value that people bring to the process is instrumental and they need to be part of it because at the end of the day they're executing it and like you said they need to have that buy in so our whole objective is let's just give them a framework that's kind of human makes it really easy for them work together to talk together to plan together and to learn together and.

And then ultimately they're going to come up with something that's optimal and that works for them and it's going to be better than not doing that that's absolutely right so anything else you like to share that I can edit here now but it's always fun talking to me I think we got to do some more of these I agree but it's great actually doing it face to face your tech is much better than my tech in terms of doing this stuff well thanks again Michael thank you always a pleasure talking with you it's awesome appreciate it if you liked this episode check it out at constructor.com slash LC.

I Michael car that's LC I am I CH a EL CA are if you learn something valuable in this episode share it with your friends and colleagues because also let me know if you enjoyed our discussion by connecting with me on Twitter at Brittany underscore CT or find me on LinkedIn you can also email me at Brittany constructor.com that's BRT and IE at construct double R. Next week you'll hear more from the LC I congress recap we'll be talking about feedback on the congress and plans for taking it further one step at a time if you haven't subscribed to the podcast you can do so at your favorite podcast player I'm looking forward to sharing a continuation of this recap next week.