okay everybody i think uh we're ready to begin i first of all want to thank everybody for joining us today i think it's uh it's going to be a really very very interesting very very timely discussion uh for uh for us hi i'm jay myers i'm the ceo of next generation manufacturing canada and as everybody knows our uh our objective is building world leading advanced manufacturing capabilities in canada uh in doing that we're we're supporting projects at the cutting edge of industry 4.0 but we're also looking at how do we enable better deployment of advanced technologies uh in manufacturing operations across canada uh may not be leading edge applications uh but that the adoption the deployment the profitable deployment of technologies and uh and of course the business requirements the skills requirements that on the part of companies in order to successfully deploy technologies uh is a key part of what uh of what we're aiming to do and supporting the uh the sector as a whole uh so i'm really really glad to be joined today by uh by a fantastic panel of experts i'm going to introduce them uh in a minute and our our guest speaker i want to thank uh first of all salesforce for their uh their sponsorship for uh today's session let me say a few things uh i guess in terms of some of the overall trends that uh that we've been seeing we've been speaking about industry 4.0 for uh for some time and you know many many companies today many manufacturers are collecting data maybe from different types of uh of equipment from different uh systems from different operating systems but collecting collecting data nevertheless um they're connecting uh systems they're connecting with each other uh as well connecting uh across factories i think the you know we're we're well into industry 4.0 but we really haven't seen the uh uh what can be delivered yet and turn especially in terms of better systems of identifying what data is important to uh to understand what data is important to collect what data is important to analyze in order to achieve better performance improvement how to begin to predict outcomes uh and and in processes and then uh alternatively or maybe sometime down the road how to uh build in greater autonomy uh not only in processes but uh but in products as well and uh to some extent we're already seeing that in many uh consumer products uh today so here to tell us uh where we're going uh where where some of the opportunities are where some of the challenges are uh and uh and maybe giving some some really great advice uh a long way in terms of what it takes for successful deployment of uh uh of uh uh smart factory technologies to um uh to improve uh uh digital and production operations we've got a great um a great team here first of all let me pass it over to uh peter coffey uh peter is um uh with salesforce uh our our sponsor for today's event and uh and i can tell you from uh from our discussions previously with peter uh a great perspective on on the industry as a whole uh but particularly on some of the opportunities that uh that manufacturers uh have and and also some best practices in terms of successful uh deployment uh and implementation practices so peter over to you thank you thank you very much jay um and if someone can confirm for me that the slides are showing they are excellent um it's so important to emphasize what you've just been saying that we need to go beyond traditional notions of mere efficiency uh peter drucker of course has famously said there is no greater sin than doing the wrong thing better and cutting the cost and improving the speed of going in the wrong direction is is not a win even if your performance indicators may tell you you're doing a great job my team at salesforce occasionally uses the the comment that we don't predict we reveal i'm not here to show you any vision of the future here so much as i am to illuminate where we are and what the current states of change are so we can kind of do the math in our heads and say you know these are really not predictions we're making this is a future that's already unfolding as we speak and in fact the world economic forum has been very good about eliminating some of these specific ideas this is a little bit of an eye chart it's got an awful lot of things that may not seem to be in the wheelhouse of the group that's with us today so let me focus on two of them the evident issues that have been illuminated during the last two years of pandemic but that really have a lot of of origin and likely continuation and things that are happening globally of demand uncertainty and disruption and the already before the pandemic things that were underway in terms of generational changes demographic change demographic change is a perfect example where you're not really making a prediction if i know how many 20-year-olds are graduating from college today i have a pretty good handle on how many 30-year-old experienced workers i'm going to have 10 years from now it's really just doing the math so these two combinations of the external environment around manufacturing and the talent pools available and the opportunities and challenge that they represent to manufacturers i really feel are two of the things that that deserve most of the um the attention that i'll be uh trying to share with you now mackenzie has been very clear that we shouldn't think of the last two years as in any way exceptional except for the particular cause of the disruption being a a novel pathogen that by any measure and i mean measure very literally the imf world uncertainty index federal reserve board geopolitical risk frequency and intensity of cyber attacks natural disasters driven by climate change and i do not want to get into a discussion of whose fault climate change is it is merely a visible reality that these things are happening this level of rising volatility in the environment the army war college in the us uses the term vuca for volatility uncertainty complexity and ambiguity these these four things drive a need for discussion and again mckenzie is observed we spent the decade pre-pandemic with a general ethos that the idea was to optimize your supply chain for minimum cost we built just-in-time supply chains that spanned oceans and when one boat got stuck in the suez canal some of the fragility and brittleness of of the resulting highly optimized systems was certainly cast into a bright light the idea as again mckinsey suggests that you should anticipate a significant multi-month supply chain disruption roughly every four years and build systems with the resilience to handle that is is somewhat different from what has been being trumpeted in the magazines and taught in the business schools uh during the last several years or even decades and the human component of this really can't be overemphasized the rapidity of platform and process change means that today a newly graduated engineer needs to be spending equivalent of hours or even weeks per year on ongoing study merely to stay current in their field the idea that your college degree was essentially a capital asset that you would depreciate over the life of your uh career is doubly mistaken because one the skills are changing more quickly and two the length of the career is itself greater so that's a compounding issue 70 percent of employers believe that employees need continuous education and training and i'm not talking about taking people off the line for one week a year to sit in the classroom being told things maybe 80 or 90 of which they already know or maybe even know better than the instructor and the rest of which they don't yet know why they need it we need to introduce education and training in line into the workplace that as people realize they have a need there's a micro moment that's available they can be supported by technology that's literally in their hands and upskilling needs to be a continuous process a continuous flow and not a batch process that's done at high cost with low effectiveness another really key observation is that when employees feel that they are a investment by their employer that they are being upskilled and made more valuable over time their likelihood of remaining with an employer significantly rises and today keeping the talent you've got is increasingly a challenge the mobility of the workforce their ability to discover of their opportunities their ability to take that job somewhere else because they can do it remotely all of these are really key as we look at opportunities like natural resources in the more remote parts of north america we realize that a lot of these jobs will be done by telepresence by people operating remotely call them robots if you will by operating machinery they don't need to be on site to operate their ability to change employment is going to be more frictionless than it's ever been before and yet my friend paul dougherty over at accenture has said half of companies while agreeing that their people need more skills only three percent are increasing training investments and that's a that's a gap that's just enormous we know why people aren't increasing training investments because training looks like a pure cost that takes people off the the production process the idea that we can infuse continuous upskilling continuous availability of new knowledge directly into the workday is a novel opportunity and one that i think is probably not being fully pursued by most manufacturers who don't really appreciate that this is a reality this is a future that's now the resources are in fact at hand and i'll show you what i literally mean by at hand the convergence of i.t and operations technology that can the convergence of things that used to be considered back office or sales force tools and of course the company's name literally is salesforce but the idea that these technologies can become relevant throughout the production process that the cycle can be in fact a closed loop where instead of production creating a product that is sold by sales and supported by service well now this is a continuous circle where the product becomes a channel for delivering an ongoing stream of services and the consumption of those services becomes input to the next generation of product design this is a holistic integration among the units of the manufacturer that replaces what was previously much more of a pipeline we use the word pipeline a lot in business up until now for sequential you know handoff to hand off to handoff it's really got to be much more of an integrated and continuing operation from now on another key point is that this is no longer monolithic effort by a single major firm to build and sell and support a product this notion that we're going to be building products that are kernels of ongoing delivery of services a simple example would be something like the ipod which when it was first introduced entered a marketplace that had many companies making digital music players as products and they left it to the user to go out and find the music to put on there sony thought it had an edge because it owned a large content library which is thought of as a proprietary feature of its digital music players and the radical notion of apple breaking open the device to make it just the kernel of the ipod plus itunes ecosystem and pulling in all of the existing content providers in a way that had really never been attempted before let alone achieved this idea that you build an ecosystem that redefines a market and i do mean literally redefines because three years after the first introduction of the ipod the word podcast was introduced by the oxford english dictionary or i should say recognized by the oxford english dictionary simplifying deployment is key it can't take two years to roll out a solution to a problem that's probably not going to be recognizable as the one that was originally envisioned at the time that that process began the idea that we need to move to continuous flow as readily as you upgrade the the apps on your phone now and think about that because only 10 years ago getting an unscheduled update to a piece of software was on the supply side a confession that someone had made a mistake and on the consumer side an irritating nuisance that was going to get in the way and now we've made that essentially an ongoing cadence of continuous improvement bringing that into the industrial environment is still a work in progress but it clearly is achievable once we realize that that is the definition of success another key idea is not to have massively centralized systems but to move as much of the intelligence to the edge as possible one of my favorite examples is that when a tesla is driving down the street and its machine vision sees a pothole and dodges around it well okay that's an impressive little piece of micro ai what's less visible but in many ways much more interesting is that the next tesla coming down the street doesn't need to see the pothole it already knows about it and this has been described by people i know who are intimately involved with 5g as the critical understanding of why 5g is not 4g made faster 5g is a distribution of control seeing what's being built as a computing fabric that moves data when it needs to do that instead of a dumb network that has everything collecting the data in the middle the reason this really matters in industry and manufacturing environments is that we all understand that having dozens hundreds thousands of devices telling the center every tenth of a second yes everything is still fine having vast quantities of yeah everything is still working fine traffic that does not scale what we need to have is systems that can be relied upon to notify the center when something interesting is happening to have information shared out to the edge so that local decision making can be made and you know an example i often offer is this if you've ever touched a hot object and jerk your finger away and then notice that your finger hurts you've already seen this happen there's that little local loop that goes fingertip to spinal cord and back the so-called reflex arc and then after the finger is out of trouble a much slower signal can propagate its way up to your brain and you can say ouch and then a few critical fractions of a second later another slower piece of your brain can say that should never be that hot i'm going to file a trouble ticket just in that one moment you've seen three different levels of network interact to do simple things quickly without involving the center at all to get the the the problem under control but then to have an opportunity to add cognitive value and centralized consideration so that you can you know tell everybody else whatever you do don't touch that thing for the next hour we're working on it this is just something that happens every day well this is what can happen in factories as well a tool is not a dumb object anymore this is something that we worked on with bosch when they introduced their iot cloud to say you know a hand tool is really the end point of a process and collecting data off the tool so you can then do something like say okay this tool is running a little bit hot it's still within tolerance but it's been getting hotter we should look at it before it has to be taken out of production these are ideas that have been around for decades w edward stemming and statistical process control if you find that on average a drill that is used to drill 110 holes begins to drill them out of tolerance well you start replacing the drill bits preemptively at the 100th hole well now we can do this in a much more dynamic and intelligent way instead of relying on old static numbers like that and be able to roll a service case automatically created when the data says it's likely to be needed at some time in the near future and this is an actual literal tool that you can put in a worker's hand it's got a barcode scanner in the base it's got a web browser built in at the top and when the worker goes up to a workpiece it scans a qr code and it tells the worker these four bolts need to be torqued to this level they do that operation which operator torque those bolts to correct spec on which workstation on which day and time is automatically logged and all sorts of issues of correct recording of correct performance of operations quality control issues that we've wrestled with for decades can now be dramatically made less less frictional and and more accurate and these are platforms i was over in berlin for the um launch of this and we and we actually did a hackathon for a hand tool say what kind of applications would it be useful to put literally in the worker's hand well you know people get bored and start making mistakes when they do the same thing over and over again we can actually have the tool command changes in sequence of operations where they don't change the the outcome just to keep the worker engaged we can gamify these operations we can let people you know see leaderboards for who's maintaining the highest combination of quality and throughput there's so much opportunity to have the intelligence leave the back office and be diffused throughout the operation so workers can feel like they are actively engaged in a system that's doing better work and manufacturers recognize this 86 percent of manufacturers believe that smart factory initiatives will be the main driver of competitiveness within the next five years again this is a future that's already underway 83 of them believe it's going to change the way products are made this is uh data from a study that deloitte contributed substantially to and they expect that they're going to be spending the money to do this because this is not an expense to be minimized this is an investment to be pursued for the leverage it brings in taking historically razor thin profit margins and bringing the the modern factory up to if not apple level profit margins then certainly much more profitable provisions of value-adding services over the lifetime of an evolving product so the objectives for this in digital production operational resilience growth with automation enabling the visibility uh i sometimes say digital transformation is a very vague phrase but we can unpack into four very specific ideas connection so you can get the data awareness that you have some notion of what the data should mean for you s for smartness that is to say using machine learning and other tools to make this scalable and finally trust where people have to rely on the data and be prepared to take action based on so connected aware smart and trusted c-a-s-t it's a simple acronym for what makes digital transformation actually happen and it creates new streams and this is key eighty percent of the money that's spent on on initiatives called digital transformation has been estimated to really be targeted at cost reduction and streamlining of existing operations but jeannie ross at mit has said the point is that you have to take the technology and create new value and not merely cut the cost of the kind of value you delivered before because otherwise you're trying to succeed in the digital world with a pre-digital value proposition manufacturers understand this they think the returns are going to be there they are going to make these investments among the things we've learned how to do during the last few years is how to create a digital headquarters to make it possible for more and more people to contribute from anywhere for more people to do their work remotely or to do it with more flexible work hours and there are any number of ways in which you can do this i want you to think about one thing that's probably familiar to all of you try to cast your mind back several decades before we had spreadsheets and when if you needed to do number crunching at scale you had to go down the hall and talk to someone who probably wrote cobalt the democratization of access to that calculation power allowed people on many desks and in many mobile environments to start asking and answering questions about math and now we can start asking and answering questions about a much richer variety of things including production rates quality materials specifications and so on and this is something my friend dion hinchcliffe has observed that we need to move from the old notion that workplace automation tells people what to do to remind ourselves of how transformational it has been when people have tools put in their hands that let them do more that let them use what they know from frontline knowledge and be scalable and visible and making contributions that really get them excited and make them want to stay with you longer and share what they've learned interestingly our own research at salesforce has found that people don't fear automation when they are the ones creating it when someone says okay i've done this same thing five times today if i can find a way in 30 seconds to make that an automatic process i don't have to think about that again i can build a little automation on the spot you do it when you do a spreadsheet formula well now we can do this in tools like slack and other workflow environments so an automation is created as readily as learning a new way of tying your shoes ai plays such an important role in this and so often ai is feared by people as something that's going to replace them or overestimated by people as something that's going to give them things that candidly really still only exist in movies but this is an example that i recently ran across the complexity of the board game of go is legendary and it's been characterized as having a so-called game tree complexity of about 170 which is pretty big number the game tree complexity of sailing in america's cup yacht is about three thousand it's far more complicated and to train the crews the elite competitive sailing crews they built a simulator with some machine learning components to it so the sailors could sail against the simulator and practice some of the tactics and gambits that are used in competitive sailing and something happened the simulator started generating original strategies and the sailors would look at and say hey that's kind of cool we never did that and they started studying the simulator for insights into new competitive strategy this interplay between what people do well and what machines do faster and what machines do without preconceived biases about what they think they know doesn't work this is a different way of thinking about what the ai we actually do have this is not fantasy stuff this is stuff that can be deployed now is being deployed now and the idea that we can introduce this into real world environments there's nothing much more real world than being out there on an america's cup yeah that's that's the real world right in your face i think this is really exciting and it's a different way of thinking about what ai can do we want to be contributors to this and i'm looking forward to the discussion with the uh the rest of the group here i think people think because after all our company name is salesforce they think that that's what you know what we're there to support the idea this is really about moving data putting verbs in there as well as nouns capitalizing on the power to build ecosystems and the ecosystems that we build with many other companies like i've already mentioned bosch for example getting the intelligence out there literally in the hands of the production worker and into the distribution and supply chains really dealing with that world that mckenzie is showing us we will be seeing of a high challenge high volatility growing uncertainty this is an opportunity that's right here and the tools are literally at hand to do things with it and i'm so pleased to have a chance to discuss these things with with the rest of the group that's here today and i i thank everyone involved in making that available to us for what's about to come thank you for the chance to be with you that's super peter thank you i think you're your focus on getting the information the intelligence to workers so that they can do their job more and better uh is is really key uh as well as this is about making better decisions uh and fi better faster decisions more highly qualified decisions and and even better prediction as you say not only to improve productivity and improve uh efficiency but uh to drive more value as well and it's really connecting that to those business objectives uh that is uh is really really key so thank you very much for um for that presentation and i and thank you for also for joining our panel so um peter is the vice president of strategic research uh at salesforce and he's going to be joined uh by three good friends of mine uh penguin san cao who is vice president at ats um paul boris who's the president and ceo of premo and craig holden who's the ceo of uh uh penivo and uh perivo is a company that we've had the pleasure of supporting and uh implementing uh some very very innovative systems optimization uh tools and uh in agri-food so i i'm going to turn it over to sean uh now to moderate our panel and uh and thank you all for for participating looking forward to the discussion great so thank you very much jason and as we get everybody up on the screen so i am going to uh disappear turn my camera off because this is about the panel uh but i thought i would just make sure we take a couple seconds here first what we're going to do for those of you joining us the panel discussion is going to go on until about 2 15. during that time i'm going to be monitoring the q a box which you should see at the bottom of your screen if you have questions please insert them there at that time at about 2 15 when i asked the last question here of the panel i'm going to turn it over and i'm going to ask your questions of our panel so again use that q a box as we go forward to make sure and if we don't get to your question we're going to capture those questions and share them with our panelists so we can make sure we get a response back to you okay so the q and a panel i'll be monitoring that and with that said i'm going to turn my camera off so it's just a panel that's highlighted and i'm going to start so this let me get into this here you'll still be able to hear me it's kind of neat a technology so i'm going to start with our first question now i you know it's always a kind of trying to figure out who's the best person to ask the first question of but really i decided to go alphabetically because that was the most fair way to do it so craig you're our first question for the panel so let me ask you the question why is a smart factory important for manufacturing from your perspective well thanks sean and thanks for having me as well uh nj um so yeah i thought i'd probably best way to answer this is just to provide some context first and help with some of the terminology that we're seeing in this space so the term smart factory factory 4.0 and industry 4.0 are all essentially interchangeable terms so how did we get to industry 4.0 well this this started with the first industrial revolution the steam engine uh so on steam being utilized in factories for example for like steam powered looms for weaving fabric then industry 2.0 was the electrification of those manufacturing processes which was a massive step change in in productivity and performance following that industry 3.0 that's the automation of those systems the introduction of robotics pcs and um erp systems just in time systems so again a spectacular step change in in performance and productivity so now interest rate 4.0 which involves things like the industrial internet of things deploying millions of sensors implementing ai machine learning that ranges into 4.0 ranges from simple digitalization so taking a single sensor connecting that to an iot gateway sending the data to the cloud and using that data to make decisions in real time and provide information uh on on performance or perhaps some predictive analytics it ranges from a simple implementation like that to a fully adapted factory uh connected and linking to the supply chain having ar robots running around fulfilling orders etc and also a big differentiator for industry 3.0 is the incorporation of customer data in decision making so with that in mind why is the why is the smart factory important so those productivity gains we've seen previously from the from the various transitions we're going to see them again we're going to see improved performance better d bottlenecking less waste and less unless you than ever so here's is a a warning but also a big opportunity so so imagine going to a steam driven factory and trying to sell electrification when that started or going to an automotive plant with manual assembly processing and trying to sell robots and getting getting pushed back that sounds crazy now but humans humans don't like change now factory 4.0 we're going to see another step change in performance and canadian manufacturing core needs to not only embrace factory 4.0 and smart factory but also lead on it thanks great thanks craig um now let me just i'm going to ask our other two panelists for their perspective because this is a pretty broad question um so peng do you have anything to add relative to why is smart factory important for manufacturing from your perspective sorry there you go uh yeah absolutely i think it all boils down to like every corporation we're you know making profit and that means how do you allow your employees to work smarter and more efficient and to what craig has just said you know the automation aspect it's up to this point has all been very um compartmentalized so you have the front-end automation you have the back-end automation nothing's really connected and the whole point of a smart factory is to be able to get your factory to the point where all these systems you have in place are connected and why do we need to do that well it all boils down to uh those who are able to react to the fast changing market conditions um able to win more business you know if you're able to uh predict what your uh consumer is going to buy and consumer i think in many ways getting more fickle right they want they want what they want when they want and the excuse of we can't deliver is not acceptable so um i think for anybody who wants to win more market shares or win the uh the rat race um you don't have a choice but to get your factory um you know introduce the manufacturing intelligence that's required to allow you to adapt to market conditions and secondly i think what's happening with the global supply chain that we're seeing is that um the old way of setting to cheaper labor country in high volume shipping to your local market that's also going out the door and there is a you know we're not going back to a factory full of employees doing hand assemble or whatever the case may be so again adapting smart factory be able to link all that especially with the labor shortage that we're seeing really it's the only solution forward i don't think we have a choice so if you want to be successful you want to succeed i don't think it's uh not adopting smart factory is an option anymore great thanks peng so paul what would you add to the the question why is a smart factory important for manufacturing sure look i've i've been in this space for a long time uh i've been blessed to be in the manufacturing space for a long time and you know i look at it a little bit differently i think all the things that have been mentioned already are critically important but i think about a smarter factory right in these operations you know there's discipline teams there's a lot of data being accumulated there are a lot of systems there's a lot of new visibility it's about taking those tools and really moving them to the next level and i think peter covered off on some of that things when he talked about training people right you know everybody i don't know how good you can get in excel and if you want to get trained on excel you can do it online for free but these teams have really run into a wall of productivity if you look at the saint louis fed they got a great website for economic data they've got a chart where they show productivity moving very briskly growing for decades and for the last 11 years 12 years it has been flat to down so we've added systems we've added sensors we've added data we've integrated things and i think we've missed the opportunity to engage these teams these continuous improvement teams and to hand them a new set of tools and i picked up a little bit of that in i was taking notes vigorously as as peter was walking through his slides because you know that's what he's talking about is how do you get all these parts and pieces together how do you create a new uh interaction for these teams that are ultimately going to drive the performance of these operations so you know i i think it's a smarter factory i think this concept of a smarter factory is going to go from important to absolutely critical because of all the disruptions that are coming down the pipe and and how people need to need to look at these things so we got to get them past this wall that they've bumped into and and get some new tools in their hands i think great thanks paul so now we're going to kind of i'll open it up i want to make sure everybody got a chance here to kind of start with that fundamental question so i'm going to open it up to the panelists and just ask the second question which is what are some examples of how you've seen or maybe experienced or deployed some smart technologies in manufacturers does anybody have any examples you could share yeah go ahead paul sorry peng uh so first of all i feel like it's one of those game where whoever can hit the sound i feel like i'm on hollywood squares nobody knows what that is if they're if they're not my generation uh peng would you like to would you like to start okay so so first thanks to ngn and thanks to to salesforce for doing this i was it was kind of you know ruminating a little bit on these questions this morning and i shot a note to peter and it struck me as interesting that uh my impression of salesforce and where where salesforce started was by being able to put tools in the hands of the artisan the sales person and their and their management team and and give them better visibility to data that they actually already had i don't know that there was a lot of new data they got in there but it was the ability to swipe a card and actually uh turn on the system that got them this better tool this better ability to be productive so i think if we look at where we are right now a lot of these ai and machine learning initiatives tend to get bottlenecked at the top of an operation at the at the board level where we're going to build peter's got great slides on this he didn't cover build a real foundation spend 18 months 24 months building a foundation and then we try to build something on top of that pyramid foundation and it it may not match so i'll give you a specific example that's the point of this question where we've seen this this actually work is uh we're working with it with a tier one automotive company they've got a well-deployed modern mes manufacturing execution system they've got controls and automation they have very complex process uh that does multiple steps it's it's i think of these things as artwork when you watch these machines running and they had come to us they had a bottleneck they said we need to clear this bottleneck we're really confounded by the data uh we there's there's so many different streams of data coming in we can't analyze them all at the same time and they said we think we're missing data we said no we'll take a look at the data that you have look at any new data what we were able to discern ring out of the existing data they had with zero preparation was there was a fault that occurs that their mes is not tuned to trigger on and that their controls and automation are not tuned to trigger on so the machine learning engine in the background goes through churns through all of these different sets of data as is it's it's kind of acting like this continuous improvement mentor in the background and it taps on the shoulder and it says hey pay attention because when this occurs and this occurs you have a negative consequence and we start to see this trend occurring someone needs to pay attention how that manifests itself with the team is now the team has taken data they already had they thought they had fully exploited and you know these are smart people they're not lazy they're not not doing their job they're working hard every day trying to trying to ring every dollar out of everything they have the the machine learning engine is allowed is pointing them in a direction then the ci team that continues improvement team launches on that problem they address the problem then they run that as a program launch across all of these similar operations and then the machine learning engine is sitting in the background listening for some variation some deviation from what they've now established is the new baseline so data they had systems they already had zero bandwidth constraint from their engineering team put it into the machine learning engine pop these insights out and people sit there and they say okay that's we need to go and check that yes check find these problems yes these are issues we need to make some changes to the to the control limits on this particular part of the process we need to understand how mes works with automation and in a matter of quite literally weeks they had done an ai project that had a tactical profound was the word their engineer used impact on their operation so think in terms of that how quickly can you spin that flywheel up how quickly can you get these learning turns right just like inventory turns learning turns where people say i think i understand what ai can do for me now in operations if it could do that why don't we go over here well i don't really have the right data or i don't have it collected in the historian it points you to where to go next uh and and that's to me not trivial and very typical for the types of interactions that we've seen so it's a i think it's a great example it's a great way for people to get started quickly thank you thanks paul it's so pang you were second so we'll let you uh share yours here as well sure i think for me there's probably uh three examples that kind of define smart factory and kind of where it's going right it may not all be there yet i think the first one i mean talked about is i was in brazil about 10 years ago and uh happened to be working with an agra company a multinational agricompany and they were showing me or you know uh what john deere was uh doing and this is ten years ago and john deere um as i said is no longer a equipment company if you can believe it but more of an i.t company because they have all these infrastructure in place um to monitor their equipment in uh uh lan and farms to uh you know track um the number of seats that are being planted and be able to give these really critical data back to the farmers you know how their crops are doing what's the water that they need it like it's it's too complex for me to totally understand but you know it goes back to this idea uh if you think take what they have done and apply to factory it's a very similar idea is adapting all these different technologies the equipment itself the uh intelligence behind the equipment the gathering of these data and to be able to analyze these datas to mean something to improve your operation or you improve your reaction to customers whatever the case may be so that's one example the other example um is a technology that my former company developed at ats acquired called symphony and we basically again adopt technology to replace a lot of uh mechanical system and that allows it to be extremely flexible it's really agile and you integrate that with um smart factory technology again it's just to be able to capture analytics that executive can look at to be able to react to their equipment before it becomes a problem because you don't want to stop your production that's very costly and i think the third example which um relatively new technology but very exciting is um out in the u.s a entrepreneur is working on using bitcoin as a background technology to develop um again uh smart factory uh software uh analytic and the reason the bitcoin pit that is important is because you don't want to lose the history right so now it's going further into making sure that that information is never lost so when you start looking at all the different pieces that are being out there that are you know that's happening and taking place um there's a lot happening i think what's missing right now is how do you how do companies pull it all together so that it is end-to-end integration and again going back to what i said earlier from the factory floor to the front end to you know reacting to consumer what peter talked about with salesforce how do you pull all of that together i think is the next big challenge hey peng sorry can i ask you to expand on just that thought you talk about this end to end and getting all this information available does it take an event a bad event for them to say hey now i get it or are they spending that money ahead of all the the the events occurring do they have a real defined roi sorry i hope you don't mind me jumping in there but could you touch on that you've got a pretty pretty interesting purview on that yeah so uh paul i come from the back end right i've been in automation for 24 years and been very passionate about it about this sector um i i don't i think the challenge should be blunt and i've said this right this whole idea of the industry 4.0 great concept great idea i'm all for it but the problem is that unless there is a big enough organization or government mandated where it's forced um all these different players to share the same um you know ip protocol where all these are able to connect i think it's really really challenging because everybody kind of talked everybody does their own little pieces of the puzzle right so even when i see like l'oreal and and um in linkedin talked about uh working with ibm and their they call it their iot and you know they're all talking about how exciting this is for them well that's great but it's still just a front end it's still just reacting to consumer how do you connect that data then to let's say their automation equipment and if you stay with your traditional automation equipment of very mechanical system how do you react to the smart these uh consumer demand that you know i want a lipstick or whatever that is just my color how do you create automation that can produce in that kind of volume bluntly you cannot if you stay with the same old traditional very mechanical system right you just can't it's just react it's just a reality because a our labor force is um it's harder to get technicians and engineers and all of that but how do you change mechanical stuff that's that quickly you have to go to machine shop blah blah blah and that's why this technology that ats acquired call symphony is so critical because the mechanical cam is all replaced by digital cam so you can actually change your robot motion by on the hmi you don't have to buy the two axis robot or the six axis robot you have an anaxia robot and so it's it's a decision that i think is harder for small business because where do you start i think that's another question we can touch on later harder for larger corporation because it require sorry hard also for large corporation because unless the ceo has a vision of saying this is where i'm going all right let's put together a team end to end what do we need to do everybody's operating in their own silo so they're buying software they're buying hardware they're buying stuff that works for their silo but it doesn't work for the entire organization and bluntly a lot of these decisions are long term so if you're operating quarter of a quarter based on what your promise your shareholders and your share price how do you make decisions that's going to be impactful four years five years down the road and not next quarter because these are not next quarter decisions and so there are a lot of challenges regardless of how large your organizations are because they're different challenges make sense yeah so thanks paul for asking paying and thanks paying for that i think that's really good because the more you talk about shareholder value the more you really get to the root of this because in some situations there is a cost associated with this and there needs to be value produced uh craig so same question for you what are some examples of how you've seen or experienced or deployed uh smart factory technologies yeah thanks yeah i'll give a few examples and i really like paul's framing um of the opportunity how would you make a smart factory smarter if you if you want to google forbes and pepsico machine learning you'll see an article there with some great examples and one of them is taking lasers which might have have been used to detect product quality and the simple nargo is to integrate machine learning into that process to uh perform better quality checks and to move away from a simple like go above no go from evo's perspective working with engine and on our project which is currently underway we've developed a real-time lean manufacturing platform where we'll deploy sensors uh at a facility or take existing data from the plc network uh we'll implement iot gateways law cost and gateways and so it says low-cost solution to collect and create existing data we'll send that to the microsoft azure cloud in in canada and then it's sending back data to operators on on tablets or hmas or or pcs at the facility virtually in real time the delay to the cloud and back is like one second or so plus perhaps a second for the firewall so that that ability to use the cloud to make true real-time decisions is here and you know this is early days uh for us we've been live for a year or so with this system we are already processing over 10 million messages a day so you can see how this is going to scale and to put servers local at a facility and to make maintain them to do this kind of service across multiple facilities yes you could do some edge computing but the cloud is a is a great solution to this so we're we're at the stage where we can report and understand overall equipment thicknesses in real time and we're looking at integrating machine learning for use cases such as production scheduling for example how can that be optimized the data sets are just too big when you've got tens of thousands of product skus you just can't do that so machine learning is a great use case for that and we're also utilizing machine learning to um help optimize utility performance at facilities so services at facilities like refrigeration systems compressed air steam boilers etc through traditional energy conservation measures and energy management systems you might be able to yield cost and utilita and carbon emission savings of eight to ninety percent of of like global best practice machine learning comes in to squeeze out that last last few percent uh what we've not been able to do thus far is understand how we integrate blockchain into our business model but if any of the panelists have got any smart ideas for that of all these things thanks craig i was thinking as you shared how many messages per day are being captured the first thing i thought of is maybe this is a good tool to manage my inbox so we're going to ask the same question now of just relative to you know your own examples if you will peter that you've experienced where smart technologies are being deployed well i've never seen a more inspiring customer story than the one from tough shed that's t-u-f-f they make you know backyard storage buildings and you might have seen them it at you know a home depot or a lowe's or some such place they turned their model inside out from making a variety of standardized buildings in in certain sizes and so on to doing an online configuration tool where customers can come in pick out different elements put together a bespoke design and purchase it at a price competitive with the standardized buildings you don't get that by adding automation to your existing silos you get that by saying wait a minute we're going to invent a different customer experience we're going to do what alvin toffler literally was writing about 50 years ago in future shock when he used the expression mass production in quantity 1 in one of the most low-tech products you can imagine and create a completely differentiated idea where instead of competing on price and quality with other standardized offerings they were able to become genuinely unique in the mind of the customer but crucially that story begins with one change leader at tough shed writing a letter to the ceo saying this is my my my date left off resignation i want the opportunity to do this if it doesn't work fire me someone took the initiative to say i want to turn everything we do inside out change our criteria for what we optimize for from producing our product line at minimum cost to changing the definition of what we sell and that's that's what it takes to to get people out of the detent i mentioned l'oreal l'oreal you know you know how cosmetics are sold you've got the counter at the department store where people come in and you'll try on the different shades during the pandemic they went to a completely online tool for being able to try on different shades on pictures of yourself that they estimate they got done in eight weeks when without the acceleration the pandemic their own estimate is that it would have taken them three years to get that done so the accelerations of the pandemic are real the definition of a different customer experience driving a recombination of your existing silos rather than an improvement of their individual performance as peng was saying is is absolutely key and paul you talked about the fragmentation of data our own estimate is that in the pre-sales force era and working with deloitte on this they came up with this number the typical company if it really wanted to know everything it's ever done or has ever hoped to do or is has is currently doing with any given customer would literally need to look in over three dozen places there's be an email thread here a marketing campaign response there a spreadsheet over there an erp record over here and so on bringing that all together into one view of the customer with different facets of that relationship visible where they need to be for design for production for distribution for after sale service recognizing pandemic acceleration and beginning the question with what's the unique experience we want to offer to our team and our customer not what's the lubrication we can put into the gears of our existing machine is how that happens and it's why the label of smart factory does mean more i think peng and craig have both been emphasizing this it means so much more than adding better technology to the current status quo model but visualizing the opportunity for a genuinely different model great thanks appreciate that peter um so the next question i'm going to take the second part of the next question i know everybody prepared but i think it's important let's talk for a minute about pitfalls now i know we're talking about a very broad topic and it's it's kind of hard to just jump on specific pitfalls but i guess the question is if if you were a manufacturer today and you're considering uh you know you want to move towards a smart factory uh you want to you know talk we've discussed different ways and means by which we'll introduce this technology what are some things you would suggest that manufacturers need to be aware of some potential pitfalls some areas where they might fall down that you could give them some tips or strategies to to avoid whether it be a delay or a costly mistake that they might make and you know if you have an example to tie in with that even better so i'll ask because i changed the question slightly is there somebody who wants to to start with that hang go ahead yeah so i think i'm going to use my own personal experience because that's safe if it's me um thinking back when i was ceo of uh transformix um and you know as we were trying to adopt technology and in this particular case was uh erp uh we were still a small company and you know i thinking back probably the greatest pitfall we had was not appreciating the deep learning curve that our whole team had to go through and the hidden cost of um having to bring their company expert in all the time and that far exceeded the actual hardware cost so i think that would be my cautionary tale to um companies that are looking to adopt different technology to try to bring all the different pieces of the puzzles together to be able to build this smart factory is you gotta be more forward like more forward thinking in term of how do the different technology you want to adopt gonna integrate together do you have the resources what's the training program is this something that is um user-friendly or is it a constant um having to bring in their expert because it's so unique to them and you know even though we we actually spent probably a year analyzing all the different uh software out there in hindsight that would probably be um the most challenging was the the how long it took to adopt okay appreciate that pain craig you you had uh raised your hand there what have you found or what would you suggest what are pitfalls or things to be aware of yeah uh again i think i'll revert to paul's making a smart factory smarter because if you've not got the infrastructure into place to support um some of these initiatives we've talked about and you've got the management buy-in it's just not going to go anywhere and you're not going to get the benefit of it either so um looking at the process of big data machine learning making decisions uh based on that so how how do you get the data out of this out to a facility well if you've not got a sound um infrastructure in place and a sound um like network plc network historian if that's where you're storing the data um gateway to get the data off the facility wireless network et cetera if you've not got that in place the data is is going to is going to fall down that that's feeding into your into those models into that decision making and then um just taking big data and pointing machine learning at it is not the way to go is that have we got data integrity that's the first you know first question can you guarantee what you're getting is is is sound and that might be you know a wrong statement might be uh it's it's just playing wrong and it's the the tolerance song um good enough there's dare to drop out etc which feeds back into the into the network issue so you've got to you've got to um before you start machine learning for example uh it's making sure that the data um set is sound but also really well defining the the problem and what you want to get out of it so it might be that you can you can generate the majority of savings by optimized process control for example so you don't need a ai algorithm to you know to tell you to turn the valve into a particular particular way and then finally i'd say you're going to get out of it what you're what you put in again having the management buying top down and engaging operators and everybody across the business is certainly where to go to ensure success great thanks craig paul what are your perspectives on potential pitfalls so i'm sitting here trying to think about how to politely indicate that i'm going to be a contrarian here right i think this is this is a tribute to salesforce and engine for bringing to people bringing together people with some some different views look at if you're if you're altering your interaction with a customer and you're going to make a single tube lipstick and that's going to come through uh and be manufactured and shipped on the same day that that's i think one aspect of the issue you know peter talked about go being a 170 and sailing being a 2900 i've been in manufacturing long enough it's got to be you know you know maybe an order of magnitude more than that because you have the plan and then the doors open and you get punched in the face and it used to just be it's snowed or it's uh deer season or something and now it's it's snowed as deer season and we just tested positive at the front gate and somebody slipped through that wasn't tested and now we've got all of this other stuff so we create this complexity and and where i see the pitfalls and remember i might i i had the the real pleasure of working at one of the largest industrials cio for advanced manufacturing strategy i think i mentioned that there were there's a lot of people who tend to think you can kick the door open to the plant and say hey pay attention we're here to transform you we're going to make things better i know this i've i've got a xyz mba and i'm glad you know in it it's it's logical there's training there's an understanding of a business model you're trying to deploy but what they generally miss is that if i don't have an unruly team if i have an unruly team and bad management there's nothing here to do for you get your team squared away get better managers if you've got some level of discipline if you're shipping parts into an automotive supply chain you have discipline you have rigor you have processes they may not be perfect they might may not be all aligned i think and and we've bet on the fact that you can you can take those pockets of the lowest level data and i think craig was i think craig this is where you're heading a little bit if if what you have to do is harmonize all that data it's going to take you forever to get one little pocket of data that you can run machine learning against our approach is to say leave it as it is where it is vacuum up whatever you got maybe it's on a thumb drive maybe it's on wire before we start this transformation because now we're going to point you to what process needs to be impacted what area needs to be affected and i think the the pitfall that i see is you know i'm not going to tell you who they were but i mean they're really smart people and they're they're great industry analysts and it was a webinar i was on and they said well first you have to do a readiness assessment across your whole country company and then once you've got that you've got to pick the problem you're going to solve and then you've got to harmonize all the data and all you've done is you've built bias into a system you've embedded it that you're never going to get out if you know the machines running and i'm shipping the product and i know what good product looks like i should be able to exploit that data instantly the tools are not widely available to do that that's where we've had our head down doing those things and and when i popped my head up the other day and just jumped on this webinar and i'm listening to it and i'm hearing man we talked about this in 2005 when i was at lighthammer going to sap we're having the same conversation then i look at the thread chart and i go i kind of get why we're flat right we got to take a little bit of different thinking so the pitfall is to to make the leap that the teams are not equipped that the teams aren't capable of doing something if we could equip them with a little better tool what if we what if we do a little experiment what if we salt in a little bit of technology not technology like a new machine or a new system but a way to analyze and kind of wring out value from this data that's our argument and then you know what step to take next is it this direction to add more sensing is it this direction to add more connectivity is it this direction to to actually just scrap the line because you just don't have the capability to run that line a particular fashion so i i it's a great risk that i say that one of the pitfalls is listening too much to the experts on the webinars and not the experts in the plant and say what is it that you're missing what tool do you need if i could grind up that data and pop out some insights tapping on the shoulder could you do something with it do you have enough discipline and rigor in your processing system to do it because we've seen time after time that that's capable now that doesn't diminish what craig's talking about which is the ability to get that data and actually make it easier to run through that data because that's going to be a critical step maybe the first step if you have nothing and what peng's talking about which is a better process they saw the symphony stuff i mean it looks incredible the ability to say send a signal and say here's how i want you to manufacture that and have a system reconfigure itself but those are you know are they on the edge of the curve right and most everybody else is kind of i'm still running a process that i that i need to be able to make better every given day so you're you're you're smart so uh uh one one one comment i'll leave you with jeff and mel said this we're storing a small meeting and he said look just remember that whenever you hear the experts writing or blogging or talking you're probably as smart as they are on the tip on the topic that you spend your day focused on don't don't necessarily drift off and just take what people are saying find ways to do really quick low-risk experiments that let you again spin the flywheel of continuous improvement up those learning turns and that are going to actually impact the process so that i think is a pitfall is if we say yep the plant isn't going to do anything we're going to wait for another 18 months until the foundation is built so that we know what tools and technology i'll be able to access and drive i i have to do something tomorrow i mean i've got to make this place run better the next shift uh so so don't lose the focus on what's what's going on uh right in front of you that makes sense great thanks paul appreciate that uh peter your perspective on kind of some potential pitfalls to be aware of yeah if i could offer the shortest possible versions of two stories from two different domains when i was at the baton rouge chemical plant and we were working with some machinery that handled you know a form of rubber that had a tendency to break down during production and leave the stuff that we technically called a goo all over some of the rotating equipment one of the ways to clean the equipment is to run ice through it which made the rubber get get hard and then kind of flake off so having an ice machine next to the extruder was something he needed to do when we were building a new one and i was in a meeting once where someone was saying yeah well we've got the specification for the ice machine almost ready to go out for bid and one of the guys just blew up and said it's an ice machine they have one next to the elevator and every holiday in what are you doing writing a spec for it you buy this out of a catalog and the lesson to be shared is we're in a world now of interoperable services cloud capabilities repurposable components the tendency to do this the way we've done this before of thinking we need to do a an elaborate bespoke solution when in many cases composing things out of readily available and highly configurable services is a novel approach and and doing it the old way as well as it can be done is doing in fact the wrong thing the other story is from oil and gas field services where we were talking with someone who said you know one of the problems is people keep wanting to say and now we can put the equivalent of a laboratory analyzer at every wellhead and we know exactly what chemicals are coming out of the ground and the guy raised a sense that at the beginning of the day i got a guy in a pickup truck and the question is out of the 10 square mile area for which he's responsible should he go north south east or west to get the maximum amount of work done that morning all he really needs is red light green light is the pump pumping or is it not pumping because that's what's going to really control where his time goes and that i think speaks to the point that's already made of get as close to the front line as you can and find out what relatively small number of pieces of data delivered reliably in a timely way and an understandable way will have the maximum effect on getting actual work done instead of allowing people to get really excited about all the the bells and whistles that they could conceivably inject into their quote smart factory on boat great thank you peter appreciate that so we're gonna move into the last question and then open up for some q a we do have a couple questions that have come in here but before we do so and based on our time i'm gonna term this the rapid fire response okay because i want to give everybody a chance for kind of some final thoughts here so the question is if you were sitting in front of let's say which you are some manufacturers and and maybe they're struggling with where should they begin if they're trying to create a strategy relative to their smart factory or developing a smart factory what are some some tips or ideas you would share with them as a good starting point uh and i'll just start well let me start with i'll just work around if that's okay i'll end on you peter uh so that uh so paying what are your uh your thoughts i would say uh just start on whatever it is to help your employees uh do their job quicker faster and be able to adapt to customer demand great uh paul so i i have a really complicated long-winded transformation strategy it's start in a corner and have a little better hors d'oeuvres and music and get people excited about the one thing that you did that was impactful and other people will peel off from every other party and join yours and it creates a viral effect done it i've seen it it works every time so while you have a grand vision you have to have the value that's delivered from that vision but start at the tactic start at a single spot fix that and you want to get people excited and drive things that's that's a great way to go great thanks paul craig yeah i'm fully on board with start small and scale it scale it from there once you've validated it and and created that excitement so to do that you can actually lean on suppliers a lot so for example asking for free trials like yeah great technology come here and prove it and then away we go across the entire facility and the entire business if that does well so i think one of the one of the barriers to making the start is like actually understanding the business case you know how much money am i going to save sometimes that's super difficult to to kind of get over that so doing that that free time trial and having the uh supplier and some skin in the game is is a great way to go about that to to validate that and another way is to get is to get funding so across across canada provincially and federally there's lots of funding that could be accessed to support that uh initial engagement so for example canadian manufacturing exporters cme they've got their technology assessment program in bc where you can do a free lean manufacturing audit of the facility which can help frame some of these issues potentially in in saskatchewan there's the lean improvements in manufacturing program where the uh successful government or fund up to half a million dollar at 50 for process optimization um initiatives so there's some great funding out there that's just a couple examples that you can access to get up and running on this great thanks craig so peter i know you had your hand up earlier there so oh yeah your perspective walk around with a camera and take a picture of every place that you see someone taking their eyes off their work to handle a terminal or do something in a spreadsheet or do something else and say how do we make that a connected data flow in which the data flows in without someone having to manually you know take their eyes off of what they're doing to enter something on a keyboard the device itself should be measuring that data and reporting it in a timely accurate consistent way and the actions to be taken should be delivered as close as possible to the point of that action and every time you take out one of those interface layers you're taking out a failure mode you're taking out a training requirement you're taking out a a frictional barrier to adoption great thanks peter appreciate that all right so we do have a couple questions that have come in uh so what i'm going to do is just ask the panel i'll open it up so first person to hit the go first person to hit the button i guess or raise their hand it'll go to you so we'll start with this uh how are manufacturers finding the right uh collaborators to share and accelerate their industry 4.0 capabilities i'll repeat that again how are manufacturers finding the right collaborators or else even add the right partners to share and accelerate their capabilities specifically in the areas we're talking about and craig you kind of touched on this too right you can get free trials are there areas you'd suggest companies start with uh paying sorry if somebody else did raise your hand i'm sorry peter actually had his hand first i was just going to say the the more front line workers you bring into the room and the more they actually feel like what they say is going to matter we saw this happen at delta airlines for example where they brought the cabin crews into the room to say this is what needs to be on the smart device that's going to help us deliver that better customer experience the more they're in the room and being heard the more likely you are to have not only the right thing being built but a more rapid adopt adoption of it once it has been built thanks peter hank did you want to yeah i think what peter said with respect to bringing key employees from each division to figure out what is it that your organization need but outside of that i would suggest maybe to form your own um you know support team of other executives from other organizations i've done that in my past life when i as a ceo that was you know relatively young to the game and i had a ceo club where it's just half a dozen of us that we meet regularly to talk about specific issues so you could form your own you know sort of smart factory clustered of um other executives and share ideas that way as well to learn from each other's mistakes craig did you have uh any thoughts on that yeah i'm surprised you didn't want to answer that sean i would have thought the ideal answer was to ask engine become an engine member and leverage the engine network we would love that but i didn't want to leave with that now member and somebody who's gotten involved you can speak to that experience so thank you for that comment uh paul any any thoughts on that question yourself yeah okay same thing and jen i mean there's there's some good groups out there there's some there's some good events but you know people are gonna suffer from information overload buzzword overload um you know put people on the bubble find somebody tell them you want to talk to them for 15 minutes and then hit the buzzer and throw them off the phone if they don't understand operations if they don't understand the challenges i mean you you need to you need to dig around a little bit and it would be great if there were better uh better ways for manufacturers to find them but if you see a story that resonates if you see a vendor that resonates with the types of problems you have reach out and and uh like i said but but time limit right don't don't lose half your day uh taking a sales pitch from somebody if they can't get to it in 15 minutes pitch them yeah that's great appreciate that paul all right so what i want to do the last question here was very similar to that it was you know finding people that can assist you in moving forward with your journey and and i think we've kind of addressed that here i mean it really depends on what journey you're taking uh obviously and and and and a variety of other organizations and associations who are plugged in are a great place to start because you can connect with smart folks like the panelists we have here today who can help you along the way and give insights and in some cases actually assist so i'll leave those questions uh i think we've answered those to the extent necessary so what i want to do just based on our time here i want to give you give yourself peter uh you know give us some some closing thoughts based on what you've heard here today what you came you were our keynote so i want to turn it over to you for the next five minutes or so just to share some thoughts and ideas well if i can go back to one of the things that craig was talking about which is kind of thinking of the sequence of accumulation it hasn't been a succession of technologies in the industrial revolutions it's been it's been an accumulation we made power scalable with steam we made it distributable with electricity we made it intelligent with electronics and automation and now we're making it guided and focused with introduction of ai these are cumulative revolutions and smart manufacturing uses all of them effectively but if you took that original 19th century first industrial revolution factory with the shafts overhead and the pulleys and the belts and everything driven by one massive steam engine and you unplug that steam engine and put in an electric motor wow cleaner quieter cheaper everyone would say what a great success it was and yet the basic architecture of the factory would still be those overhead shafts and pulleys and all the machines going on and off and running at the same speed during exactly the same hours of the day you wouldn't really have gotten what electricity could give you you'd only have used electricity to do the old things somewhat better and eric brennelson at mit and lauren hit it pen asked is the same thing true for it and they did a really thorough study of dozens and dozens of companies that you know the companies that reorganize to use the new capability to do things in a different way produce hugely better results than the ones who merely introduce the technology to do the things they've been doing but to do those somewhat more cheaply and efficiently and so the first line of the first slide i shared today about beyond efficiency is so key you have to be prepared to look at what you do through different eyes of given this capability how would we have visualized what it is we do here and the other thing i want to add is there's a tendency to think of the labor component and the technology component as both being ingredients that get pulled in and the customer is someone outside the idea that the people who work for you are a customer that you are selling them the package of pay and benefits and they're paying for it literally with the best years of their lives and you're competing for that customer who could take the best years of their lives and go spend them somewhere else that's a completely different way of thinking about what we're trying to do when we create a work environment that makes people feel engaged the appeals to their sense of being of mastery that people get actual pleasure from doing things well and knowing they're doing them well and creating the factory is an environment where people feel empowered and and enriched so they will in turn give you their best insight and be be eager to say you know we could be doing this differently and not feel they're going to be smacked down this really ties a lot of this together so smart factories include and i think uh this is very much in line with what paul and and penguin both been saying is that smart factories include and totally depend on smart leadership and and and not aggressive vigorous adoption of technology to do the old thing better and if we take that away and say what could i do tomorrow without new head count without new capital budget without permission from anyone a simple question of who don't i talk to who don't i listen to what assumptions am i making every day without even thinking of them as being assumptions those are all things that we can do on an individual level and the the sum of those things creates amazing organizations that that redefine their industries great thanks very much peter so i want to take the last few minutes here since we have a few minutes and and maybe this is a surprise to the panelists but thinking maybe give you a minute if you want to say some closing comments um and before we do that and then i'll thank everybody but i do want to take a moment obviously to thank all of our panelists here so first off obviously thank you peter for your presentation today it was great shared some a lot of great insights thank you to craig paul and pang for your time i know you're all extremely busy so on behalf of everybody that's joined us here today as well as zenjen i want to thank you for your time uh so just some closing brief obviously so some closing thoughts uh from today's discussion paul yeah sure so look uh uh you know you can assign this to the shameless plug category right uh we're an organization that takes that operational data as is as it exists in the operation brings additional value from it and then uses it to drive this uh acceleration of continuous improvement and uh you know we've not worked with any company that had that doesn't have enough data doesn't have reasonably disciplined data that can drive it and it helps create like i said that small initiative that has the opportunity to go viral just today one of our customers uh that we were working this process with uh we got an email that said we've had a very significant issue occur uh would you have been able to determine this and the answer is yes and now everybody's scratching their heads saying geez i wish we would have gone faster so this goes back to the start now start small start fast my faith my saying is do something that's small enough that if it's that's large enough that if it's valuable it's going to have an impact and small enough that if it doesn't work we can all bury it out back and get to keep our jobs and if you think that way and if you think that you've got some some issues i mean this is an exciting time for manufacturers so and thanks thanks again to the to the panel and thanks again to salesforce and jen for for allowing me to join thanks paul uh paying some closing uh comments in the next minute or so sure uh yeah so following up on paul thank you and jen for inviting me and the audience for uh showing up um i don't know how much more to add to what peter and paul said other than you know maybe because of my entrepreneurial background and created r d company is no matter where you start the next time you do it you're gonna learn from that and you're gonna be that much better so get started um and i think you know peter touched on a really important point there can be no changes until leadership's willing to take chance there can be no innovations until leadership's willing to say you know what we may fail and we're not going to punish those who fail i used to say jokingly but kind of serious to my staff you know when you make the first mistakes what did you learn because it's going to make us that much better the next time right that's innovation that's anything you want to do if you make this the same mistake twice either you're an idiot or i'm an idiot so let's not make me like let's not make either of us an idiot so let's learn from that so whatever it is that you chose to do whatever pieces of your entire business depend on where you are in your journey and the size of your organization pick something that is going to be impactful that you know no matter what it is you're going to be one step ahead even if you have planned for five steps and you gain one you're still further ahead so just start on the journey great thanks fang and we've got about a minute and a half or so here craig final or closing thoughts yeah thanks i'm gonna actually go back in and answer one of the uh previous questions that didn't get covered and that's what what success have you experienced so um a large uh sector focus for benivo is the food and bev sector and a lot of the food and bev sector is actually operating a factory 2.0 so our challenge or or the opportunity for canadian manufacturing is how to transition uh food processors from 2.0 to 3.0 to to 4.0 so if you if you're for example a blower molding facility you might have an oee score overall equipment effectiveness of like 99 to something you know you're tweaking that last few that last 0.1.2 to squeeze out that last bit performance if you're a factory 2.0 food and bev processor your oe might be 50 so you can by implementing lean manufacturing principles and starting that transition right up to factory 4.0 you can expect double digit percentage savings so there's just a phenomenal prize effort um for canadian manufacturing in particular the food and best place of operating in the great sport of canadian manufacturers without journey great well thanks craig so i and peter i'm assuming we're kind of good with uh unless you have anything else to add you're good perfect well listen once again i want to thank our panelists today paul peng craig peter i really appreciate your time we appreciate all of you for joining us this video this recording will be made available to anybody who couldn't join us and i want to wish everybody a great rest of your day we'll see you again next time take care everybody thank you thank you