TribalScale First Name Basis - ROI on AI & the Future of Supply Chain Management

00:44:13
https://www.youtube.com/watch?v=vAbRpy9J5fs

概要

TLDRThe conversation explores the transformation of the manufacturing and warehousing sectors, emphasizing the transition from manual processes to digitalization and AI. The speaker reflects on their experience at Nestle, noting the importance of data management and the benefits of automation in enhancing operational efficiency. They discuss the challenges faced by mid-market manufacturers and the significance of change management in adopting new technologies. The discussion highlights the future potential of AI, including the use of digital twins and the impact on sustainability. Companies are encouraged to take an incremental approach to digital transformation, focusing on data-driven decision-making and strategic planning for AI integration.

収穫

  • 🚀 Digital transformation is essential for modern manufacturing.
  • 📊 Data management is crucial for effective decision-making.
  • 🤖 AI integration can enhance operational efficiency.
  • 💡 Smaller manufacturers should start with data cleanup.
  • 🔄 Incremental change eases the transition to digital processes.
  • 🌱 Sustainability can be improved through AI optimization.
  • 📈 Automation frees up labor for value-added tasks.
  • 🔍 Identifying a single source of truth is key.
  • 🛠️ Digital twins offer new opportunities for optimization.
  • 📅 Timing is important; consider off-seasons for implementation.

タイムライン

  • 00:00:00 - 00:05:00

    The speaker reflects on their extensive experience in warehousing and manufacturing, noting a significant shift from manual, paper-based operations to digitalization, particularly in the last three years. They emphasize the importance of AI and digital tools in making operations more efficient and accessible, especially for smaller companies.

  • 00:05:00 - 00:10:00

    The speaker attributes the rapid changes in the industry to the decreasing costs of digitalization and the chaotic business environment caused by COVID-19. Companies are now more focused on eliminating inefficiencies and allowing staff to concentrate on value-added tasks rather than mundane ones.

  • 00:10:00 - 00:15:00

    As companies embark on digitalization, the first step is to clean up data to establish a single source of truth. This leads to automated dashboards and KPIs, resulting in real-time data access and labor hour savings, allowing teams to focus on more strategic initiatives.

  • 00:15:00 - 00:20:00

    The speaker discusses the benefits of digitalization, including labor cost savings and the ability to tackle new projects. Companies can either reduce headcount or redirect labor towards higher-value tasks, ultimately driving growth and efficiency.

  • 00:20:00 - 00:25:00

    For mid-market manufacturers, the speaker advises starting with data cleanup and leveraging the agility of smaller teams for quicker decision-making. They highlight the decreasing costs of AI and digitalization as a potential opportunity for these companies.

  • 00:25:00 - 00:30:00

    The speaker notes that technology advancements have improved supply chain synchronization, with systems now integrating more effectively. Automation in warehouses is also increasing, with electronic yard management tools and automated unloading processes becoming more common.

  • 00:30:00 - 00:35:00

    The speaker emphasizes that automation should not necessarily lead to job losses; instead, it can allow experienced labor to focus on training AI and enhancing product design. The changing labor market may also lead to a natural reduction in traditional factory roles.

  • 00:35:00 - 00:44:13

    The speaker suggests that companies should start their digital transformation journey by identifying inefficiencies and establishing a single source of truth for data. This process should be gradual to ease employees into the change and validate the new systems.

もっと見る

マインドマップ

ビデオQ&A

  • What are the key changes in the manufacturing industry over the years?

    The industry has shifted from manual, paper-based processes to digitalization and AI integration, improving efficiency and decision-making.

  • How can smaller manufacturers approach digital transformation?

    Smaller manufacturers should start by cleaning up their data and identifying their one source of truth, then gradually digitize their operations.

  • What are the benefits of automation in manufacturing?

    Automation can reduce labor costs, improve efficiency, and allow staff to focus on more value-added tasks.

  • What role does AI play in the future of manufacturing?

    AI will help optimize operations, enhance decision-making, and improve sustainability by reducing carbon footprints.

  • How should companies manage the change to digital processes?

    Companies should take an incremental approach, involving employees in the process to ease the transition and validate new systems.

  • What are the risks of not adopting digital transformation?

    Companies risk being left behind in a competitive market and may struggle with inefficiencies and higher operational costs.

  • What is a digital twin in manufacturing?

    A digital twin is a virtual representation of a physical system that can be used to optimize operations and train AI.

  • How can companies ensure they have enough data for AI?

    Companies should identify critical data streams relevant to their operations and gradually build a database for AI training.

  • What is the significance of sustainability in digital transformation?

    Sustainability can be enhanced through AI by optimizing processes to reduce carbon footprints and improve resource efficiency.

  • What should companies consider when implementing AI?

    Companies need to define their AI strategy, understand their data flows, and determine how AI can best fit into their business model.

ビデオをもっと見る

AIを活用したYouTubeの無料動画要約に即アクセス！
字幕
en
オートスクロール:
  • 00:00:04
    you've had some incredible experience uh
  • 00:00:06
    especially in the warehousing side,
  • 00:00:08
    manufacturing side. Um how have kind of
  • 00:00:10
    you seen the industry change throughout
  • 00:00:13
    your entire career, you know, working at
  • 00:00:15
    Nestle, working manufacturing and and
  • 00:00:18
    what are some really key changes that
  • 00:00:20
    you'd love to share with people getting
  • 00:00:22
    into that? When I first started uh
  • 00:00:24
    Nestle, I would say um especially on the
  • 00:00:28
    operation side, a lot of the stuff was
  • 00:00:30
    still pretty much manual um paper based
  • 00:00:33
    um and then however I would say based on
  • 00:00:37
    tribal knowledge for one of a better
  • 00:00:39
    experience of individuals. Correct.
  • 00:00:41
    Right. Right. Um, as the years went by,
  • 00:00:45
    I think going digital became more
  • 00:00:47
    important and became became more the
  • 00:00:50
    norm and and that kind of accelerated, I
  • 00:00:54
    would say, in the last three years I've
  • 00:00:55
    been there where we really move full
  • 00:00:57
    tilt into digitalization of our
  • 00:00:59
    information and our and our records and
  • 00:01:02
    everything. Um I would say the events of
  • 00:01:04
    the last few weeks with uh you know
  • 00:01:09
    deepseek and the suddenly the
  • 00:01:10
    availability of an open AI model another
  • 00:01:13
    one that's that's performance-based as
  • 00:01:16
    well as a lower cost of entry. I think
  • 00:01:18
    that kind of says pretty loudly that AI
  • 00:01:21
    is here to stay and it's making it more
  • 00:01:23
    accessible for everyone.
  • 00:01:26
    uh smaller companies now probably have
  • 00:01:29
    better a greater chance of getting um
  • 00:01:33
    financially um acceptable access or
  • 00:01:36
    access at a financially acceptable cost.
  • 00:01:38
    Yeah. Um the programmers now have who a
  • 00:01:42
    whole new playground which they can
  • 00:01:45
    experiment in. So I think now I think
  • 00:01:47
    the doors are open. Um, and frankly, I
  • 00:01:50
    think it's now up to companies,
  • 00:01:52
    individual companies to decide, hey,
  • 00:01:54
    what's my AI
  • 00:01:55
    strategy and then from there decide what
  • 00:01:59
    their implementation game plan is. Yeah.
  • 00:02:01
    No, that's fascinating. And and on that
  • 00:02:03
    point, you mentioned there's been lots
  • 00:02:04
    of changes in the last three years. Um,
  • 00:02:07
    you know, I'm curious really to get your
  • 00:02:08
    insight. Why why do you think the last
  • 00:02:10
    three years have been so significant?
  • 00:02:12
    What's been happening? I think the uh
  • 00:02:16
    the cost of digitalization has come down
  • 00:02:18
    as well, right? The ability to go in uh
  • 00:02:20
    to digitalize has come down. I think
  • 00:02:24
    um things have become a lot more chaotic
  • 00:02:26
    from a business environment. Uh I mean
  • 00:02:29
    with the onset of COVID and stuff, we
  • 00:02:31
    saw
  • 00:02:33
    um swings in uh both uh uh the costs of
  • 00:02:38
    of commodities as well as just the
  • 00:02:40
    overall business environment become more
  • 00:02:41
    chaotic. So everyone's looking for an
  • 00:02:44
    edge, right? Uh how do I take
  • 00:02:47
    inefficiency out of my operation, right?
  • 00:02:50
    Uh how do I make things uh take out the
  • 00:02:53
    mundane task out of my organization so
  • 00:02:56
    my staff can focus on the the more value
  • 00:02:59
    added stuff or the or the the big world
  • 00:03:02
    changing events that we need to focus
  • 00:03:03
    on? Yeah, of course. Yes. And that's
  • 00:03:06
    interesting you say that because you
  • 00:03:08
    know we're very much immersed in those
  • 00:03:10
    environments. Speaking with
  • 00:03:11
    manufacturers across Canada, we're
  • 00:03:12
    hearing a lot of those same same things
  • 00:03:14
    as well, especially when it comes to
  • 00:03:15
    reducing inefficiencies. Um so from a
  • 00:03:18
    technology perspective, I I know that
  • 00:03:19
    you touched on AI. Um but you know, I
  • 00:03:22
    think a lot of manufacturers um they can
  • 00:03:25
    be very tangible people as well, right?
  • 00:03:27
    So where have you kind of seen a lot of
  • 00:03:29
    these adjustments a lot of these kind of
  • 00:03:31
    waste reducers optimization uh in a
  • 00:03:34
    tangible sense from your experience what
  • 00:03:36
    does that look like? So from my
  • 00:03:38
    experience I would say uh as we as you
  • 00:03:41
    start your journey uh through
  • 00:03:43
    digitalization heading towards AI I
  • 00:03:45
    think the first the first thing that
  • 00:03:47
    would happen is as you clean up your
  • 00:03:48
    data you suddenly determine what your
  • 00:03:50
    one source of truth is in terms of data
  • 00:03:53
    and that's that's a big step. So
  • 00:03:55
    suddenly you get rid of the the the
  • 00:03:57
    noise in your data and your team's
  • 00:03:59
    already starting to focus on what's the
  • 00:04:02
    uh one source of truth to make decisions
  • 00:04:03
    to predict etc. Um the next step is
  • 00:04:07
    obviously the actual digitalization
  • 00:04:09
    where suddenly um dashboards and uh KPIs
  • 00:04:13
    are now produced by systems rather than
  • 00:04:16
    people whether when I say people whether
  • 00:04:19
    it be on paper or keying into Excel
  • 00:04:21
    worksheets suddenly now you're pulling
  • 00:04:23
    this stuff out of data links uh
  • 00:04:25
    automatically and you and you're you're
  • 00:04:27
    doing your visuals uh automatically as
  • 00:04:30
    well. So suddenly now you have real-time
  • 00:04:32
    data or more real-time data um uh done
  • 00:04:35
    with less uh with less effort. So there
  • 00:04:38
    you already start to see uh savings in
  • 00:04:40
    terms of uh labor hours or to put it
  • 00:04:42
    another way uh uh you see labor hours
  • 00:04:46
    freed up so that can be focused on more
  • 00:04:48
    value added stuff. Right. Um
  • 00:04:52
    and then also as well suddenly you see
  • 00:04:55
    uh um your ability to make decisions
  • 00:04:57
    quicker and with more confidence
  • 00:05:00
    increases all that adds value in the
  • 00:05:02
    background right yeah yeah so it's quite
  • 00:05:05
    interesting so it's almost like you
  • 00:05:07
    solve one problem it opens up space to
  • 00:05:09
    solve more problems more resources
  • 00:05:12
    absolutely um you did may mention kind
  • 00:05:14
    of labor cost savings what other kind of
  • 00:05:17
    really benefits are there to taking that
  • 00:05:19
    approach approach. So in terms of the
  • 00:05:21
    labor cost savings, obviously as you get
  • 00:05:24
    rid of manual work uh or manual input as
  • 00:05:27
    well as the reconciliations that happen
  • 00:05:30
    before you before you digitalize,
  • 00:05:32
    suddenly you have your labor has a lot
  • 00:05:35
    you can either as the manager of the
  • 00:05:38
    business have the option of either
  • 00:05:39
    reducing a headcount or in in other
  • 00:05:42
    instances
  • 00:05:44
    uh it will allow you to tackle projects
  • 00:05:46
    probably you didn't have time to tackle
  • 00:05:47
    before. Uh for example, if if the
  • 00:05:50
    enhancements you're doing suddenly frees
  • 00:05:52
    up your salespeople, you know, maybe
  • 00:05:54
    before your company focused on tackling
  • 00:05:56
    the higher end of a market, now maybe
  • 00:05:58
    your sales guys can say, "Okay, we got
  • 00:06:00
    the higher end of the market. Let's tack
  • 00:06:01
    the middle end of the market now since
  • 00:06:03
    my sales team can now focus on that." So
  • 00:06:06
    you can look at it two ways. Either cost
  • 00:06:07
    savings or it gives you uh labor now you
  • 00:06:11
    can focus on more value added stuff.
  • 00:06:13
    Yeah. Right. Or projects that you didn't
  • 00:06:15
    have time to work on before. Yeah.
  • 00:06:17
    Right. or initiatives and strategies you
  • 00:06:19
    didn't have time to look at in detail
  • 00:06:20
    before. Suddenly that that's your
  • 00:06:22
    option. I can now put it towards these
  • 00:06:24
    items. Yeah. And drive further growth.
  • 00:06:26
    Oh, that's incredible. And obviously
  • 00:06:28
    you've seen that at large companies uh
  • 00:06:30
    correct global companies. Do you think
  • 00:06:32
    that that translates just as easily to
  • 00:06:34
    like let's say a mid-market manufacturer
  • 00:06:36
    in North America and you know they have
  • 00:06:39
    limited team, limited resources. What
  • 00:06:41
    kind of advice would you give them for
  • 00:06:43
    kind of approaching something like this
  • 00:06:45
    but they've never done this before? So
  • 00:06:47
    for for middle market or smaller
  • 00:06:49
    companies versus the multinationals I
  • 00:06:52
    would say in their case and and this is
  • 00:06:54
    probably an oversimplification
  • 00:06:57
    uh that for them the first step is
  • 00:06:58
    probably cleaning up the data. Um they
  • 00:07:02
    would probably have a greater tendency
  • 00:07:03
    to have more stuff on paper or more
  • 00:07:06
    stuff in ad hoc informal systems. So
  • 00:07:09
    that's their first gain. Um I think as
  • 00:07:12
    well their advantage over larger
  • 00:07:14
    companies is that with the smaller
  • 00:07:16
    companies you probably the decision-
  • 00:07:18
    making is probably focused in one or two
  • 00:07:20
    people and you should be able to make
  • 00:07:23
    decisions quicker about what direction
  • 00:07:25
    you want to go in. Uh you raise a good
  • 00:07:27
    point about resources depending on the
  • 00:07:30
    strategy.
  • 00:07:31
    um the the the cost of doing what they
  • 00:07:34
    want to do may may
  • 00:07:36
    uh entail significant cash flow outflow
  • 00:07:40
    for them. Uh and in that case, yes, that
  • 00:07:44
    might be a barrier. But as I mentioned
  • 00:07:46
    earlier, with the cost of doing some of
  • 00:07:49
    the AI related stuff for digitalization
  • 00:07:51
    continuing to go down or the potential
  • 00:07:53
    for that to go down now with the
  • 00:07:54
    increased competition, uh that barrier
  • 00:07:57
    may decline over time as well. Yeah. No,
  • 00:08:00
    it's fascinating and and you know
  • 00:08:02
    obviously you have a strong um history
  • 00:08:05
    in the supply chain area as well and you
  • 00:08:08
    understand especially within Canada the
  • 00:08:10
    complexities of that. How have you seen
  • 00:08:13
    technologies um at least recent
  • 00:08:15
    advancements kind of improve or change
  • 00:08:18
    that market? um without really getting
  • 00:08:20
    into all the political events that we're
  • 00:08:21
    seeing kind of a affect this as well and
  • 00:08:25
    um what kind of changes do you think
  • 00:08:27
    that would make for let's say a large
  • 00:08:29
    global company working out of Canada the
  • 00:08:31
    US or even anme in in that case so on
  • 00:08:35
    the supply chain side what I've seen in
  • 00:08:37
    uh until since uh until a couple of
  • 00:08:40
    months before I retired what was
  • 00:08:41
    happening is that
  • 00:08:44
    um systems were being integrated a lot
  • 00:08:47
    more so for example any any company with
  • 00:08:50
    a co-manufacturer for example
  • 00:08:53
    historically it was it was pretty much a
  • 00:08:56
    a paper flow back and forth now we're
  • 00:08:58
    finding we're using uh whether it be
  • 00:09:00
    something as simple as EDI or actually
  • 00:09:03
    formally linking systems we're seeing
  • 00:09:05
    the systems talk to each other so
  • 00:09:07
    there's a lot more synchronization
  • 00:09:08
    between say a company's commands and uh
  • 00:09:11
    and a company's uh and and the main the
  • 00:09:14
    company who's purchasing those services
  • 00:09:16
    right uh on on the transportation side
  • 00:09:20
    where we're seeing digitalization is uh
  • 00:09:23
    um for example in a warehouse uh when
  • 00:09:26
    you to manage the traffic within a yard
  • 00:09:30
    historically that was done manually you
  • 00:09:31
    know i.e. How many trucks are in the
  • 00:09:33
    yard, where they're parked, when trucks
  • 00:09:35
    come in, where do you put the next truck
  • 00:09:36
    in to make sure your yard doesn't get uh
  • 00:09:39
    uh um what do you call it? Jammed,
  • 00:09:42
    right? And and and nonfluid. Now, that's
  • 00:09:45
    being done uh by the basically um uh
  • 00:09:49
    electronic uh yard management tools,
  • 00:09:51
    right? Where where basically a truck is
  • 00:09:54
    uh checks in electronically and it is
  • 00:09:56
    tracked in the yard. um as well related
  • 00:09:58
    to transport instead of giving uh paper
  • 00:10:01
    BS now they're electronic BS right so um
  • 00:10:07
    of course you can uh there is some paper
  • 00:10:09
    like for example in customs documents I
  • 00:10:11
    think there will always still be some
  • 00:10:12
    paper but uh again we're heading in a
  • 00:10:15
    direction where in in in a lot of areas
  • 00:10:17
    in warehousing we're moving
  • 00:10:20
    paperless right um I think uh as well in
  • 00:10:24
    the warehouse itself um
  • 00:10:27
    uh especially the new warehouses um uh
  • 00:10:32
    automation is is beginning to take hold.
  • 00:10:34
    Um once a truck arrives at the
  • 00:10:35
    warehouse, you have AGVs uh unloading
  • 00:10:38
    the truck and putting away the product.
  • 00:10:40
    Um I would I would say though that for
  • 00:10:43
    the older warehouses um sometimes uh the
  • 00:10:47
    physical structure is not always uh
  • 00:10:50
    conducive to automation and uh and
  • 00:10:53
    obviously the AI that drives that
  • 00:10:54
    automation. But uh I would say certainly
  • 00:10:57
    the new warehouses um uh that's that's
  • 00:11:01
    for sure going to happen. Yeah. Um I
  • 00:11:03
    think the technology as I mentioned is
  • 00:11:04
    here to stay. It'll only get cheaper and
  • 00:11:06
    it's now for companies to decide when do
  • 00:11:08
    I jump in. Right. Yeah. That's
  • 00:11:10
    incredible. And I think um you know some
  • 00:11:13
    of the questions we can possibly
  • 00:11:15
    anticipate around automation is its
  • 00:11:18
    relation to labor. North America. We see
  • 00:11:21
    that affecting different manufacturers
  • 00:11:22
    and different degrees. Um, some see it
  • 00:11:25
    as a way to enhance labor, the way
  • 00:11:28
    things get done, making environment
  • 00:11:29
    safer. Um, optimizing efficiencies,
  • 00:11:32
    cutting down time, things like that. Um,
  • 00:11:35
    what would you suggest for, you know,
  • 00:11:37
    smaller companies? I think global
  • 00:11:39
    companies have a very firm understanding
  • 00:11:41
    of where they need to improve and
  • 00:11:42
    optimize those things. But companies
  • 00:11:44
    working with smaller teams, how can they
  • 00:11:46
    leverage automation after defining what
  • 00:11:48
    that might be just the software side?
  • 00:11:51
    Maybe discussing hardware a little bit.
  • 00:11:53
    Um how would they kind of leverage that
  • 00:11:56
    to really optimize how their teams
  • 00:11:58
    operate to drive efficiency without
  • 00:12:01
    looking at reducing staff for example?
  • 00:12:04
    So in so in that in that case um I would
  • 00:12:08
    say their opportunity there is to
  • 00:12:10
    redeploy that labor. Um once they drive
  • 00:12:12
    out inefficiency I think they have an
  • 00:12:14
    opportunity there to use that uh labor
  • 00:12:17
    especially if it's labor that's
  • 00:12:18
    experienced and uh has a lot of uh
  • 00:12:20
    historical knowledge that labor can be
  • 00:12:22
    used to uh further enhance training of
  • 00:12:25
    the AI that that governs these automated
  • 00:12:28
    machines. uh that labor can be used to
  • 00:12:31
    uh help the company seek out new
  • 00:12:33
    opportunities. For example, if they're
  • 00:12:35
    looking at launching a new product, um
  • 00:12:37
    that labor can be involved in designing
  • 00:12:38
    that new product and how best to design
  • 00:12:40
    that product so that it flows
  • 00:12:42
    efficiently through their manufacturing
  • 00:12:43
    system. Right. Yeah. Um so I would I I
  • 00:12:46
    wouldn't always say it's a uh you would
  • 00:12:49
    automatically lose labor. Also keep in
  • 00:12:51
    mind labor market's changing as well. I
  • 00:12:54
    think what we're seeing as well both in
  • 00:12:56
    factories and in warehouses is that uh
  • 00:12:59
    as the the next generations come on less
  • 00:13:01
    and less people are willing to work or
  • 00:13:03
    wish to work in a factory environment in
  • 00:13:06
    the traditional factory environment or
  • 00:13:07
    the traditional warehousing environment
  • 00:13:09
    and so um I think there will also be a
  • 00:13:12
    normal reduction on over time in the
  • 00:13:15
    supply of that labor. So that fits in
  • 00:13:17
    nicely as automation comes in and
  • 00:13:18
    reduces the need. Yeah. And uh maybe
  • 00:13:21
    instead of a guy driving a forklift,
  • 00:13:23
    maybe you have some sort of software
  • 00:13:24
    technician instead, which which which
  • 00:13:27
    people may be more willing moving
  • 00:13:28
    towards. It's incredible. Yeah. Really
  • 00:13:30
    the factory of the future IoT industry
  • 00:13:32
    4.0. Exactly. Um you know, I think
  • 00:13:35
    there's many ways you can kind of get
  • 00:13:37
    lost in the imagination of how that
  • 00:13:38
    would look. U and that kind of leads to
  • 00:13:41
    my next question. One thing that we are
  • 00:13:42
    finding, you know, speaking with
  • 00:13:44
    manufacturers, Canada, the US, um, as
  • 00:13:47
    you're saying kind of from the labor
  • 00:13:48
    perspective, we're finding more people
  • 00:13:50
    are retiring. Correct. And some
  • 00:13:51
    companies are saying, especially to us
  • 00:13:54
    being tribal scale, our objective is,
  • 00:13:58
    you know, either to go paperless, go
  • 00:14:00
    digital, but also to take that extensive
  • 00:14:03
    information of what's in Joe's head on
  • 00:14:05
    the manufacturing floor, correct? And
  • 00:14:07
    bring that into a digital system. How do
  • 00:14:09
    we do that?
  • 00:14:10
    So what would your suggest and what kind
  • 00:14:12
    of focus or approaches or even
  • 00:14:14
    experience um would you make for
  • 00:14:16
    companies trying to leverage that? So in
  • 00:14:19
    terms of what getting what's in Joe's
  • 00:14:20
    head or or Joan's head
  • 00:14:24
    um I think uh I think that's that's when
  • 00:14:27
    when a company takes its first step
  • 00:14:29
    towards digitizing the data that it has.
  • 00:14:31
    I think there's lots of tools uh
  • 00:14:33
    currently that can help them start also
  • 00:14:35
    digitalizing what's in Jet. Um I think
  • 00:14:39
    the use the use of IoT um on equipment.
  • 00:14:43
    So then yeah, for examp I'm just picking
  • 00:14:46
    a very random example. Um, if if Joe's
  • 00:14:50
    if Joe's uh uh expertise was um uh being
  • 00:14:56
    able to determine what products coming
  • 00:14:58
    off the line uh below standard quality
  • 00:15:00
    by looking at it. If the IoT starts to
  • 00:15:02
    track whenever those whenever Joe pulls
  • 00:15:05
    those products off and what's specific
  • 00:15:08
    to that product that Joe pulls off, you
  • 00:15:10
    can start then tracking. Okay. Start
  • 00:15:13
    gathering information that your AI can
  • 00:15:14
    start learning on. Right? you're using
  • 00:15:16
    Joe but you're using the technology to
  • 00:15:19
    record that you know when Joe makes the
  • 00:15:21
    decision what are the parameters in
  • 00:15:22
    place right right so I think that that's
  • 00:15:25
    a big one um I and I re and I'm not
  • 00:15:29
    speaking now from a technical
  • 00:15:30
    perspective but I I came across uh uh an
  • 00:15:33
    article recently on digital twins where
  • 00:15:36
    uh digital twins now are being used to
  • 00:15:38
    to train AI so for example using the
  • 00:15:41
    same uh same example as Joe say Joe is a
  • 00:15:46
    uh quality checker on a line uh looking
  • 00:15:48
    at observing product and and assuming
  • 00:15:50
    what what's good and what's bad. Uh uh
  • 00:15:54
    you can basically make him a digital
  • 00:15:55
    twin and track his every move, what he
  • 00:15:58
    looks at, what he does uh digitally and
  • 00:16:00
    track that all and then have that and
  • 00:16:03
    then determine when he decides
  • 00:16:04
    something's not good and and moves it to
  • 00:16:06
    uh the rejection lead. And again there
  • 00:16:08
    again recording the parameters around
  • 00:16:10
    where Joe makes decisions and you have
  • 00:16:13
    something to uh you have something to uh
  • 00:16:15
    to then train your eye on. And and again
  • 00:16:19
    coming back to people I think Joe will
  • 00:16:20
    always be valuable because I think one
  • 00:16:22
    of the key things with AI is that uh it
  • 00:16:25
    is good to always have a human in the
  • 00:16:27
    loop. Always have a human being to to uh
  • 00:16:29
    sense check. So maybe that's where
  • 00:16:31
    that's Joe's new job now, right? Instead
  • 00:16:33
    of being on the line, he's
  • 00:16:35
    double-checking uh or doing checks on uh
  • 00:16:38
    on the AI so often and and validating
  • 00:16:41
    that the AI is is is working in
  • 00:16:43
    accordance as the way we think it should
  • 00:16:45
    be. Right. Yeah. So almost just guiding
  • 00:16:46
    that technology. Exactly. So you know, a
  • 00:16:49
    human is still in charge. he's not doing
  • 00:16:51
    the the repetitive manual work but he's
  • 00:16:54
    now more involved in in ensuring that uh
  • 00:16:57
    doing being the quality check on the AI
  • 00:17:00
    in sense. Yeah, of course. And I think
  • 00:17:02
    there's all types of questions around
  • 00:17:04
    obviously efficiencies and how safety
  • 00:17:06
    would play out and I'm sure there's all
  • 00:17:08
    kinds of articles that have been written
  • 00:17:09
    on how that would operate um and and
  • 00:17:11
    guiding those systems which is pretty
  • 00:17:13
    incredible. And I understand that you've
  • 00:17:15
    had some experience as well, you know,
  • 00:17:18
    seeing how this automation works and and
  • 00:17:21
    the efficiencies of that. Um how have
  • 00:17:24
    you seen at least from an ROI
  • 00:17:26
    perspective at a larger company um being
  • 00:17:29
    in the industry when that was all manual
  • 00:17:31
    versus to an automation was brought in
  • 00:17:33
    have you noticed a big change in um
  • 00:17:35
    improvements at the overall kind of
  • 00:17:38
    operations and health of an organization
  • 00:17:40
    like that? In terms of the ROI on the
  • 00:17:42
    actual investment I would say yes there
  • 00:17:45
    there has been an improvement. It's been
  • 00:17:46
    a bit up and down uh to be honest. Um, I
  • 00:17:50
    would say
  • 00:17:51
    uh it's it's it's decent when I say that
  • 00:17:55
    decent about 10 years which is not bad
  • 00:17:57
    if you're if you're an organization that
  • 00:17:59
    uh plans to be a going concern um it is
  • 00:18:01
    an investment in the future and I think
  • 00:18:03
    that's where we need to be clear it's
  • 00:18:05
    not an investment for short-term gain
  • 00:18:07
    it's an investment to give you a
  • 00:18:09
    strategic advantage right um um I think
  • 00:18:13
    uh some of the uh ups and downs we've
  • 00:18:15
    seen recently for example uh during the
  • 00:18:18
    for one of a better with the covid years
  • 00:18:20
    when uh companies suddenly started to
  • 00:18:23
    have this deep interest in automation as
  • 00:18:25
    uh labor became an issue um you saw the
  • 00:18:28
    cost of these things spike up any
  • 00:18:30
    automation equipment I think that will
  • 00:18:32
    probably normalize um and com and kind
  • 00:18:35
    of come back down to uh more normal
  • 00:18:37
    levels as we get as we've gotten out of
  • 00:18:39
    co now probably this is what year two as
  • 00:18:41
    we're going into um and so uh ROI should
  • 00:18:44
    come back uh to something that's more um
  • 00:18:49
    digestible for most companies, right? Um
  • 00:18:52
    but again, I would look at uh any
  • 00:18:54
    investment in AI or automation as as
  • 00:18:57
    really a strategic investment. This is
  • 00:18:58
    not something you're doing for quick
  • 00:18:59
    ROI. Yeah. Right. You're doing it to
  • 00:19:02
    position your company uh for future
  • 00:19:04
    success and you're doing it as part of
  • 00:19:06
    an overall strategy for your company,
  • 00:19:08
    not a one-off. And that's the other
  • 00:19:10
    thing. You need to do it as part of an
  • 00:19:11
    overall strategy, not of a oh this is
  • 00:19:14
    the flavor of the week, let's do it. you
  • 00:19:15
    know, it's got to fit into a longer term
  • 00:19:17
    strategy. Yeah, of course. And I think
  • 00:19:19
    that raises an interesting point as
  • 00:19:21
    well. Um, you know, being in
  • 00:19:24
    manufacturing, speaking
  • 00:19:27
    through tribal scale with different
  • 00:19:28
    manufacturers in North America. Um, a
  • 00:19:31
    lot of the questions that we're finding
  • 00:19:33
    is, you know, a lot of manufacturers
  • 00:19:35
    understand this need for digital
  • 00:19:36
    transformation. They see the value in
  • 00:19:38
    it. They know it's an overnight thing to
  • 00:19:40
    do. I think there's a lot of fear behind
  • 00:19:42
    that as well. You know there's a lot of
  • 00:19:44
    anxiety and I think a lot of it just
  • 00:19:46
    comes down to where do we start right
  • 00:19:49
    how do we identify really those instant
  • 00:19:52
    wins for us what you define instant as
  • 00:19:55
    uh over a time frame and we find that
  • 00:19:58
    many of them take like a phased approach
  • 00:19:59
    to that what kind of suggestions would
  • 00:20:02
    you make for these types of companies
  • 00:20:03
    who are first just trying to understand
  • 00:20:05
    what digital transformation is so I
  • 00:20:06
    think it's kind of a buzzword in the
  • 00:20:07
    manufacturing industry right now we're
  • 00:20:09
    finding even large organizations that
  • 00:20:11
    are still very paperbased
  • 00:20:13
    um don't really understand what it is or
  • 00:20:15
    how it works. And when you're learning
  • 00:20:17
    about that while trying to decide how to
  • 00:20:19
    act on it, I feel like that could
  • 00:20:20
    potentially be a lot for some
  • 00:20:22
    organizations while trying to obviously
  • 00:20:25
    make sure that they're staying optimal
  • 00:20:26
    in their day-to-day without having
  • 00:20:28
    resources being pulled away from these
  • 00:20:30
    things. What kind of suggestions would
  • 00:20:32
    you have for those types of
  • 00:20:33
    manufacturing companies for just how how
  • 00:20:35
    to how to get started and how to find
  • 00:20:37
    those kind of quick wins if they're
  • 00:20:39
    going from you know the old way to now
  • 00:20:42
    the current age I would suggest first uh
  • 00:20:45
    they have a look at do a bit of a
  • 00:20:48
    introspection on their operations uh
  • 00:20:51
    mostly because they know they
  • 00:20:53
    theoretically know their operations
  • 00:20:54
    better than anyone else try and
  • 00:20:56
    understand and whether you talk about
  • 00:20:58
    doing process flow mapping or whatever
  • 00:21:00
    uh where the inefficiencies are in the
  • 00:21:02
    operation. Um I think they're pretty
  • 00:21:05
    obvious ones. For example, if someone's
  • 00:21:07
    still using paper, uh they probably uh
  • 00:21:10
    they instinctively know, okay, I I
  • 00:21:12
    already know I'm not being the most
  • 00:21:13
    efficient from a in in today's world if
  • 00:21:17
    I'm still using paper, right? Um um and
  • 00:21:20
    once that introspection happens, I mean,
  • 00:21:22
    I think uh the first step uh is is
  • 00:21:26
    really how do I get rid of
  • 00:21:30
    any nonvalue added transactions which
  • 00:21:33
    would be paper. um how do I and another
  • 00:21:36
    another thing with smaller companies is
  • 00:21:38
    how do I get to my my one source of
  • 00:21:41
    truth or the information stream that I
  • 00:21:43
    want to use as my one source of truth
  • 00:21:45
    right and that's where again the joe's
  • 00:21:48
    come in the experience guys right uh you
  • 00:21:50
    use your team to determine hey guys this
  • 00:21:54
    shift uses this to do make decision the
  • 00:21:56
    night shift uses this uh let's decide on
  • 00:21:58
    what's my one source of fruit and and
  • 00:22:00
    that that whole process is already
  • 00:22:02
    getting him on the road to being more
  • 00:22:04
    efficient. Right. Yeah. Right. They're
  • 00:22:06
    it's removing uh uncertainty and it's
  • 00:22:09
    it's creating transparency. Right. Yes.
  • 00:22:11
    Right. Um and and so that's kind of the
  • 00:22:14
    f once that first step happens and so
  • 00:22:16
    it's and I'm probably oversimplifying
  • 00:22:19
    not every company has to go through this
  • 00:22:20
    but say they they decide okay these are
  • 00:22:22
    the flows are my one source of truth.
  • 00:22:24
    Everything else I should disregard or
  • 00:22:27
    degrade right or dep prioritize. Then
  • 00:22:31
    that's the that's the streams you you
  • 00:22:33
    you start uh digitizing and that's the
  • 00:22:35
    streams you start using to feed into
  • 00:22:38
    your digital dashboards or digital KPIs
  • 00:22:42
    and digital dashboards could mean you
  • 00:22:44
    know real-time performance of your line
  • 00:22:47
    right not waiting for the end of the
  • 00:22:48
    shift for somebody to tally up the
  • 00:22:50
    numbers on a piece of paper right um so
  • 00:22:52
    that then your supervisors on the line
  • 00:22:54
    can make can make tweaks as as the as as
  • 00:22:57
    a shift goes through by just looking at
  • 00:22:58
    that dashboard yeah right so that's
  • 00:23:00
    where the dig digitalization stage come
  • 00:23:02
    digitization stage comes in right um and
  • 00:23:06
    then once that's all set and you have
  • 00:23:08
    all this data digitized
  • 00:23:10
    um you then you have basically the
  • 00:23:12
    foundations of feeding your AI or if if
  • 00:23:15
    that's your next step right so now so
  • 00:23:18
    now you're basically using uh uh
  • 00:23:20
    digitization to help you predict and
  • 00:23:22
    make decisions right the next step with
  • 00:23:25
    with AI is now for help AI helping you
  • 00:23:27
    make those decisions using that data of
  • 00:23:29
    course right Um so I think I think
  • 00:23:32
    that's was the steps I would recommend
  • 00:23:34
    for them because I think by doing these
  • 00:23:36
    steps as well they start to see for
  • 00:23:38
    themselves um benefits um in in and the
  • 00:23:43
    and these early steps are also very low
  • 00:23:45
    cost generally and also you get the the
  • 00:23:47
    easy wins right you see the retrain
  • 00:23:49
    right so it helps reinforce that hey I'm
  • 00:23:51
    in the right direction um as well I
  • 00:23:54
    think what it does is it starts because
  • 00:23:57
    to what you mentioned there is a lot of
  • 00:23:59
    uh hesit meditation about making huge
  • 00:24:01
    changes. It eases them into the change
  • 00:24:04
    of the final change, right? It's in and
  • 00:24:06
    in a way it's it's change management.
  • 00:24:08
    It's helping them it's helping them get
  • 00:24:09
    used to the change. It's helping more
  • 00:24:11
    importantly the folks who would be using
  • 00:24:14
    the actual tools get used to the change
  • 00:24:16
    and not having it happen in one big
  • 00:24:18
    bang. Right. Yeah. Of course. Uh and
  • 00:24:21
    helping them. It also helps them and and
  • 00:24:24
    it it's good in a way because it helps
  • 00:24:26
    the folks who are using those tools
  • 00:24:27
    validate them from
  • 00:24:30
    from the start by determining what what
  • 00:24:32
    source of data is your one source of
  • 00:24:33
    truth all the way through to now as you
  • 00:24:36
    go full-fledged AI you have this team
  • 00:24:38
    validating the data and so there is also
  • 00:24:40
    confidence on their side that hey right
  • 00:24:43
    yeah so that when you start to train AI
  • 00:24:45
    they go yes for sure we'll use this
  • 00:24:46
    because you know we we cleaned it up
  • 00:24:49
    we've been using it now uh more from a
  • 00:24:52
    digital perspective now let's let's
  • 00:24:54
    start applying AI to yeah yeah obviously
  • 00:24:57
    you need to decisions, large
  • 00:25:00
    organizations, small organizations, you
  • 00:25:02
    know, data driven data driven decisions.
  • 00:25:05
    Correct. Um, and that's that's really
  • 00:25:07
    key. Um, and I think you've kind of
  • 00:25:09
    addressed brought up a little bit
  • 00:25:11
    timing, you know, and I think the
  • 00:25:13
    manufacturing industry as a whole,
  • 00:25:15
    especially as we've learned, um, really
  • 00:25:18
    seasonality plays a big key in the
  • 00:25:21
    operations of any manufacturing facility
  • 00:25:24
    throughout the year. Correct. Is there
  • 00:25:26
    ever a good time to start looking at
  • 00:25:28
    implementing these changes? And when
  • 00:25:30
    would that be? Oh, is uh I guess you're
  • 00:25:33
    talking about uh one of the I guess
  • 00:25:37
    barriers or hesitations you get from
  • 00:25:38
    folks is that uh uh this is my busy
  • 00:25:41
    period. Um no way I'm going to stop my
  • 00:25:44
    lines first to play with it.
  • 00:25:47
    I I
  • 00:25:48
    would that that's actually a tough one
  • 00:25:51
    because especially for a a smaller
  • 00:25:53
    company. Um they can they're less likely
  • 00:25:56
    to want to lose part of a year. Um I
  • 00:25:59
    would suggest that you start the work in
  • 00:26:02
    the off on the off seasons. Um and and
  • 00:26:06
    depending on the type the the the
  • 00:26:09
    project
  • 00:26:11
    um you would probably want to go live
  • 00:26:13
    before the the big season and so I would
  • 00:26:18
    say probably what you would do in their
  • 00:26:20
    busy season if anything is probably some
  • 00:26:22
    of the exploratory work and and and some
  • 00:26:25
    of the mapping right okay or or whatever
  • 00:26:27
    you need to do system and actually it's
  • 00:26:29
    a good time to do it because that's the
  • 00:26:30
    highest activity and that's where when
  • 00:26:33
    you're doing mapping
  • 00:26:34
    uh you would see every possible scenario
  • 00:26:36
    that could come up um or or if you're
  • 00:26:39
    training using it using it to train data
  • 00:26:42
    to train an AI uh it's also where you
  • 00:26:44
    could see every possible scenario and
  • 00:26:46
    then once you get into the off season
  • 00:26:48
    that's when you probably want to you
  • 00:26:50
    could probably risk taking down systems
  • 00:26:52
    or lines if you need to install uh
  • 00:26:54
    software uh or if there's a risk that uh
  • 00:26:57
    as you test software that there might be
  • 00:26:59
    downtime on lines I think then then
  • 00:27:02
    that's probably the time Um uh and then
  • 00:27:05
    uh obviously you could do
  • 00:27:07
    a depending on okay this depends on how
  • 00:27:10
    side how big your client is as well. I
  • 00:27:12
    mean if there are multiple lines you
  • 00:27:14
    could probably start on the smaller
  • 00:27:15
    lines and and kind of uh get your
  • 00:27:18
    learnings there first before loading out
  • 00:27:20
    to the the larger more more um more um
  • 00:27:24
    more strategic lines that they have
  • 00:27:25
    production lines when I say lines right.
  • 00:27:28
    Yeah. Yeah. So that's another option. Um
  • 00:27:31
    I would also suggest and I'm not sure if
  • 00:27:33
    this is possible in the type of uh
  • 00:27:35
    rollouts you do but uh if you have a
  • 00:27:38
    what we call a pre-prod or test
  • 00:27:40
    environment where you could test the
  • 00:27:42
    logic of what you're trying to implement
  • 00:27:44
    uh given a given a what do you call it a
  • 00:27:47
    a sample set of the parameters that it
  • 00:27:49
    will face and run it through at least on
  • 00:27:51
    the software side to see if uh things
  • 00:27:53
    are running smoothly before putting it
  • 00:27:55
    into production on the on the actual
  • 00:27:57
    line. Yeah, of course. Right. I I think
  • 00:28:00
    that's that's also important. Um but uh
  • 00:28:02
    yeah, you you raise a really I think
  • 00:28:04
    valid concern for any company uh that
  • 00:28:06
    has a seasonal business and uh to risk
  • 00:28:09
    uh um not being able to produce or sell
  • 00:28:12
    during their uh their season, right?
  • 00:28:14
    Yeah. Yeah. I know of course and I think
  • 00:28:16
    that's obviously those anxieties are
  • 00:28:18
    very valid. Um understanding how do I
  • 00:28:22
    implement this? Um that's one approach
  • 00:28:24
    even here at tribal scale that we're
  • 00:28:26
    taking is kind of the incremental
  • 00:28:28
    approach. Correct. um we're not
  • 00:28:30
    replacing so much, we're supporting, you
  • 00:28:32
    know, and I feel like that's really the
  • 00:28:34
    approach to doing anything digital. Um
  • 00:28:37
    and uh you know, as you're saying, using
  • 00:28:40
    that data to validate it just to make
  • 00:28:42
    sure that you're on the right path and
  • 00:28:43
    timing is critical for that for
  • 00:28:45
    especially manufacturers
  • 00:28:47
    um and that approach. Another side
  • 00:28:50
    question, but sure, how much is enough
  • 00:28:52
    data?
  • 00:28:55
    that will depend on your use and I think
  • 00:28:57
    and and in the organization. Um but I
  • 00:29:01
    also think how much data you would need
  • 00:29:04
    will also be uncovered when you take the
  • 00:29:06
    incremental approach because by then the
  • 00:29:09
    comp the your client would themselves
  • 00:29:11
    know how much I need to train my AI,
  • 00:29:14
    right? Um and another another actually
  • 00:29:17
    another advantage of the incremental
  • 00:29:19
    approach it reduces the potential of any
  • 00:29:21
    downtime, right? when you digitize the
  • 00:29:24
    risks there are probably lower than if
  • 00:29:25
    you were to go
  • 00:29:27
    full-blown AI right because uh if you
  • 00:29:30
    take the incremental approach by the
  • 00:29:31
    time you're ready to get to the most
  • 00:29:33
    sophisticated part of it you know your
  • 00:29:35
    foundation solid and really now is
  • 00:29:38
    you're now just handing over the
  • 00:29:39
    decision- making now to AI and you have
  • 00:29:42
    humans uh doing the double checking but
  • 00:29:45
    uh how much data that that's that's very
  • 00:29:47
    I would say that's very um that would
  • 00:29:49
    probably be very customer and use
  • 00:29:52
    specific Yeah. Yeah. Um and uh I would
  • 00:29:56
    also say that's also why it's
  • 00:29:59
    probably important for the company to go
  • 00:30:02
    through that stage where they
  • 00:30:04
    decide what what pieces of data are the
  • 00:30:08
    most relevant for them. Yeah. Yeah.
  • 00:30:11
    Yeah. Yeah. And that's something that uh
  • 00:30:14
    you can't decide but the company itself
  • 00:30:15
    needs needs to tell you that hey you
  • 00:30:18
    know this is what I need to make my
  • 00:30:19
    decisions right.
  • 00:30:22
    Yeah, of course. Yeah. Focusing on
  • 00:30:23
    bottlenecks, understanding Exactly. They
  • 00:30:26
    need to understand where where where
  • 00:30:28
    their critical pain points are and what
  • 00:30:30
    are their critical streams of data that
  • 00:30:32
    they must have or they need to make
  • 00:30:34
    their decisions. Yeah. Right. Yeah.
  • 00:30:36
    Yeah. Of course. No, I think that makes
  • 00:30:38
    a lot of sense. And something else you
  • 00:30:41
    kind of addressed is as well is is risk.
  • 00:30:43
    You know, I think part of our approach
  • 00:30:45
    here at Tribal Scales, we're always
  • 00:30:46
    trying to be very transparent. you know,
  • 00:30:48
    these are the risks of changing and
  • 00:30:50
    these are the risks of not changing.
  • 00:30:52
    Correct. And you know, you show that
  • 00:30:54
    with the benefits, everything is kind of
  • 00:30:55
    weighted decision. I think that that's
  • 00:30:57
    just kind of the appropriate way to to
  • 00:30:59
    take it and you know, that's that's what
  • 00:31:01
    we're seeing as well. Um, what are some
  • 00:31:03
    of the risks would you say for
  • 00:31:05
    manufacturers in conducting for the
  • 00:31:08
    first time any kind of digital
  • 00:31:10
    transformation focusing on AI for
  • 00:31:12
    example? And and what are the risks of
  • 00:31:14
    not changing would you say? I I would
  • 00:31:17
    say the the risks of changing is
  • 00:31:23
    uh there aren't I wouldn't call them
  • 00:31:26
    risks. I would say they're probably
  • 00:31:28
    there's some pain points of changing. Um
  • 00:31:31
    there is the change management aspect
  • 00:31:33
    with their employees. Um I think
  • 00:31:37
    um internally they would have to take a
  • 00:31:40
    good hard look of at the data and how
  • 00:31:42
    they're making decisions to determine
  • 00:31:44
    you know uh strategically what's the way
  • 00:31:46
    forward. Um I I I wouldn't say there are
  • 00:31:50
    necessarily risks. um the only the only
  • 00:31:53
    risk I would say probably is that they
  • 00:31:57
    in their strategy they choose the right
  • 00:31:59
    wrong use case for what they want to
  • 00:32:01
    digitize automate right and that's where
  • 00:32:04
    probably some work needs to be done and
  • 00:32:06
    where and where frankly their employees
  • 00:32:08
    and their
  • 00:32:09
    staff are critical obviously with with
  • 00:32:11
    your help as well in determining what's
  • 00:32:13
    the best use case in terms of chasing as
  • 00:32:16
    a as an init as an initiative
  • 00:32:19
    yeah I would say in terms of not doing
  • 00:32:21
    anything. I think uh that's the future.
  • 00:32:24
    Um uh unless you're very niche industry,
  • 00:32:29
    um you would probably be left behind if
  • 00:32:32
    if you don't at some point uh decide to
  • 00:32:35
    move in uh in that direction or embrace
  • 00:32:37
    AI. Yeah, that's interesting. Yeah. And
  • 00:32:41
    uh you know, I I think somebody in your
  • 00:32:43
    position, you've seen a lot of
  • 00:32:45
    technologies come and go, you know, a
  • 00:32:47
    lot of processes change. um you know
  • 00:32:50
    starting from a clipboard, pen and paper
  • 00:32:53
    you know all the way through full
  • 00:32:55
    modernization and cobots automation
  • 00:32:58
    things like that. Um what are some
  • 00:33:00
    technologies that you've seen that have
  • 00:33:02
    really kind of stood out and real kind
  • 00:33:04
    of game changers since you started until
  • 00:33:07
    now?
  • 00:33:09
    I would say actually the stuff that I'm
  • 00:33:11
    seeing as a real game changers is the
  • 00:33:13
    stuff I've I've most recently seen. I
  • 00:33:15
    mean, I think the stuff about digital uh
  • 00:33:19
    what they call digital factory twins or
  • 00:33:21
    it could be a digital warehouse twin. Um
  • 00:33:24
    I think that uh opens up uh a whole uh a
  • 00:33:27
    whole lot of opportunity for uh for
  • 00:33:30
    automation as well as uh using AI to to
  • 00:33:34
    optimize uh operations and it's already
  • 00:33:37
    being used if I'm not mistaken. that but
  • 00:33:39
    uh like where parts of factories are
  • 00:33:42
    digital twin. I'm not sure if there's
  • 00:33:43
    any factories or or or that are fully
  • 00:33:47
    digital twin but obviously it's it's a
  • 00:33:49
    new technology I think relatively new
  • 00:33:51
    and I think uh that that for me is is
  • 00:33:54
    most impressive because it also allows
  • 00:33:55
    you to train AI to train uh robots or
  • 00:33:59
    cobots um very easily versus the versus
  • 00:34:03
    uh how prior uh previous years we used
  • 00:34:05
    to train them. And that for me is a um
  • 00:34:08
    um terribly exciting. Obviously, when I
  • 00:34:10
    say uh digital twins, I also imply AI is
  • 00:34:14
    intertwined in that, right? Yeah. Yeah.
  • 00:34:16
    No, I think that's that's fantastic. Um
  • 00:34:19
    you know, and obviously seeing that
  • 00:34:21
    happen in person, that's that must be
  • 00:34:23
    pretty exciting as well. I I've seen it
  • 00:34:26
    experimented in person, but uh recently
  • 00:34:28
    I think it's uh it's pretty much now in
  • 00:34:30
    production in some factories. Yeah.
  • 00:34:33
    Yeah. Yeah. Fantastic. Talked about AI
  • 00:34:35
    for supply chain, optimizing roots,
  • 00:34:37
    things like that. Uh, even going from
  • 00:34:39
    paper to digital, you know, as you're
  • 00:34:41
    saying with with trucks and how they
  • 00:34:43
    operate. So, I think um, if you were to
  • 00:34:46
    kind of put time period for yourself
  • 00:34:48
    that you're curious to kind of see these
  • 00:34:50
    technologies grow, you know, you
  • 00:34:53
    mentioned three years, there's been a
  • 00:34:54
    lot of change in the the markets. What
  • 00:34:56
    do you think the next three years might
  • 00:34:57
    look like for manufacturing or maybe
  • 00:35:01
    supply chain? Is there anything that you
  • 00:35:03
    might be a little bit excited for at
  • 00:35:05
    this time or or looking forward to or
  • 00:35:07
    even your eyes? Oh, I would say on on
  • 00:35:11
    the manufacturing side, I am really
  • 00:35:13
    looking forward to what uh um automation
  • 00:35:16
    and combined with digital factory twins
  • 00:35:18
    will do. I think that that that would be
  • 00:35:20
    really interesting. Um if if we can
  • 00:35:23
    really uh turn that into something
  • 00:35:25
    really practical. Yeah, it certainly
  • 00:35:27
    looks good on paper and in the isolated
  • 00:35:29
    use cases that we see, but uh for if
  • 00:35:32
    that was to become widespread, it would
  • 00:35:33
    be really interesting uh um what that
  • 00:35:36
    would do. I mean uh actually and sorry
  • 00:35:39
    I'm just coming back to something you
  • 00:35:40
    said earlier on uh on uh on Canada and
  • 00:35:44
    the US uh I would say and this is
  • 00:35:48
    nothing to do with technology now is the
  • 00:35:51
    only other way Canada I would say could
  • 00:35:52
    get us
  • 00:35:54
    productively competitive with the US is
  • 00:35:56
    if
  • 00:35:58
    we actively started to target export
  • 00:36:01
    markets. So yes, we don't have the
  • 00:36:03
    population, but what if we would become
  • 00:36:05
    a manufacturer for to export products to
  • 00:36:08
    other companies that other countries,
  • 00:36:11
    right? And that's where we can make up
  • 00:36:13
    the volume. Yeah. And it doesn't
  • 00:36:15
    necessarily mean going south. It could
  • 00:36:17
    be going Europe, Asia, wherever. Yeah.
  • 00:36:19
    Right. And that's pretty interesting.
  • 00:36:21
    Yeah. And maybe that's an opportunity as
  • 00:36:23
    technology makes transportation easier.
  • 00:36:25
    Yes. Right. I don't know. Um but uh
  • 00:36:28
    Yeah. Yeah. Of course. And a lot of the
  • 00:36:31
    conversations, a lot of the information
  • 00:36:32
    we've been seeing coming out, especially
  • 00:36:34
    on AI and supply chain, I think a lot of
  • 00:36:36
    the activity even within Canada within
  • 00:36:39
    opportunities to enhance how supply
  • 00:36:41
    chains work. Um, it's always really
  • 00:36:43
    looked at, you know, really the trucking
  • 00:36:46
    industry. I think train as well to
  • 00:36:48
    extent, you know, going over land
  • 00:36:51
    masses. Um, I think it'd be incredible
  • 00:36:53
    to see how that could affect even
  • 00:36:55
    shipping, um, or, you know, potentially
  • 00:36:58
    flying. products internationally as
  • 00:37:01
    well, how that would optimize things as
  • 00:37:03
    well and uh perhaps a conversation for
  • 00:37:06
    later day as well because you can get
  • 00:37:09
    into some pretty deep rabbit holes about
  • 00:37:10
    what things could look like and what
  • 00:37:12
    they currently look like. There's
  • 00:37:14
    already talk of uh driverless trucks
  • 00:37:17
    going across, right?
  • 00:37:19
    Going across the continent, right? um
  • 00:37:22
    and driverless trains, which is probably
  • 00:37:24
    more of a uh a closer reality because
  • 00:37:28
    the trains are already on a track. Yes.
  • 00:37:30
    Right. They don't have to deal with all
  • 00:37:32
    the uh the uh the the non-planned
  • 00:37:35
    incidents a truck would entail on a
  • 00:37:37
    road, right? Yeah. Yeah. Right. Yes. And
  • 00:37:40
    and we've had conversations in the past,
  • 00:37:43
    you know, working in the food industry.
  • 00:37:45
    Um we understand there's different types
  • 00:37:47
    of trucks. You can have cooling trucks
  • 00:37:50
    and the cost of all that varies. So, you
  • 00:37:53
    know, perhaps maybe it's a little bit of
  • 00:37:55
    an obvious question, but are there
  • 00:37:57
    certain industries within manufacturing
  • 00:37:59
    that will see higher gains for things
  • 00:38:01
    like AI for supply chain optimization
  • 00:38:04
    knowing that there are different costs
  • 00:38:06
    operating in those markets?
  • 00:38:10
    Unfortunately, not really from what I
  • 00:38:11
    can tell.
  • 00:38:14
    Um because strangely enough uh the
  • 00:38:17
    frozen channel which is the most
  • 00:38:19
    expensive to distribute because of you
  • 00:38:21
    need to keep all the products at minus
  • 00:38:22
    25. Uh currently that's that's probably
  • 00:38:26
    the hardest to automate because uh as
  • 00:38:28
    you know once you when you throw lithium
  • 00:38:30
    batteries and uh and uh and and uh
  • 00:38:34
    certainly I um automated machinery
  • 00:38:36
    within a super cold environment um
  • 00:38:39
    things
  • 00:38:42
    a little tougher on the machines. Um,
  • 00:38:44
    for example, lithium batteries lose
  • 00:38:46
    their power quickly once you go below a
  • 00:38:48
    certain temperature. So, you lose some
  • 00:38:50
    of the efficiencies. Um, so
  • 00:38:53
    unfortunately, I would say I wouldn't
  • 00:38:55
    see a differentiation in the benefits
  • 00:38:56
    necessarily between industries. Um, it
  • 00:38:59
    will benefit all. So, in a sense, for
  • 00:39:01
    example, uh, it it will make
  • 00:39:03
    manufacturing deep frozen products more
  • 00:39:06
    efficient because you'll be able to make
  • 00:39:07
    decisions quicker. Uh however the same
  • 00:39:10
    will happen with if you're making
  • 00:39:11
    ambient products right
  • 00:39:14
    yeah that's pretty incredible um and I
  • 00:39:16
    think that one question or one approach
  • 00:39:19
    that always comes to mind again which is
  • 00:39:21
    always government related um is
  • 00:39:23
    sustainability you know do you think
  • 00:39:25
    that there's new
  • 00:39:26
    opportunities you know looking at
  • 00:39:28
    sustainability with these technologies
  • 00:39:30
    and if so what would that what would the
  • 00:39:32
    kind of use cases for that look like
  • 00:39:34
    early yes there would be opportunities
  • 00:39:37
    in sustainability Um from an AI
  • 00:39:40
    perspective, now suddenly you you can
  • 00:39:43
    have uh AI optimize your profitability
  • 00:39:47
    and one of the parameters that it
  • 00:39:48
    optimizes is carbon
  • 00:39:50
    foot right and that in terms and that is
  • 00:39:53
    that especially comes to uh say for
  • 00:39:56
    example planning out to distribute a
  • 00:39:58
    product what's the most uh cost least
  • 00:40:01
    costly as well as the the the approach
  • 00:40:03
    that has the lowest carbon footprint and
  • 00:40:05
    you don't have to have a human do it now
  • 00:40:06
    you can have an AI
  • 00:40:08
    countless iterations of how to do that.
  • 00:40:10
    I think that's that that there will
  • 00:40:12
    definitely be a benefit there. Um there
  • 00:40:14
    will also same thing with the
  • 00:40:15
    manufacturing if for example um uh an AI
  • 00:40:20
    makes a warehouse more efficient uh and
  • 00:40:23
    usually a warehouse is more efficient
  • 00:40:24
    there when there is less movement of
  • 00:40:26
    product within the warehouse. You
  • 00:40:27
    basically optimize the movement of
  • 00:40:29
    product around the warehouse then you'll
  • 00:40:31
    be using less electricity. Yeah.
  • 00:40:32
    Likewise a production you'll be using
  • 00:40:34
    and therefore less carbon right. Yeah.
  • 00:40:36
    uh even on the the AI side now if if
  • 00:40:42
    uh most recent developments uh continues
  • 00:40:45
    then theoretically we would be using
  • 00:40:48
    less power less powerful trips chips to
  • 00:40:50
    uh get the same AI oomph and therefore
  • 00:40:53
    the so-called uh data centers uh only
  • 00:40:56
    their own power stations as right now is
  • 00:40:58
    being forecasted right so theoretically
  • 00:41:00
    all that the the AI footprint could also
  • 00:41:03
    come down from what we're seeing right
  • 00:41:04
    now uh right common footprint, right?
  • 00:41:07
    Yeah. No, it's pretty incredible. I
  • 00:41:08
    think that opens up a lot of different
  • 00:41:10
    discussions and avenues to explore and
  • 00:41:13
    um you know, I think I think what we're
  • 00:41:15
    seeing is that is just the overwhelming
  • 00:41:17
    changes that are happening in these
  • 00:41:18
    industries. Um you know, we've had the
  • 00:41:20
    chance to talk quite a bit and adjust
  • 00:41:23
    our our focus on different topics, which
  • 00:41:25
    is uh pretty incredible. So, thank you
  • 00:41:26
    so much for that. But are there any kind
  • 00:41:28
    of main takeaways that you would kind of
  • 00:41:31
    leave viewers with who are considering
  • 00:41:33
    digital transformation or who have maybe
  • 00:41:35
    already started it and who are just
  • 00:41:37
    starting to understand what it is and uh
  • 00:41:39
    really what they should do next? Well, I
  • 00:41:42
    I would say like AI is is here to stay.
  • 00:41:44
    Um I think it's going to be our future.
  • 00:41:47
    I think uh organizations need to
  • 00:41:52
    understand where they would play or best
  • 00:41:55
    fit into this or how they can best
  • 00:41:57
    leverage this is probably another word
  • 00:41:59
    um to their advantage and what
  • 00:42:01
    strategies they want or how they want to
  • 00:42:03
    use that in their AI in their
  • 00:42:04
    strategies. I think uh
  • 00:42:07
    um although it does not necessarily mean
  • 00:42:10
    they need to jump in uh all the way into
  • 00:42:13
    the deep end in one shot, I think
  • 00:42:15
    certainly they need to start looking at
  • 00:42:19
    how they make decisions and their data
  • 00:42:21
    flows and decide what information for
  • 00:42:23
    them is critical, what's key and start
  • 00:42:26
    moving towards at least if it's not
  • 00:42:28
    already there on onto a digital sort of
  • 00:42:31
    platform and start using it uh digitally
  • 00:42:35
    because uh once you start doing that you
  • 00:42:37
    start recording it automatically and how
  • 00:42:39
    you make decisions right it's it start
  • 00:42:41
    that starts being recorded automatically
  • 00:42:43
    so that when you do the transition to AI
  • 00:42:45
    you already have your database on which
  • 00:42:47
    to train your AI. Yeah. Right. That that
  • 00:42:50
    that would be my advice. I don't I don't
  • 00:42:52
    think there's any turning back. I think
  • 00:42:54
    uh AI using AI or leveraging AI will
  • 00:42:57
    become cheaper if what we're seeing so
  • 00:42:58
    far in the last couple weeks continues.
  • 00:43:00
    Um it will become more competitive as
  • 00:43:02
    well. So again with comp competitive
  • 00:43:05
    comes more cost. If it becomes more open
  • 00:43:07
    source it also means it'll probably
  • 00:43:09
    develop or become more sophisticated
  • 00:43:12
    quicker. Um um and as well as uh
  • 00:43:16
    programmers will start developing
  • 00:43:19
    various applications for it a lot
  • 00:43:20
    quicker too right so I think all those
  • 00:43:23
    are good things and uh I think it's it's
  • 00:43:25
    now for individual companies to decide
  • 00:43:28
    okay where do I how do I leverage this
  • 00:43:30
    and uh how can I best leverage this
  • 00:43:32
    given my business strategy and my
  • 00:43:36
    industry I don't I don't think there's
  • 00:43:38
    any turning back yeah that's fantastic
  • 00:43:41
    thank you so much Ron's been a huge
  • 00:43:43
    asset I in the Canadian
  • 00:43:46
    manufacturing across the board. So,
  • 00:43:48
    thank you so much. Thanks for having me
  • 00:43:50
    for contributing and uh contributing to
  • 00:43:52
    this talk. Uh I'm sure we're gonna have
  • 00:43:53
    lots of questions come in later. Um
  • 00:43:55
    we'll be happy to address those with you
  • 00:43:57
    maybe at some point. Always really
  • 00:43:59
    appreciate your insights and it's been a
  • 00:44:01
    real pleasure. Thank you, Jeremy, for
  • 00:44:03
    having me. I mean, it's was great having
  • 00:44:05
    this conversation. It's actually uh
  • 00:44:06
    given me a chance to also contemplate uh
  • 00:44:09
    some of the things I kind of have
  • 00:44:10
    sitting around in my head. So yeah, so
  • 00:44:12
    much for having
タグ
  • manufacturing
  • digital transformation
  • AI
  • automation
  • data management
  • sustainability
  • change management
  • digital twins
  • warehousing
  • efficiency