Future Product Lifecycle and Digital Networks

Future Product Lifecycle and Digital Networks


Integration is an important element of manufacturing. Time ago, the meaning of integration was to build a manufacturing factory which created virtually everything needed for manufacturing at the same place. Rouge River Plant by Ford Motors was one of them. Finished in 1928 it became the largest integrated factory in the world. Henry Ford’s dream was to organize a factory combining precision manufacturing with standardized and interchangeable parts and a clear division of labor. Here is quote from Robinson Library article.

The largest industrial complex in the world at the time, the complex would eventually include a steel mill, glass factory, and automobile assembly line. Virtually everything required to build and assemble a Model T was manufactured on the premises. The final assembly line was transferred from Highland Park in 1927. In 1930, the Rogue River Plant covered 6,952,484 square feet and employed 81,000 men.

Fast forward almost 100 years. Manufacturing is different. Integrated plants are history now. Global manufacturing and supply chain is a reality of today’s world. In my presentation at COFES 2015 earlier this year, I’ve been discussing data networks as a new way to support a complexity of manufacturing environment. The picture below shows suppliers of new BMW 3 series.


Each product is a system.The growing complexity of manufacturing environment and products creates the demand for new product lifecycle architectures. These architectures will be able to support management of multidisciplinary product data (mechanical, electronic, software) and will operate as a global distributed data network.

I’ve been thinking about a role of networks in the future of manufacturing and product lifecycle. I’ve been reading Edge article – Digital Reality A Conversation with Neil Gershenfeld [1.23.15]. It is a bit long, but I would recommend it to anybody working on engineering and manufacturing technologies these days. Neil Gershenfeld, Director of MIT Center of Bits and Atoms gives you a perspective of future rethinking of manufacturing.

Engineering and manufacturing processes will be digitally integrated, but physically distributed.In this digital future, we will be able to ship data to make manufacturing local. Here is the passage which explains that:

To rewind now, you can send something to Shenzhen and mass manufacture it. There’s a more interesting thing you can do, which is you go to market by shipping data and you produce it on demand locally, and so you produce it all around the world. There’s a parallel with HP and inkjet printing. HP’s inkjet division is in Corvallis, Oregon because they had to hide from Palo Alto because they were told that inkjet printing would never scale, it would never be fast enough. But their point was a lot of printers producing beautiful pages slowly scales if all the pages are different. In the same sense it scales to fabricate globally by doing it locally, not by shipping the products but shipping the data.

But this is not a final stop if apply a complete product lifecycle paradigm to that. In the world of Ford-T product left the factory and never came back. We have some elements of recycling today in manufacturing. Here is a glimpse of the future:

If you take this alignment between mainframes, minicomputers, hobbyist computers, PCs, the research tools we’re using are like the mainframes, the fab labs are the minicomputers. They’re being used to do the equivalent of invent the Internet. The next step is we’re doing a lot of work on machines that make machines. You don’t go to a fab lab to get access to the machine; you go to the fab lab to make the machine. To do that we’ve had to rip up CAD-CAM, machine control, motion control, all the ways you make stuff, to make machines that make machines. That’s the next step. Over the next maybe five years we’ll be transitioning from buying machines to using machines to make machines. Self-reproducing machines. But they still have consumables like the motors, and they still cut or squirt. Then the interesting transition comes when we go from cutting or printing to assembling and disassembling, to moving to discretely assembled materials. And that’s when you do tabletop chip fab or make airplanes.

What is my conclusion? Majority of manufacturing today is point-to-point. We can define things, design and engineer them, plan the production and make things at the end. There are lot of hype of internet of things today. To me the most important element of IoT and PLM convergence is how we connect them together into digital network in the way that each element virtual or digital will be able to define the network and empower it. Similar to internet, it will grow into a network that won’t require a central control – a departure from point-to-point manufacturing in the future of product lifecycle. Just my thoughts…

Best, Oleg


Share This Post

  • I do wonder about these assertions around ‘it scales to fabricate globally by doing it
    locally, not by shipping the products but shipping the data.’ Sometimes it
    seems to assume that material will ‘just be there’ in the right place and of
    the right type.

    At the very least there are nuances which might stem from performing some kind of ‘Material flow’ analysis- comparing the centralized design/production model to the (idealised) centralised design/distributed fabrication(/assembly) one. My thoughts are in flux – but revolve around considering the transport of raw (often bulk) materials. Not mentioned – but energy delivery is another analysis to lay over the top.

    Let’s assume that materials are ‘semi processed’ – so not mainly made up of ‘ore’ – so ‘waste’ – may be considered small for metals and some plastics and vanishing to 0 for ‘ideal’ 3DP. We can assume bulk transport is generally by sea. I think some questions
    help us think a bit more deeply about this:

    So how will the(multiple) materials be distributed? – probably from a spoke and hub – wheel model– and also recycling /waste recovery will need to be performed also. – options locally or centrally (via the return trip?). Digital actually goes both ways – local requirements can be centralised back – and combined with the efficient and (relatively)
    cheap transport costs. Locally tailored design can be centralised (or maybe at the local HUB (by local I mean country or state)).

    My thoughts here are that a spoke and hub model (with manufacturing at or near the hubs)actually makes more sense – than what many people mean by local. Especially as there is some benefit to large industrial processes (incorporating some form of mass customisation) in automated assembly. Think Amazons distribution hubs here)

    My point is that eventually a balance similar to the (efficient) logistics model may emerge – rather than the ‘true’ local one – one that is often driven by some people’s desire to ‘want’ the local over the central in a quasi political statement [not you Oleg] – to supposedly save energy and resources and maybe tilt back to idealised ‘craft’ production and allow them to take a break from the ‘modern world’ .
    Of course all the above needs better modelling – sounds like an idea for a student project.

  • beyondplm

    Paul, thanks for your comment! I think you touched an important point – materials and distribution. This is clearly something to think about. I can only imagine that the same network, which is taking care about optimal local manufacturing can count the optimal material distribution. Something similar to the model with airline hubs and transportation services.

  • Thanks Oleg Hub and Spoke is what I meant by airline, logistics- that’s why to me it goes way beyond IoT and PLM convergence – and well into E(M)RP. Infact it goes beyond ERP into SRP (‘State’ resource planning??) – well at least beyond the individual enterprise. What ever happens I think this is beyond the existing boundaries of what we refer to as PLM, ERP etc Logistical planning – and will need more than a few get ‘rich quick’ startups to get going.

    There is a role for a new Digital Infrastructure (not just routers and fibres) to be laid down – before what we now call Digital manufacturing actually can achieve its ‘True’ promise. I think that maybe Logistical ‘material’ movement – may actually be a better place to start – than ‘traditional’ PLM – and may be a better guide to what gets produced where and How (eg Trad subtractive/AM) than just some idealistic 3DPers wanting it to be produced locally based on the idea that local is more efficient/ appropriate. Build the new PLM on top of the LLM (Logistical Lifecycle Management??)

    A free (but appropriate) link from my employers (different brand) http://www.quintiq.com/

  • beyondplm

    Paul, you are right, it goes beyond traditional PLM which is in the best case point-to-point delivery from design to manufacturing. I’m sure we are going to see multiple attempts to start from different places – PLM, Logistic, Supply Chain, materials, manufacturing services and more.

    I wonder what do you mean by “new digital infrastructure”. I thought we have one – internet. Do you see another infrastructure for manufacturing?

  • Oleg, by NDI
    ( I should have said New Digital Manufacturing Infrastructure) – I mean what could be built up on top of the Internet (as I said router and fibres)and WWW – and even above that with xml, json, HTML, Javascript and general IT technologies that are used by all industry sectors outside (as well as inside manufacturing – eg insurance, finance. Travel and holiday bookings (ie words and numbers).

    I think when we start talking about ‘things’ or services (such as CAD/CAE) which are directed at making things, and involve the logistics (and procurement maybe?) of large amounts of Stuff) – the current providers of Digital manufacturing ‘infrastructure’
    are actually behind the curve – but with a largely captive market. Vendor tie in and (wilfully?) poor levels of interoperability – should have been sorted out (via the market of via regulation) years ago – but there’s been no clear winners and unlike the Victorians who say imposed (an imperfect) standard(ish) rail gauge which helped to expand and unify the industry– govts are too afraid to intervene even in cases where it may make some sense.

    So I’m saying let’s take the at as a start analysis (via say rfid) of material flows – and then as a next step the effective routing of material flows as a county or societial level (tied in with digital design flows as you original article was discussing) – as the start of a New Digital Manufacturing Infrastructure- back bone – which businesses can use (as a kind of broker initially I guess?) – and society can deliver ‘stuff’ in a more optimal way.
    Any existing Digital manufacturing vendors – can then join (or contribute to) that digital infrastructure – but one that has a combined material and design flow. Things such as STEP, semantic web/RDF etc have all be false starts – but now maybe with things such as NoSQL databases and with the (often ) higher Product complexity of manufactured goods – and the knowledge of how to produce s/w to produce them- it’s time for traditional Digital Manufacturing companies to produce another layer (of Product Definition models for one) on top of the WWW stack (and it is technically possible) and if they don’t then for people outside the traditional Digital manufacturing world to
    shake things up. Aligning the Digital and the Real world Infrastructure architectures appeals too – although that is just instinct.

    I’m just saying it maybe worth taking material logistics as a possible- non obvious starting point for that as a way to break out of ‘traditional’ Digital Manufacturing models (eg Cloud) – which tend to play out as a repeated history of how the WWW developed (PC/Browser) on every desk, open source etc, open everything) . but enough for now – I will mull…

  • beyondplm

    Paul, thanks for sharing your insight! This is great conversation!

    I agree with your view about building future manufacturing infrastructure on top of existing www standards- totally make sense.

    There is one thing, which is unclear to me. You are saying

    “the start of a New Digital Manufacturing Infrastructure- back bone – which businesses can use (as a kind of broker initially I guess?) – and society can deliver ‘stuff’ in a more optimal way. Any existing Digital manufacturing vendors – can then join (or contribute to) that digital infrastructure – but one that has a combined material and design flow.”

    So, who will build an infrastructure and why others (especially existing PLM vendors) will join? And what is the meaning of such ‘join’?

    One of the www advantages is that no central system is required to keep it up and running. Do you assume some sort of “central” digital manufacturing infrastructure?

    Another thing that is not clear to me is your statement about NoSQL. I guess, it is better that what we had before, but why this layer will be better than RDBMS, RDF, STEP, etc.? For example Eurostep is providing solution on top of STEP. I’m not saying this is a platform for future manufacturing, but how NoSQL solution will be different from existing ones?

    PS. It looks like we put enough comments to grab beer or coffee :). Are you visiting DS office in Waltham, MA? I live just few miles down in Newton.

  • HI Oleg, sorry for the late reply. The question you ask is
    difficult to answer. “Who will build the infrastructure”. There are several ‘potentials’.

    The STEP/ISO community is an obvious one – and as you
    mentioned RDF has been used as a possible back bone – but has the issues of being seen as ‘too complex’ – plus STEP has problems (I recall) over the mapping between different vendor data representations- especially when each vendors
    wants to Add value – through its own technologies and geometric (and non geometric) representations. I also recall that RDF and associated technologies (esp SPARQL) did not have write support (to DB). Neo4J for example does (via CYPHER) in a similar way to SQL.

    RDBMS will not go away (and can talk to Graph DBS anyway) but I see that with Graph representations being ‘natural’ for both geometric and non geometric Product/engineering data – then having a graph representation ‘All the way down’ – UI- bunessness/modelling layer->data layer – could both
    reduce the amount of code involved (increasing reliability) and I understand potentially allow for better scalability in the cloud. Research would need doing here of course. When I see a feature tree, a Boolean op, as assembly tree
    or an update ‘Graph’ – they are all implicit or implemented in different forms where the ‘back bone’ is really a Graph – with operations/ data at each node.

    This leads to the alternative ‘Who’:

    The other alternative is some form of start up (or set of
    start ups) – who without having the
    legacy (and overheads) of the existing vendors) are more interested in providing a Service than selling their ‘product’ (in this case CAD/CAM/CAE/PLM[ I will use [CAE] for short]). So this goes back to your point of ‘starting from (non CAE ) data Logistics, materials etc’ – to I guess ‘Set the example’ with simpler data structures than CAE – but I feel with many of the ways that Google, twitter etc have being doing things –
    possibly involving ‘Graph Thinking’ the CAE industry could learn a lot in forming the Infrastructure back bone.

    Traditional vendors will resist of course – but as a partial
    analogy even Microsoft and partially embraced open source to be involved in the general software infrastructure – but are concentrating more on their core ‘value
    add’ to ‘Join’ the other (two) infrastructures Android and iOS – and also make money on hosting via Azure. Perhaps this is the part the traditional vendors can play a part – by hosting the data and serving other (smaller) players ‘value add’ apps on a common (ish) platform.

    Amazon is a good ‘third way’ possibility/ model for what
    might happen – in that they built up their AWS platform for internal use (not their core business) but made sure that it could act as a service for others.

    Ironically Microsoft could be the ones who developed or coordinated the build up of a new ‘Manufacturing infrastructure’- again going back the the AD man – materials/ logistics discussion which started this – especially as the CAE levels of sophistication in terms of algorithms are relatively or often ‘low level’.

    Enough – too long again and too confusing. Unfortunately
    I am in Cambridge UK – so a coffee will have to wait for a while. thanks for reading.

  • beyondplm

    Hi Paul, thanks for sharing your comments! Many valid points. I will keep in mind you’re in Cambridge for my next travels to UK. Would love to get together and discuss. Best, Oleg

  • Pingback: Beyond PLM (Product Lifecycle Management) Blog » PLM and Disintegration Challenge in 2016()

  • Pingback: Beyond PLM (Product Lifecycle Management) Blog Data and Connected Enterprise (inspired by PDT 2017 call for abstracts) - Beyond PLM (Product Lifecycle Management) Blog()