Resolved
1 votes
As Sandy Kemsley states in this blog:
We need to turn the internet of things into the process of everything. A sea of events needs to feed into a sense/respond engine that drives towards outcomes, whether a simple status outcome, a repair request, or automation and control. BPM, or at least the broad definition of intelligent BPM that includes decisions and analytics, is the perfect match for that sense and respond capability.
What do you think?
Thursday, June 12 2014, 09:46 AM
Share this post:
Responses (7)
  • Accepted Answer

    Thursday, June 12 2014, 10:41 AM - #Permalink
    Resolved
    5 votes
    Absolutely. There is no doubt that the Internet of Things represents a great moment for BPM. Gartner estimates that by 2020 the number of smartphones tablets and PCs in use will reach about 7.3 billion units while the IoT will have grown much faster resulting in 26 billion units at that time. More connected devices mean more intelligent systems, and these intelligent systems need to be controlled by an intelligent engine/designer combo. BPMS companies are well positioned at the eye of this storm. Whether you are embedding devices or installing systems, someone or some software needs to be adding some intelligence to the IoT systems. This is nothing different than what BPM does today...call it BPM on Steroids because there will simply be a massive increase in endpoint devices. Not only will business boom for BPMS companies during the next few years, I suspect we will see a lot more M&A as companies start rewriting their business plans to focus on IoT and realize that they need to add a designer and an engine to their product strategy. The interest thing is that many companies are rushing headfirst into creating engines and designers to create these ecosystems. But most of the designers at these companies don't have BPM backgrounds so they tend to oversimplify and underestimate the enormous value that lesser understood BPMN 2.0 elements bring to the table. Take for example the event-based gateway. I don't run into too many people on the street that know what this gateway does. But ask anyone at a BPM conference about it, and their face lights up as they describe its importance. After all, how do you enforce an SLA without an event-based gateway? In other words, your NEST reports a smoke. If the source of the smoke is not detected/eliminated within 5 minutes then you want to dial 5 cellphones, call the fire department, and open the garage door. Or, if in 4 minutes you get confirmation that it was a false alarm, then do nothing and let everyone in the house continue sleeping. How do you do this without an SLA enforcing BPMN element? Every assigned action has an implicit SLA whether we admit to it or not, and in most scenarios I need to tell the rest of my connected ecosystem what to do if that SLA is not met. The event-based gateway is the most elegant way of doing this, and I have never seen a non BPM centric designer handle this type of scenario in such an efficient way. One bit of bad news, though, though for most BPMS providers – you are going to have to rewrite a lot of code. 30 billion devices is a lot more than 10 thousand users. This means that only the most scalable, fastest, and resilient technologies will prosper in this brave new world. .
    Like
    1
    • Max J. Pucher
      more than a month ago
      Brian, You are absolutely right there is no BPM solution on the market today that can handle the event stream of many devices ... actually BPM can at best handle ONE EVENT at a time. Thus I have no idea what you are talking about.

      BPM was invented from manufacturing, so lets see what is happening there in robotics. Many robots are no longer programmed by someone to perform a series of movements. Some are trained by users moving their heads and some find the optimal path themselves from target definitions. Some have a multitude of actor arms and they coordinate their movement and sequence through highly adaptive algorithms, most certainly not flow diagrams. BPM is so outdated that being stuck on it is kind of getting hilarious.

      BPM proponents cannot imagine more than a flowchart, so thats all they see ...
    • Scott Francis
      more than a month ago
      how's the check-scanning and OCR stuff working out?
      i mean honestly. Saying BPM is out of date is like saying Word Processing is out of date. specific tools will go out of vogue or be supplanted by better tools, but people will still want to type out (or dictate) their thoughts into a more lasting form. Similarly they'll want to capture how work gets done. Today call that BPM. maybe we'll call it something different in the future.
    The reply is currently minimized Show
  • Accepted Answer

    Thursday, June 12 2014, 11:01 AM - #Permalink
    Resolved
    6 votes
    Absolutely agree. There's no point generating all of this data from smart devices unless it's put to work. By taking the captured data and applying it directly to processes we can significantly enhance the value of IOT devices. An interesting question (for me anyway) is what will the role of BPM platform be in this new new era of smart IOT solutions? I expect IoT devices to be supported by BPM platforms with BPaaS (Business process as a service) capabilities i.e. cloud enabled and multitenant with huge scalability (as stated previously by Brian). In effect the IoT smart process market will be like the current BPO market on steroids.
    Like
    1
    • Max J. Pucher
      more than a month ago
      All that BPM produces is fools with tools. Can someone tell me where the 'intelligence' in BPM is supposd to be coming from that will handle these devices? A godly process designer and each time new devices are added processes are being redesigned from a holy vision? BPM can't even easily deal with supporting new data on the fly so the data from these devices is going to be a nightmare for a process programmer.

      The natural complexity of an 'internet of things' requires an adaptive solution that supports complex systems. BPM is just a complicated mesh of flow diagram straightjackets. It has no handle on data, content and rules which will all be key to deal with the dynamics.

      The future will be machine-learning solutions, much like the NEST thermostat that Brian referenced. A management solution will need to learn what devices exist in a certain context. Learn which data elements and their values are relevant from user interaction. Then the actions relevant to achieve certain goals will be trained by users. An chain of events-contexts-actions will build goal-achieving patterns over time.

      Flow-diagrams? Intelligent BPM with predictive analytics and process mining? Dream on ...
    • Peter Whibley
      more than a month ago
      Max, you are thinking of IoT process solutions from your perspective of human centric, unpredictable, dynamic business processes. For the most part IoT will be invisible to users. I already spend too much time on the web so I don’t need another 26Bn smart devices competing for my attention. The initial wave of IoT solutions will be simple processes that eliminate important yet mundane processes e.g. inventory management, scheduling service and repair, health monitoring. For these type of processes BPM is ideally suited.
    The reply is currently minimized Show
  • Accepted Answer

    Thursday, June 12 2014, 11:55 AM - #Permalink
    Resolved
    2 votes
    For many of us, the internet of things is already here. The people in the waterfall - Analysts, architects and developers - along with the manager whose ego project it usually is, treat staff as mere switches in the process – an inconvenience which interrupts their fully automated process. To them people are just things - nuisances which stop their process working as smoothly as it might. But it these three guys who will make sure BPM is sidelined again. The whole development process is too expensive, too cumbersome and often both user and customer unfriendly. One lesson from mobile phones and tablets is that ease of use is the new competitive advantage. Apps which make connecting data as easy as finding something in a list and selecting it. Swipes, pretty pictures, zooms and fades and all for free – that’s how the web is educating customers and users. You can’t take embedded processors costing pennies and charge £50k just to create a process to connect them. Here there is a race between opposing methods and groups of people. I’ve recently been drafted in to manage the Meetup for Thames Valley Web Developers and I can tell you that these guys are good at creating process-driven, customer friendly workflows. They are already feasting on process improvement people’s lunches. There is a massive opportunity to link the sensors into the company, to deliver instant big data and a process which learns, adapts and re-iterates itself – a truly artificially intelligent process to provide “what the customer needs before he knows he needs it”. But, if he doesn’t know he needs it, how can he be persuaded to pay lots of money for someone to design a process from scratch to make it happen? And who is going to come round his house to install the servers, take the licence fees and set up the 20% maintenance contract???! BPM needs to learn to design once and deploy often, without charging for re-inventing the wheel each and every time. Banks may fall for that – thing designers won’t.
    The reply is currently minimized Show
  • Accepted Answer

    Thursday, June 12 2014, 12:39 PM - #Permalink
    Resolved
    2 votes
    Sandy Kemsley's BPM-centric quote on the potential of the Internet of Things is terrific. "A sea of events"! Marvellous. "Sense and respond" driving towards "outcomes"... it's an ideal vision, grounded in capabilities that are available today, and meeting business requirements that are real. What's the catch? I'm not sure there is a "catch", except to note that a lot of specialized work needs to be done. And Sandy alludes to that work in the reference to "intelligence", "decisions" and "analytics". This is the work particularly of the business analyst. (Business analysts unite! Your ship has just come in . . .). Why the business analyst? Because the BA "profession" is all about exactly the kind of modeling work that needs to be done. IoT adoption will only accelerate insofar as all the requisite business intelligence is developed. That's the choke point. I suspect competitive pressure will eventually be the engine pushing adoption forward.
    • Max J. Pucher
      more than a month ago
      John, Sandy and others are reiterating things I have said for the last ten years. They have nothing to do with todays BPM. It is a smoke screen behind which the limitations of todays BPM methodology are hidden. In what way does typical BPM today deal with goal and outcomes? In ACM we use goals but they are worked towards by humans not by flow-diagrams, not by business or artificial intelligence that needs expert programming to work. We are the only ones to have a working pattern recognition engine that makes suggestions to performers.

      An internet-of-things will not be controlled and managed by business analysts and yes, while some will want to mine that information to do analytics, I can guarantee you that most communication that is valuable will be protected from such mining. Secure protocols that allows devices to instigate communication privately and without others spying. My NEST, fridge or any other device in my home or car will not post private information on the internet. I expect my devices to authenticate themselves to communicate each other information that they can interpret and utilize cleverly without being open to spying.

      At the University of Vienna research is in progress to define and test brokerage protocols that allow devices to freely collaborate without any upfront design work. They do not utilize flow diagrams but algorithmic approaches that are closer to game theory. It will be those brokerage hubs that will organize devices to collaborate.

      So I find the suggestion that BPM will be the control of an IoT unimaginative to say the least.
    • Scott Francis
      more than a month ago
      we're all just regurgitating max's ideas and the sooner we get used to that the better :)
    The reply is currently minimized Show
  • Accepted Answer

    Thursday, June 12 2014, 01:35 PM - #Permalink
    Resolved
    3 votes
    Max You are so wrong. First “BPM” was not invented from manufacturing. “ERP” came from manufacturing which created a gap in business between people and hardwired records of history that make an accounting system or back office “processing”. And so “BPM” as a discipline filled that gap and is now set to help sort out the mess of legacy and go into new arenas such as the internet of everything. BPMS need to handle front and back office seamlessly which of course will readily handle “the internet of everything” including digitisation of business operations. However not all internet of everything will be a process? Sure early iterations of BPM supporting technology may well fit your outdated descriptions but not now. We and I expect others can handle multiple concurrent and asynchronous events. We certainly have all the adaptive capabilities ready to build any process including case management. Can’t see supporting new data on the fly is an issue! We already can build intelligent process that can recognise past actions and there will be more to come. BPM as a principle has no limitations in that “gap”. Some of the supporting BPMS may yet have some way to go to cover all the requirements and as Brian suggests some vendors may have a technology “challenge”? But please do not impose limitations that have been addressed by some even if not all?
    • Max J. Pucher
      more than a month ago
      David, then I suggest you read Michael Hammer ...

      Software on the Internet will be very different in ten years time and will not resemble anything that the limited perspective of enterprise architects and process designers can imagine. It most certainly will have to do nothing with anything anyone might chose to name BPM today.

      Applications and software in enterprise applications are slow, tedious and outdated and that is the only reason that anyone could be excited about today typical BPM approach ... or 'methodology' if you will.

      You remind me of Steve Ballmer, who laughed about the iPhone claiming that it would fail because their market research said so ...
    • David Chassels
      more than a month ago
      Max
      Yes Hammer and Champy with BPR in the early 90s tried to bring process to the fore but it coincided with IT dominating the business systems that promised much but as we now know failed. And so that dream of people and process optimisation faded into the shadow of “IT”.

      I liked Peter Keen in his Process Edge thinking highlighting the need to look at salient processes and recognise non salient ones which could be standardised even outsourced. He recognised that good processes were assets but if failed to adapt with changing circumstances they could be come a liability. Again IT failed to address this adaptive requirement and this is the “gap” that “BPM fills as I describe. Now the supporting BPM software can deliver this quickly and efficiently.

      There is nothing new in “process” it is how business really works! As I have said before in the early 70s I was mapping out processes walking round whisky companies understanding how their processes worked looking for weaknesses …and I found them…..!

      No matter what fancy delivery mechanisms the future holds business logic has not and never will change! It is about people doing things to create an outcome. Yes you are right “applications and software in enterprise applications are slow, tedious and outdate” and represents the big opportunity for BPM led new adaptive applications at a fraction of the cost of old “IT”. But it will be evolution not revolution as it will take years to manage out legacy and silo duplication.

      I think I and many co contributors are the ones with vision of a new way and you take the Steve Ballmer stance….time will tell.....
    The reply is currently minimized Show
  • Accepted Answer

    Thursday, June 12 2014, 02:20 PM - #Permalink
    Resolved
    2 votes
    “process of everything” sounds very familiar for me – “person’s life-as-a-process”, “person's situation-as-a-process” - see http://improving-bpm-systems.blogspot.ch/2013/06/practical-process-patterns-cxaap.html Sure, any serious endeavor is a system of processes, but “sense and respond” sounds very reactive for me. Maybe this is good for things, but not for people (at least for some of them who I know). Again, I believe that BPM (as a trio of discipline, architecture/practice and tools) may help (although Max may disagree) us, the people, leaving and working beyond "sense and respond" . Thanks, AS
    • Max J. Pucher
      more than a month ago
      Yes, I do disagree. Any real world system is a complex network of entities that communicate through states and events that propagate and react freely. Show me a natural system that follows a flow diagram.

      An internet-of-things will either resemble such natural adaptiveness or it will simple be utterly useless.
    The reply is currently minimized Show
  • Accepted Answer

    Friday, June 13 2014, 03:23 PM - #Permalink
    Resolved
    2 votes
    Max, please, do not assume that we not talking ONLY natural systems (at least, I not consider people on this planet as a "natural" system) and we are not talking of flow diagrams as ONLY the coordination technique. Again, we, architect, do not assume. Thanks, AS
    The reply is currently minimized Show
Your Reply

Join the Discussion

Want to join the discussion?

Login, or create an account to participate in the forum.

Top Participants

Dr Alexander Samarin
278 Replies
01/10/2016
David Chassels
271 Replies
30/09/2016
Emiel Kelly
223 Replies
30/09/2016
Bogdan Nafornita
210 Replies
30/09/2016
E Scott Menter
182 Replies
28/09/2016