We use cookies and other similar technologies (Cookies) to enhance your experience and to provide you with relevant content and ads. By using our website, you are agreeing to the use of Cookies. You can change your settings at any time. Cookie Policy.
Start a Discussion

Does Process Simulation Have Any Value?

Vote 0 Votes

A question that's been going around of late, and seems to have been started by Steve Towers, is does process simulation have any value?

15 Replies

| Add a Reply
  • If saving a few seconds off every transaction or activity performed by a business process worker can amount to a large saving, I'm sure that simulation has a value. Simulating potential process improvements and feeding back the results of real trials has got to be more effective than just making changes and hoping for the best.

    But really, how many businesses (outside of call-centers) have hundreds of people sat in a room doing similar activities, to make 5 second savings add up to millions of dollars? Making a process lean and removing obvious waste does not require simulation. It can be done with a brain, some experience and a little time.

    In my opinion, most companies are in the position of needing to fix their current processes to get to a baseline of something stable, repeatable and measurable, long before they can start running statistical analysis and simulations for really fine tuning it.


  • I think the question begs another question, which is, "what do you consider BPM to be?"

    I consider BPM to be a practice that focuses on the identification and manipulation of the key performance metrics that results in success or failure. I don't know of any organization that can "afford" to manipulate operational processes without any indication of the effect of the manipulation. Simulation tools are critical to gaining insight, as much as can be gleaned, about the impact of a change, prior to a change being enacted.

  • While there are certain types of processes that need and can benefit from simulation if done properly, overall I believe that simulation is over-hyped. I have written a lot about this topic in my post "The Hype about Simulation and Optimization"

    For most companies my recommendation is to focus on automation and not get distracted by simulation/optimization. Once they have their important processes automated then they can focus on simulation and optimization for the processes that can benefit from it.

    Leadership BPM

  • Hi Rashid, i would have to disagree with you on this one.

    Simulation shouldn't be an end by itself, it is just a step in the automation process. Once a workflow has been set up, simulation helps identify certain pain-points or extraneous steps that may not be apparent but results in a "d'oh" moment :-)

    Yes, i agree workflow simulation is being hyped, just like ISO standards, CMMi, BPR, etc. This should be just another fish in the BPM ecosystem.

    On a side note, can we escape the hype cycle? ;-)

  • Asking the question in absolute terms "does it ever provide any value" is rather misleading. Some organizations will see a clear benefit, but those organizations are very rare.

    It is a bit like asking "Will a spoiler (an aerodynamic device designed to reduce the drag of a vehicle) help reduce my commute time?" Theoretically, yes, a bit faster acceleration will shorten the time, but if your commute is like mine, traffic and congestion causes most of the delay and acceleration is not the issue.

    My experience is that most organizations can benefit purely by being able to see what their processes are. When we do a process discovery for an organization, all people need to do is look at the reality of their existing process, to immediately see improvements. There is no need to simulate to find ways to work better. That is where 95% of all businesses are today. Simple visibility of the process will show areas of improvement which should be addressed first.

    Some of the rest of the organizations have well designed processes which have been broadly reviewed, and nobody can find any obvious improvements. These organizations will see a clear benefit by trying simulation scenarios to find an optimum -- primarily because process behavior is not always intuitive or obvious. But it would be misleading to generalize this by saying that simulation always has value.


  • I am with Phil and Keith on this one. Most of the organizations we are dealing with are just not at the stage where it makes a big difference. These organizations are at the stage where two things make a huge difference for them:
    1) Process Discovery - This first step allows the organization to actually come to terms with their process by seeing it clearly
    2) Process Automation - Just getting the process moving inside an automated workflow engine also produces a huge yield.

    If you don't have these steps done (and a few others), forget about simulation.

    On the other hand, if you are dealing with true systems issues, STP, or processes that are very resource intensive, then I am sure shaving off a few seconds can result in big savings. But I think this is a very specific market segment.

  • In terms of where we are today, I very much agree with the previous contributors – simulation has only had a minor role to play so far. However, looking to the future I think its role will grow dramatically, because the trend towards process automation (through a combination of the adoption of both BPM and more process-centric ERP/product systems) is both a key enabler and a demand driver for process optimisation technology.

    As a manager in a process factory that has had automated processes for a number of years, I think I would expect my technology supplier to offer me a way to easily visualize processes (eg at the use case level), incorporating feedback data from the system as my team uses it, with the option of simulating change. This should go well beyond the ‘bottleneck discovery’ usually suggested as an example of success in process simulation – future process optimisation tools will highlight common keystroke errors and delays in typing to highlight areas for process improvement, perhaps also matching the usage profiles of individuals to new tasks (‘on her work to date, Anita would process loan reviews much faster and with a higher success rate than Juan, whose work patterns suggest he would be more effectively deployed in contract document production’).

    I watched Avatar last night. I wonder if someone asked one of the Hollywood futurologists responsible for it what ‘process optimisation technology’ might mean in 2020, what they would come up with?

  • The short answer is that process simulation does have value. However, it is not the most important question to answer about a business process. Businesses will get huge value from understanding, documenting and automating their processes.
    Many businesses I encounter still use paper based manual process that are inefficient and prone to all kinds of failure. The last thing they need is deep analysis and simulation. Once a process is defined, automated and working within their business they can optimize it through different simulation methodologies.

  • In order for simulation to have value, you have to be able to make accurate guesses for a lot of things (arrival rate distribution, task time distribution, data values expected, etc). Even if you are using historical data and testing a change to a process, you still have to come up with guesses for the aspects of the process that are new. This is hard work and the result is only as good as the accuracy of the guesses. I think people know this and there are just not very many companies that have the time or the modeling skills necessary to make the exercise worth the effort.

  • Having been part of the original discussion from the start, and having been pointed at by Steve Towers for my recent presentation which included some simulation results, I must jump in.

    First, I agree entirely that simulation is another tool in our BPM toolkit, and is not an end in itself. It is useful in some situations, and not in others. Figuring out when it will add greater value than the effort needed to build a good simulation is an important activity to proceed construction and execution of simulations. And having maturity in other areas of BPM is a reasonable precursor to require before going off and using simulation. It is a more sophisticated tool than many organizations are ready to benefit from.

    That said, where it is useful, simulation can be an extremely powerful tool. Let's break out of the mold of thinking that simulation is only useful for process optimization. Think business case assessment, think innovation in process (what far out ideas might I try in a simulation that I might never be willing to explore on the floor?), think pre-planning for events that we know are likely and we want to have some solutions in our hip pocket (e.g., I can show you how to save 80% on the cost to quickly mitigate a volume surge in a loan processing environment by using the non-intuitive solution I found using simulation, versus the straightforward, intuitive solution we (and many others) have always used in the past).

    And to those who think we need to nail all process parameters on the head to get a useful simulation, think again. I have gotten very useful and valid directional guidance from simulations that I know are not nailing these parameters, because the dynamics of the process are relatively insensitive to some (but not all) of those parameters, and the dynamics are exactly what I am trying to understand. You won't get that from static process diagrams. So, identifying those 'key driver' parameters is very important, and we can focus our efforts on getting those nailed down accurately, and let the other parameters be less precise, and not pay an undue price for that.

    Hype or no hype, I have had success with simulation, and care not a whit about the hype so long as I can do what I need to do for my organization and deliver real... and unique... value.

  • I've never had cause to use simulations in the fairly holistic way I've seen Mark use them, but I have had great success in applying them to discrete subprocesses.

    As one such example, i was able to demonstrate that making drivers either do their own data capture OR use automated devices after shift, might either take an extra 30 minutes, OR have a significant cost for a bit of new kit and training, BUT would improve inventory accuracy levels by ove 4000%, thus reducing COGS processing times, post-good-reciept workflows, period end closing, etc.

    All in all, a clear saving of several millions per quarter for a relatively small business.

    And thats just one little process (which admittedly, i looked at because some earliers sims had shown it was quite crucial as a bottleneck)

    So does it have value? Heck yes.
    Is it a pancea? Of course not.

  • Yes it can. Any process that has large amounts of work flowing through it can potentially benefit from simulation (it doesnt have to be processes with large amounts of people doing the similar tasks). Simulation can identify bottle necks very quickly...

    However, would I recommend simulation. No....Like many things with BPM, its great for demonstrations but in the real world it just doesnt cut it...Most bottle necks and issues can be quicky seen with good UAT. More often than not I find the investment and time in simulation doesnt provide a big enough pay off...Of course there will be exceptions, but they are just that, exceptions...

  • Where would the construction industry be without simulations?

    CPM does simulation on time, money and resource availability.

    You may or may not know that forward projections are also possible in healthcare where durations are NOT known (the operation may take 1 hour, or perhaps 4 hours) but this does not take away from the need to allocate, level and balance resources.

    Then we have a huge gap after simulation which is how do I use a software system to guide real-time processing on a day to day basis?

    Finally, there is a great opportunity to train new staff by taking best practice protocols, running these and letting staff make decisions, then seeing how long it took them to get through the workflow and what pathways they took. You can score them and rank their performance.

  • I think everyone here has given good, honest answers - everyone is right. Why? Because "it depends". See my post here - http://bit.ly/9aeUEm

  • I guess I am agreeing with most of the previous posts whan I answer it depends - especially on the type of business process.
    For integration processes then simulation can help with understanding, debugging and tuning the process. If the process is a human process, then simulation as it is done today has very little value.
    It would be interesting to think through what you need to be able to truly simulate a human centric business process.

Add a Reply

Recently Commented On

Monthly Archives