Ground-Floor BPM

Scott Menter

Simulated Benefits

user-pic
Vote 0 Votes

I once suggested on Twitter that simulation in the BPM world is a non-starter. I don't remember my exact words, but the gist of it was, "Nobody uses simulation." A famous and influential analyst called me out, asking me if I thought that his clients were "lying to [him]". Twitter's not a great place for that kind of argument, so I'll respond here.

Yes, Mr. Widely-Quoted Analyst, they are lying to you.

I am almost never asked by a prospective customer about simulation. Even when the subject arises, it is generally expressed as a checkbox item, and never as a core requirement. In fact, other vendors have told me that customers who buy the simulation modules they offer eventually, and without fail, exchange them for credit against some other component. Moreover, in my own career as an IT executive, I have simply never encountered a process that made me think, "gee, I wish I could simulate this bad boy".

Simulation delays automation, and I don't like to wait. There is an instant benefit from automation, even if the underlying process is not very efficient. To put it another way: I'd rather automate a poorly designed process today than spend six months analyzing, simulating, and optimizing it before automating. A decent BPM solution should make deployment easy, and then provide you with plenty of data on the behavior of your process--the perfect input for optimization. I'll take real data generated from actual process executions over simulation results every time.

Do BPM technology customers tell their analysts that they are using simulation? Sure: their employers paid a lot for it (sometimes at your recommendation, Mr. Analyst). The cognitive dissonance arising from the discovery of its limited value must be intense.

Are you a BPM customer? Has simulation yielded substantial benefits to you and your company? By all means: show me why I'm wrong. There's a space provided for just that purpose below.

9 Comments

| Leave a comment

"Simulation delays automation" - only a pure automation vendor could say this, I have heard similar comments before, don't bother with analysis "It is best to simply automate what you have, get rid of the people and then take a look" - Such rush to act before thinking is what has got IT type projects in a mess before, and surely will again. There are times when simulation takes too long costs too much and their are times when it makes sense or is vital. Good vendors should be able to share with you metrics and examples of both when and when not to use. Life is never as simple as appears to being made out here. Given that I don't work in the Simulation industry, but have worked with many clients and vendors who do support and have used it to great effect these observations, whilst obviously true for you and your experience of the world, obviously don't tie in the the view of the world by your "widely respected analyst" - and I happen to share views nearer to his, based on what I have seen.

Of course neither view is in fact right or wrong, they are simply our views based on our experiences. Some people base it on a narrow view e.g. a single customer base, some a wider view based on multiple customers bases and some on a very broad view totally unaligned to any vendors view.

I don't know if your post is written from the perspective of a vendor who does bot feel that analysts are giving them the great time they think they should have because of product weaknesses - I agree that if I did not have something and analysts said I should then I would have to justify why I did not have it. Or, whether it is written to stimulate debate, in which case I suspect an interesting set of comments will ensue.

Kind Regards

Mark McGregor
Independent, Author & Performance Coach

www.markmcgregor.com

user-pic

Scott, the issue with your "no simulation" approach is surely that you inevitably rely on trial and error to draw you towards a correct conclusion.

Whilst ignoring the possibilities of simulation does perhaps give you a more rapid start (by definition, as you are missing out a step), the counterargument that it is best to understand your likely destination before you embark on a journey might also apply.

I presume your software suite doesn't put a mark in the simulation check box, by the way :P

Hi Mark: You're right, in that I do think an iterative approach to process deployment makes the most sense, for a variety of reasons. And, I have to admit, I don't mind "stimulating debate" once in a while. :) But I will say that in the non-vendor periods of my career, I have worked in a variety of industries and never ran across the need for simulation.

I suppose my point is not that simulation could never be of use to anybody, but rather, that it's been overhyped to the point where far more customers are paying for it than actually will ever use it to their advantage.

Thanks for your comment!

Hello Simon! I think I'd use a different analogy. Process discovery takes care of knowing your destination, and modeling the process (using either traditional or non-traditional techniques) lays out the path to get there.

Simulation is more like trying to imagine what the traffic will be like on each leg of the trip, and then changing the route based on that analysis. Most of the time, the answer isn't going to change—you just want to make sure that you stay tuned to the traffic reports in case something comes up while you're on the way. That's what our BPM software does (since you asked :), and I honestly believe that's the best approach.

Great comment, thanks!

My company is a BPMS vendor that offers a simulation plug-in to Visio as part of our suite. We have hundreds of customers who not only bought simulation technologies but use them as well.


Our pre-sales organization also uses simulation prior to any implementation to help our clients validate their business case for future BPMS investments. This validation often leads to the future embrace of simulation technologies as part of normal process discovery efforts, and even later as part of ongoing optimization efforts.

If your arguement is that simulation has not been taken up by the masses, I will agree with you. Simulation requires a higher level of business process improvement maturity. Not all organizations express or have achieved that level of maturity, but that does not mean that none have.

[Shameless promotion] We have several case studies on our website that include Visio-based BPMN models with simulations embedded in them. These models + simulations were developed along with our customers to help validate and optimize process environments.

Hi Derek: I hear you. But I'd argue that simulation per se isn't the best way to produce data for optimization, and I've already argued that simulation in the discovery process is an obstacle to rapid deployment.

But, as a thank-you for your great comments, I'm happy to provide a platform for your shameless promotion. :)

Scott, you certainly said the right things to "stimulate debate" especially from the simulation fans/vendors! My very first experience with what would now be called BPM technology was in fact a rush to automate, and it was a disaster. It's coloured by mindset ever since. Although a sample size of one, it keeps what you say from resonating with me.

That being said, there is a lot of truth in what you say: people buy the simulation technology and don't use it, or don't use it effectively enough to make it worthwhile. In many BPMS's it's a "fake feature" as Bruce Silver would call it, again calling into question the value.

I think you might have missed the key point though: whether through automation or simulation, the real goal is optimization. Simulation is but one tool. If it gets you to optimization faster, then why not use it?

I've touched on this here:
http://sim4bpm.com/2011/03/07/keeping-the-end-in-mind/

Hello John. If you've been around as long as I have (and that's seeming like a very long time at this point), I'm guessing you've probably also had the experience of watching projects die as they are analyzed to death. Seeing that happen over and over again has left me with a strong bias to action—which is not to say that analysis shouldn't follow right behind.

I do agree that optimization is a goal. But in my view, <shameless plug> as reflected in this humorous* video </shameless plug>, what counts most in optimization is real world data generated from process executions.

Even though you did not use the appropriate tagging for your shameless plug, I do appreciate your comment, so I'm OK with it. :)

Thanks!

______
*) I hope.

OK, so maybe we are still - and for far too long, in my humble opinion - in the early adopter-stage of business process simulation.

I have many years of experience as a developer and successful applier of simulation in high tech engineering/manufacturing industries (jet engines, helicopters, HVAC, people movers, fuel cells, etc.). It tool some time before simulation became a key tool, and once it did, those industries never looked back.

We aren't there yet in service industries, and the challenges are unquestionably different. However, having been part of the simulation revolution in the engineering arena I have no doubt that the early adopters of business process simulation who direct their efforts thoughtfully and energetically towards the use of simulation stand to reap rich rewards.

That said, simulation is, or could/will be, another tool in the process design and process improvement toolkits and in and of itself is no silver bullet. But it can be an incredibly valuable tool, particularly when it enables a business to try things out that it would never, ever do in the field. I personally have used business process simulation to optimize headcount ratios for teams with multiple roles, and to identify a non-intuitive way to save 75%+ when addressing volume surges in a fulfillment process.

I remember seeing the arguments about the impossibility of running high-end graphics on PCs back in the early 90's. What is done today is amazing on its own, and certainly put the naysayers' claims to shame. I expect that some of those who adopt and effectively use business process simulation will be able to demonstrate the same sort of success to the business process simulation naysayers.

Going back to the comment "I'd rather automate a poorly designed process today than spend six months analyzing, simulating, and optimizing it before automating," automating a bad process is never a good thing in my opinion. Doing something badly, but faster, sounds like a valid definition of insanity. OK, spending six months doing nothing but simulating to find an optimal solution doesn't feel right either. Give any competent process improvement professional that process and a couple of months, and simulation or no simulation, changes to that process can be defined that will make for a much better outcome of automation than just blindly jamming it into an automation system without considering and remedying key weaknesses first.

Maybe that fact that you "have simply never encountered a process that made me think, 'gee, I wish I could simulate this bad boy'" is a statement more about you, than about the potential for simulation. I know that I have had that thought, turned it into action, and delivered valuable results.

But maybe that's just me.

Leave a comment

Scott covers ground-level BPM issues of interest to enterprise users who are tasked to do more with less while improving business processes.

Scott Menter

E. Scott Menter is the VP of Business Solutions for BP Logix, a provider of business process management (BPM) solutions to corporate, non-profit, and government organizations. In addition to technology leadership positions in financial services and higher education, Scott also spent over a decade leading his own identity management software firm. Contact Scott at Scott.Menter@bplogix.com
or http://twitter.com/ESMatBPL.

Recently Commented On

Recent Webinars

    Monthly Archives

    Blogs

    ADVERTISEMENT