Resolved
1 votes
Do you think there is a business benefit to simulating case management?
Tuesday, July 29 2014, 09:52 AM
Share this post:
Responses (9)
  • Accepted Answer

    Tuesday, July 29 2014, 10:09 AM - #Permalink
    Resolved
    3 votes
    I think simulation has a number of uses, but this is not one of them for me. I can see using using simulation for scenario planning, case planning and setting constraints for free ranging cases, but the typical discrete and optimization simulation is a long shot here. I see more of a use of process mining to find patterns of success after the fact combined with scenario/case planning. This allows organizations to stay legal and compliant while still morphing to add better practices (emergent behavior). Just my two cents. Jim
    Like
    1
    The reply is currently minimized Show
  • Accepted Answer

    Tuesday, July 29 2014, 10:18 AM - #Permalink
    Resolved
    3 votes
    Forward simulation, a.k.a. predictive analytics, in a case management is useful to alert people when troublesome conditions are approaching. An unexpected influx of cases distributed over many case managers might not be noticed. A forward simulation might be able to warn people that the current case load, based on how long it typically takes to handle such cases, can not be completed in time. The prediction need not be exact. Such a warning could alert managers to bring on extra help, or to change the mode of working to be more efficient. Simulation is used here to provide an indicator that case managers can respond to.
    Like
    1
    The reply is currently minimized Show
  • Accepted Answer

    Tim Bryce
    Tim Bryce
    Offline
    Tuesday, July 29 2014, 11:35 AM - #Permalink
    Resolved
    0 votes
    I look upon case management as more of a project management function, primarily in the area of planning. Allow me to explain. In our Enterprise Engineering Methodology (EEM), we had the ability to model the business logically and physically, thereby giving us the ability to do such things as performing an organization analysis and priority modeling. The latter being a prioritized listing of the enterprise's business objectives and their supporting projects. If the business changed, we could automatically recalculate the company's priorities (not to mention play "what if"). At the heart of our software was a Repository which housed all of the components about an organization, including business, systems, and data resources. As Jim Sinur will remember, we had the ability to perform an "impact analysis" on any information resource, be it a business component, systems or software, or data resources. By doing so, we could determine the effect of change on any component and prepare a more precise plan for implementing it. For example, if the enterprise were to reorganize a part of a business, the Repository would list all of the people and machine resources involved, projects affected, systems, and data. Conversely, if we wished to change a data element, the Repository would list all of the records, files, inputs, outputs involved, not to mention programs and much more. The intent of the Repository is to reduce redundancy of information resources and promote sharing. I cannot image simulation of any kind without such a tool.
    Like
    1
    The reply is currently minimized Show
  • Accepted Answer

    Tuesday, July 29 2014, 11:52 AM - #Permalink
    Resolved
    2 votes
    When you use “declarative” to build in the graphical process design/build environment where all business logic is pre-built “simulation” becomes a bit redundant. Because you can build and change very quickly it becomes part of the interaction with users to get feed back using a first cut of the real application as a prototype. This applies as much to Case Management as to any digital application. Any bottle necks are identified with real time analytical feed back or required changes are readily addressed just change and “re-declare”. In addition a management module allows overview of activity with capability to re allocate work as required As users gain confidence that they really do have say in the way they work so change will become the norm. All this makes “simulation” slightly academic when you could be “playing” with and refining the real application?
    The reply is currently minimized Show
  • Accepted Answer

    Tuesday, July 29 2014, 02:25 PM - #Permalink
    Resolved
    1 votes
    Interesting question. First let's ask what is meant by case simulation. Which is a specific example of just any old "simulation". Are we talking about simulation of a particular process? To what end? What is the benefit of investing time in a simulation? Will the result be a retooling and optimization of a process model? I have found that what might be called "second order business activities", such as competitive research, customer market research, business process simulation etc., all of which are usually thought of as "costs", doesn't happen that often (safety is also such an activity, unfortunately). Show me an organization where systems theory is used to decide whether or not to build another warehouse in Cleveland, and we may have found a candidate for case-oriented or any kind of process modeling. In this example (warehouse location), the scope of the process model is a whole organizational function. Most simulation that occurs isn't at such a grand level. In my experience, "supports simulation" is more of a check list item than anything used frequently. This is unfortunate, but only a reflection of where we are in a technology adoption curve.
    The reply is currently minimized Show
  • Accepted Answer

    Wednesday, July 30 2014, 02:55 AM - #Permalink
    Resolved
    2 votes
    I think that business benefit to simulation of all types of work including case-management one. Especially, line managers - http://improving-bpm-systems.blogspot.ch/2010/03/linkedin-how-do-we-measure-work-flow.html Maybe the question should be how to carry out simulation of case-management-type-of-work? For example, if the case management is about empowering the knowledge workers then only a particular knowledge worker can do some “forward simulation” for his/her case. As the same knowledge worker also executes this case then there is a potential conflict of interests. Thanks, AS
    The reply is currently minimized Show
  • Accepted Answer

    Wednesday, July 30 2014, 04:32 AM - #Permalink
    Resolved
    3 votes
    To deal with this question requires not process or case management knowledge but an understanding of mathematics. Only if you have an environment in which very strict conditions are repeatedly found, such as in physics, simulation can prove benificial. One can use also simulation in chaotic environments with random events injected to see how a system will react. So in theory one can use simulation to test a case management scenario. The reality is different. The usability of simulation is to test how a well-defined system (i.e. an airplane) will react in stress conditions in a complex environment. Once simulation provdies a satisfactory performance, both pilots and the control computer onboard a modern airplane are trained and programmed not to use that system outside these well tested conditions. So simulation is usable for hard-coded applications and process management when one is designing a well-defined logic or process structure and one wants see how it will fare in practical use. Once again the process managment must not be used outside these tested conditions as the results will be unpredictable, with usually catastrophic results as the system collapses. Therefore simulation makes little sense in case management, because there is no predefined process structure. The whole point is to deal with the unpredictable reality, which applies to 80% of business processes. So what do you want to simulate? Identifying bottlenecks? One can use simulation to test the ACM system but not the case/process structures that will be created by the business while they use it. The business performers can and should create and adapt the processes to achieve the process goals and customer outcomes. That can't be simulated and there is no point in doing so. All the business performers knowledge would have to be programmed into the simulation, which is impossible. A 'forward simulation' of what-if scenarios is thus not practically possible, because chaotic systems with a number of influencing parameters develop a huge number of potential state space patterns. The use of the word 'simulation' is utterly wrong for this because there is no predictable rule set and data set. The only thing that can be done is to collect action/outcome data from past cases and identify probable scenarios. Administering medicine to a patient is linked to a checklist of go/no-go criteria and can be looked at with a what-if scenario, but the possible reactions and outcomes are absolutely unpredictable due to system complexity. A pattern-matching engine can filter past case scenarios for like patterns and calculate a likelyhood of actions to outcomes given the current state space scenario. That's also what a good doctors does using his experience. And it is what our pattern-matching engine does for best-next action recommendations. But that IS NOT SIMULATION and it is not process mining, because there is no process to mine. These are just state spaces of information and their progression over time. These state spaces do only partially reflect the reality of complex systems that can't be modeled and thus not simulated, despite the fact that many scientists do the opposite in for example those idiotic climate predictions. So if you want to waste time and money then do simulations for process and case management ...
    • Dr Alexander Samarin
      more than a month ago
      As far as I know, Royal-Dutch-Shell group used very successfully some kind of “a 'forward simulation' of what-if scenarios” which allowed them to enter “seven sisters”. But, maybe the strategy planning is less complicated that ordinary case management.

      Thanks,
      AS
    The reply is currently minimized Show
  • Accepted Answer

    Geoff Hook
    Geoff Hook
    Offline
    Thursday, July 31 2014, 12:42 PM - #Permalink
    Resolved
    2 votes
    I would definitely agree with John Morris in that as with all analysis it is crucial to know why you are doing the simulation. This drives the selection of the appropriate conceptual model, data to be used, experiments to be conducted and results to be generated. At Lanner we recognise the Keith Swenson comment that is very much about simulation being used for ‘prediction’ so that operational managers can make better decisions on resourcing etc, based on foresight. David Chassels seems to question the value of simulation when you can change the application itself. I would argue that whilst experimenting with the real system can be an option, on occasions this can be too risky, the potential downside of a poor option being selected too great, this is where simulation ‘off-line’ helps out. If you have a business issue or purpose for which it is valid to address with simulation, the next question is about building the right model, using the best data and experimenting correctly. An interesting question for Case Management Experts is can simulation models be directly constructed from CMMN similar to what we do from BPMN? Whilst the ease in which simulation models can be constructed from BPMN is an advantage today, it can also be a hindrance….too easy can lead people to go that route even if it’s not appropriate. In summary I feel simulation can be useful in case oriented applications, but don’t necessarily assume it is the same simulation model design in all applications. It really does depend on the business question.
    The reply is currently minimized Show
  • Accepted Answer

    Friday, August 01 2014, 08:49 AM - #Permalink
    Resolved
    1 votes
    Geoff, I'm delighted that someone has mentioned Lanner! I was tempted to refer to Lanner myself, and actually did so in spring 2011, in the context of discussing BPM and simulation: When Worlds Collide -- Or Don't -- BPM & Business Simulation (see Comment No. 1), and then again in 2013 in the context of the need for modelling: Why Big Data Can Be Like A Jelly Fish: No Bones & You Might Get Stung There's a huge upside to simulation; the challenge is making the business case and investing in the skills. We're really just beginning. Lanner's work around "BPSim" is very relevant for anyone wanting to explore the opportunities.
    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
276 Replies
25/09/2016
David Chassels
269 Replies
23/09/2016
Emiel Kelly
220 Replies
23/09/2016
Bogdan Nafornita
209 Replies
25/09/2016
E Scott Menter
182 Replies
23/09/2016