Adaptive Processes and BPMN
Blog: Frank Michael Kraft's Blog
In order to understand, what Adaptive Processes mean, it makes sense to discuss classic process modeling techniques and show, what are the weaknesses of them – in order to be able to propose better solutions.
Also many people ask me: What is the relationship between BPMN and Adaptive Processes and Adaptive Case Management?
BPMN is standardized modeling notation and meta model for modeling business processes. It defines modeling elements, their relationships, attributes and their execution semantics – i.e. what happens if the model is loaded into a workflow engine. Modeling tools and workflow engines can be built based on this specification – and in fact many are. Training of BPMN mainly is learning the modeling elements, their meaning and also how to use then in context – i.e. how to solve some modeling problems with them – as for example in (Allweyer, 2009). Some go further and propagate a certain style of using BPMN, as for example the famous Bruce Silver (Silver, 2009). It is always limited to the BPMN syntax and semantic – i.e. the definition of BPMN modeling elements and the execution semantics. On the other hand, Bruce Silver claims, that if people follow his style, then their BPMN models will be in a certain sense “better” than other BPMN models, that is they fulfill more quality criteria than “just BPMN” models. For that the Bruce Silver style defines additional recommendations and rules that go beyond the BPMN syntax and semantic rules.
Adaptive Processes is a paradigm, that takes a step back and looks at all of this and asks: “Wait a moment – what are we doing here? Is this working? If not, why not? And what can or must be done in a different – better – way?” In that sense Adaptive Processes is not limited to BPMN syntax and semantics. Adaptive Processes is allowed to take any modeling language or even invent a new one, if the solution to modeling problems is better. Adaptive Processes can utilize BPMN plus other process modeling languages, but is not limited to that. It can and does also take elements from SOA and from data modeling like UML. At the same time Adaptive Processes shapes a modeling style for using each of these modeling languages individually and together. It defines a certain style to use BPMN, SOA, UML or more and all of them together. In that sense Adaptive Processes is a BPMN style, yes. It is not the Bruce Silver style. It is a different style. But yes, some of Bruce’s recommendations and rules conform to Adaptive Processes, while others won`t.
What does that mean for BPMN modeling? That means if someone just draws a BPMN model, it is nearly sure, that it is not adaptive. This is the case from the first model. And in fact, if I look into BPMN books that are around even the first models are not adaptive. This is not to blame anybody, because Adaptive Processes of course is an innovation, the result of a steep learning curve of us process experts.
But at the same time Adaptive Processes questions the limitations of BPMN and some of its basic meta model flaws. Of course the Adaptive Process BPMN style tries to avoid these flaws – for example by not allowing certain modeling elements, or by recommendations or rules as of how to use the modeling elements more effectively. But in the end the result will always be limited to what BPMN is able to do. Therefore new modeling languages are being invented, that have the Adaptive gene from the beginning. This currently is an ongoing process and currently it is not standardized as of yet. Different companies propose different solutions and so does AdaPro. After some time these will be standardized. Then I expect that we will have a new process modeling language that is – as I said – adaptive from the beginning. If this is the case, then a mapping can be defined from BPMN to the adaptive modeling language and vice versa. And guess what – the mapping will be easy if people stuck to the Adaptive Process BPMN modeling style, and it will be difficult or impossible, if they haven’t. So if I propagate the Adaptive Process modeling style, then I am anticipating what is coming as a new standard in the future for modeling business processes.
That is why I can propagate a certain BPMN modeling style and criticize BPMN at the same time – which seems paradox. But with this explanation it is not.
I intent to discuss certain models from current literature, show their limitations and problems and by this make clear, why we need Adaptive Processes. It will then be up to you to decide, if you subscribe the Adaptive Process philosophy. I want to be very open about this. It is a yes – no decision. If you decide no – then you will have more freedom to use BPMN, but also more problems as I will argue. If you subscribe to it, it is best done before your modeling project begins. Yes, it is also possible to remodel an existing landscape to be adaptive, but it is clearly more effort than if it is done from the beginning. On the other hand it solves problems you have with your current model landscape that you thought were impossible to solve.
I have experienced this with many clients. Unconsciously the models become simpler, less spaghetti-style and less repetitions, easier to maintain. Often people don`t even know that I teach them the adaptive style, and I don`t tell them if they do not want to know. But I tell you, because I want to establish and argue for the adaptive style, convince you so that you are prepared for the things to come.
Allweyer, T. (2009). BPMN 2.0 – Business Process Model and Notation: Einführung in den Standard für die Geschäftsprozessmodellierung. Books on Demand.
Silver, B. (2009). Bpmn Method and Style: A Levels-Based Methodology for Bpm Process Modeling and Improvement Using Bpmn 2.0. Cody-Cassidy Press .