iPoint Blog
The irst steps in modeling real life systems

“Let’s stop re-inventing the wheel with everybody starting from the scratch when creating a material and energy flow model!”

Do you still rerecord, what you thought and expected when you started with your first flow model of a real world production system? I remember that there were so many questions on how to model best, on which degree of detail to use, which of the flows and processes to focus on, and over all the question, if I’m really so innovative that I’m the one dealing with all these questions for the first time and why I can not build on the efforts of other, who have answered these questions for their use case earlier…

A material and energy flow model can grow to a remarkable complexity

After having modeled quite a few systems in different realms in the past years I found that including all processes of e.g. a production system in a detailed material and energy flow network, the model can grow to a remarkable complexity. Due to missing standards or norms on how to model a system in a material and energy flow network, the resulting network models from different modelers are in large part exceedingly different in design, detailedness and completeness, among others. The models might serve well for the original author and for the intended purpose (e.g. a LCA or an optimization project), but may not be adequate to enable (re-)use of the valuable knowledge and data stored in the model for analysis with different perspectives or by another person, bringing at risk the opportunity of an actually integrated, multi-perspective view of a production system.

To fully appreciate the time and money invested in the elaboration of a material and energy flow model, it should serve for all kinds of analysis and tools applicable on such a model.

The solution – a framework of rules and patterns for the modelling of production systems

A possible solution that I would like to explore further is a framework of rules and patterns for the modelling of production systems, which will help to structure the models in the best possible way and to support or at least facilitate integration of the different perspectives for eco efficiency evaluation and optimization, including among others exergy analysis or material flow cost accounting.

The design pattern approach

An analogy from the realm of computer science is the application of design patterns. These patterns evolve as the most effective, efficient and reusable solution to common design problems (Gamma, 1995). As object oriented design in computer science is also partly an approach to model a representation of real world objects and processes, it is considered if this approach can be used as well on material flow networks that represent industrial processes. The design pattern approach originated in architecture and has found its way into pedagogics and space systems architecting, among others (Bergin, 2012; Hein, 2012; Kelly, 2013).

The idea is to apply the pattern approach to the modelling of material and energy flow models, defining a framework of methodically perceived design patterns and underlying modelling rules for material and energy flow network models which facilitate the creation and reuse of well structured, comprehensible models and enable a multi perspective analysis including efficiency analysis, impact assessment, financial analysis and baseline consumption calculation. Therefor it is necessary to identify common modelling challenges and best practices for their solution (including common antipatterns, i.e. worst practices).

Help Newbies with their first steps in modeling real life systems

An initial set of design patterns for material network models should help Newbies with their first steps in modeling real life systems and facilitate the reuse of well modeled components. Another advantage of design patterns is a recognition effect, spotting already known patterns in a model makes it a lot faster and easier to understand. As all the design patterns should have a short but meaningful name there is also something known as “pattern language”, where the mentioning of the name of a design pattern might serve to communicate a rather complex circumstance in few words.

Three possible categories of design patterns

Up to now 3 possible categories of design patterns have been identified. Modelling patterns help to solve problems on how to present real life production processes in the flow network models, e.g. the “infrastructure equipment pattern” on how to include lighting or climatization in the model of a production system, when there is no direct connection to the reference flow. Efficiency patterns might represent resource-, energy- or eco-efficient solutions for common production system efficiency challenges that can be reflected in modeling (and therefor simulation and planning), e.g. waste heat reuse patterns or renewable energy integration patterns. Consistency patterns should focus on the industrial ecology approach to “close cycles” and may be applied also as integration between different models, trying to connect outputs (sub-products and wastes) of one production system with necessary resource inputs of another (Ayres, 1989).

Contribute to the research and share your experiences

If you would like to contribute to the research and share your experiences – common challenges and as the case may be well working solutions to these problems – please contact me at mfn-patterns@senstab.net or fill in the questionnaire at https://goo.gl/forms/FPW9XEIhRxAdX2b72. In exchange I offer to keep you informed about my work and give access to a first Umberto pattern module gallery, as soon as it will be available.

Add comment