Using Physical Factory Simulation Models for Business Process Management Research

Abstract

The production and manufacturing industries are currently transitioning towards more autonomous and intelligent production lines within the Fourth Industrial Revolution (Industry 4.0). Learning Factories as small scale physical models of real shop floors are realistic platforms to conduct research in the smart manufacturing area without depending on expensive real world production lines or completely simulated data. In this work, we propose to use learning factories for conducting research in the context of Business Process Management (BPM) and Internet of Things (IoT) as this combination promises to be mutually beneficial for both research areas. We introduce our physical Fischertechnik factory models simulating a complex production line and three exemplary use cases of combining BPM and IoT, namely the implementation of a BPM abstraction stack on top of a learning factory, the experiencebased adaptation and optimization of manufacturing processes, and the stream processing-based conformance checking of IoT-enabled processes.

Using Graph Embedding Techniques in Process-Oriented Case-Based Reasoning

“…This confirms the assumption that partly motivated this paper. The effect might be amplified by domains that are even more complex than CB-II, such as argumentation [25] or flexible manufacturing [4]. Additionally, it is shown that sGEM and sGMN, although integrating rich semantic information, are not able to consistently outperform FBM in MAC/FAC retrieval tasks.…”

“…Our semantic graph format allows the integration of semantic knowledge within graph structures. The format is mainly used to model processes and workflows in various domains (e.g., [3][4][5]9]). We represent all cases and queries as semantically annotated directed graphs referred to as NEST graphs, introduced by Bergmann and Gil [7].…”

“…If the atomic types are extended by a data type for timestamps, for instance, it is possible to add a new one-hot encoding for this extension. Second, only encoding the base types, i.e., string, float, list, etc., can be insufficient in certain scenarios where the domain model is very complex (see [4,5] for examples of such domains). The proposed approach can be extended in these cases to use individual one-hot encodings for specific data types in the domain model.…”

“…Case-Based Reasoning (CBR) [1,2] is used widely across different domains, e.g., for cooking assistance [3], in smart manufacturing [4], and for data mining support [5]. CBR is a methodology for problem solving where problems and their respective solutions (bundled as cases in a case base) are used to solve upcoming problems (queries).…”

“…POCBR particularly aims at managing procedural experiential knowledge, e.g., retrieving workflow models from large repositories [7]. The involved processes are usually represented as semantic graphs [4,5,9], which adds complexity to all involved tasks. For instance, in similarity-based retrieval, the main influential factor on performance is the definition of the similarity measures and the underlying case representation.…”