M4E Research #1: September 2023
This week I’ve been delayed a little bit because I was trying to initiate a new line of posts. Let’s try how this works! Each month I will include 3 papers related to manufacturing (Industry 4.0, Lean, etc.) that I found interesting. These papers will have been released the previous month, September this month (they may be not accepted yet).
So let’s start the first issue of M4E Research. This month relevant papers are the following ones:
Qualitative and quantitative evaluation of a methodology for the Digital Twin creation of brownfield production systems
Abstract: The Digital Twin is a well-known concept of industry 4.0 and is the cyber part of a cyber-physical production system providing several benefits such as virtual commissioning or predictive maintenance. The existing production systems are lacking a Digital Twin which has to be created manually in a time-consuming and error-prone process. Therefore, methods to create digital models of existing production systems and their relations between them were developed. This paper presents the implementation of the methodology for the creation of multi-disciplinary relations and a quantitative and qualitative evaluation of the benefits of the methodology.
Authors: Dominik Braun, Nasser Jazdi, Wolfgang Schloegl, Michael Weyrich
Discussion: Brownfield production systems refer to industrial facilities or sites that are already developed or under development, with existing infraestructure and resources. These are the most challenging environments for an Industry 4.0 approach, as the lack of standards or high heterogeneity of them is quite common in old machines or productions systems. Hence the importance of having viable methodologies to develop digital twins and link them correctly to real assets.
Link: https://arxiv.org/abs/2310.04422
A Comprehensive Survey on Rare Event Prediction
Abstract: Rare event prediction involves identifying and forecasting events with a low probability using machine learning and data analysis. Due to the imbalanced data distributions, where the frequency of common events vastly outweighs that of rare events, it requires using specialized methods within each step of the machine learning pipeline, i.e., from data processing to algorithms to evaluation protocols. Predicting the occurrences of rare events is important for real-world applications, such as Industry 4.0, and is an active research area in statistical and machine learning. This paper comprehensively reviews the current approaches for rare event prediction along four dimensions: rare event data, data processing, algorithmic approaches, and evaluation approaches. Specifically, we consider 73 datasets from different modalities (i.e., numerical, image, text, and audio), four major categories of data processing, five major algorithmic groupings, and two broader evaluation approaches. This paper aims to identify gaps in the current literature and highlight the challenges of predicting rare events. It also suggests potential research directions, which can help guide practitioners and researchers.
Authors: Chathurangi Shyalika, Ruwan Wickramarachchi, Amit Sheth
Discussion: One big problem for AI and data analytics applications in manufacturing is the lack of relevant data and how to predict events that are quite rare (e.g. some types of machine breakdowns that happen with low probability). This paper underscores the challenges associated with imbalanced data distributions, where occurrences of rare events are significantly outnumbered by more common events. Some notable trends in this area include the incorporation of domain knowledge, where researchers are exploring the fusion of data-driven approaches with qualitative insights, expert systems, and knowledge graphs to bolster prediction reliability, ensemble methods and also the use of multi-modal data.
Link: https://arxiv.org/abs/2309.11356
Split Federated Learning for 6G Enabled-Networks: Requirements, Challenges and Future Directions
Abstract: Sixth-generation (6G) networks anticipate intelligently supporting a wide range of smart services and innovative applications. Such a context urges a heavy usage of Machine Learning (ML) techniques, particularly Deep Learning (DL), to foster innovation and ease the deployment of intelligent network functions/operations, which are able to fulfill the various requirements of the envisioned 6G services. Specifically, collaborative ML/DL consists of deploying a set of distributed agents that collaboratively train learning models without sharing their data, thus improving data privacy and reducing the time/communication overhead. This work provides a comprehensive study on how collaborative learning can be effectively deployed over 6G wireless networks. In particular, our study focuses on Split Federated Learning (SFL), a technique recently emerged promising better performance compared with existing collaborative learning approaches. We first provide an overview of three emerging collaborative learning paradigms, including federated learning, split learning, and split federated learning, as well as of 6G networks along with their main vision and timeline of key developments. We then highlight the need for split federated learning towards the upcoming 6G networks in every aspect, including 6G technologies (e.g., intelligent physical layer, intelligent edge computing, zero-touch network management, intelligent resource management) and 6G use cases (e.g., smart grid 2.0, Industry 5.0, connected and autonomous systems). Furthermore, we review existing datasets along with frameworks that can help in implementing SFL for 6G networks. We finally identify key technical challenges, open issues, and future research directions related to SFL-enabled 6G networks.
Authors: Houda Hafi, Bouziane Brik, Pantelis A. Frangoudis, Adlen Ksentini
Discussion: Federated Learning is a machine learning approach that enables training on decentralized data residing on multiple devices or servers without the need to exchange the data itself. In traditional machine learning models, data is collected into a central server or data center, where the model is trained. However, in federated learning, the training process occurs on the local devices, and only the model updates or aggregated gradients are shared with the central server or a coordinating node. This approach is more and more interesting to protect privacy and create machine learning models in a collaborative way. This paper emphasizes the importance of these methods for novel communication systems (6G). 5G and 6G technologies will be for sure quite important for smart manufacturing use cases in the future, we will discuss more about this in later issues.
Link: https://arxiv.org/abs/2309.09086