We propose an unsupervised anomaly detection approach based on a physics-informed diffusion model for multivariate time series data. Over the past years, diffusion model has demonstrated its effectiveness in forecasting, imputation, generation, and anomaly detection in the time series domain. In this paper, we present a new approach for learning the physics-dependent temporal distribution of multivariate time series data using a weighted physics-informed loss during diffusion model training. A weighted physics-informed loss is constructed using a static weight schedule. This approach enables a diffusion model to accurately approximate underlying data distribution, which can influence the unsupervised anomaly detection performance. Our experiments on synthetic and real-world datasets show that physics-informed training improves the F1 score in anomaly detection; it generates better data diversity and log-likelihood. Our model outperforms baseline approaches, additionally, it surpasses prior physics-informed work and purely data-driven diffusion models on a synthetic dataset and one real-world dataset while remaining competitive on others.

This is an extended abstract of the manuscript 'Solving Industrial Fault Diagnosis Problems with Quantum Computers' [1] that was published in the journal Quantum Machine Intelligence in 2024. The article presents two approaches to perform fault diagnosis: (i) using Grover's algorithm, and (ii) using the Quantum Approximate Optimization Algorithm. We found the Grover algorithm generates more solutions and needs some post-processing to obtain minimal diagnoses, but it is much faster than the more accurate QAOA approach.

In this paper, the use of Large Language Models (LLMs) for the configuration of hybrid TSN/5G networks is investigated. We discuss promising use cases where LLMs offer significant potential to simplify complex configuration tasks. Particularly, we consider two important scenarios: In the first scenario, the LLM functions as an engineering component enhancing traditional network control entities such as the Centralized User Configurations (CUCs) and Centralized Network Configurations (CNCs) of TSN networks and interacting with the 5G control plane. In the second scenario, the LLM serves as an interactive assistance tool for users performing manual configuration tasks. For these use cases, an LLM-based architecture for network configuration is proposed, which consists of a Retrieval-Augmented Generation (RAG) system, a verification component, and an orchestration layer. Within the framework of this architecture, we introduce LLM-based methods to enhance the reliability of configuration in complex real-time networks, leveraging strategies such as divide and conquer, prompt engineering, and verification.

Artificial intelligence (AI) is becoming increasingly popular in the context of industrial manufacturing. However, in industrial manufacturing in particular, it is important to ensure the trustworthiness of AI. In this article, we give an overview of different aspects of trustworthy AI in this context. At first, we divide the topic into three different components, namely data, algorithm, and IT infrastructure. We identify several aspects of these components that are required for the trustworthy use of AI. Measures to achieve trustworthy AI are then derived and illustrated on the basis of a specific use case. It is further intended in the ongoing work to evaluate the impact of the individual measures.

In this article, we investigate in how far quantum computers can be leveraged to solve NP-complete fault diagnosis problems within the area of industrial cyber-physical systems. Therefore, two approaches are proposed which exploit quantum computing to solve diagnosis problems: The first method employs Grover’s algorithm, and the second is based on the Quantum Approximate Optimization Algorithm. To show the industrial application, we present an integrated approach to learn the diagnosis model from process data, check whether the model is suitable, and use it for diagnosis. The result is a method for quantum industrial fault diagnosis. For this approach, the diagnostic capabilities and the runtime have been evaluated on an IBM Falcon processor using three publicly available benchmarks from the process industry. Further, the scaling between quantum computers and classical PCs has been analyzed.

In this work, we study the task of scheduling jobs on a single machine with sequence dependent family setup times under the goal of minimizing the makespan, that is, the completion time of the last job in the schedule. This notoriously NP-hard problem is highly relevant in practical productions and requires heuristics that provide good solutions quickly in order to deal with large instances. In this paper, we present a heuristic based on the approach of parameterized local search. That is, we aim to replace a given solution by a better solution having distance at most k in a pre-defined distance measure. This is done multiple times in a hill-climbing manner, until a locally optimal solution is reached. We analyze the trade-off between the allowed distance k and the algorithm's running time for four natural distance measures. Example of allowed operations for our considered distance measures are: swapping k pairs of jobs in the sequence, or rearranging k consecutive jobs. For two distance measures, we show that finding an improvement for given k can be done in f(k) · n^𝒪(1) time, while such a running time for the other two distance measures is unlikely. We provide a preliminary experimental evaluation of our local search approaches.

In this paper, a hybrid model of regularized Long Short-Term Memory (LSTM) and autoencoder for fault detection in automated production systems is proposed. The presented LSTM autoencoder is used as a stochastic process model, which captures the normal behavior of a production system and allows to predict the probability distribution of sensor data. Discrepancies between the observed sensor data and the predicted probability density distribution are detected as potential faults. The approach combines the advantages of LSTMs and autoencoders: The correlations between individual sensor signals are exploited by an autoencoder, while the temporal dependencies are captured by LSTM neurons. A key challenge in training such a process model from historical data is to control the information passed through the latent space of the autoencoder. Different regularization methods are investigated for this purpose. Fault detection with the proposed LSTM autoencoder has been evaluated on the use case of an industrial penicillin production, achieving significantly improved results in comparison to the baseline LSTM.

This paper introduces a new diagnostic approach for concurrent discrete event systems (DES) in automated production plants. The proposed diagnostic procedure is based on Petri net models of the DES. Discrepancies between the Petri net models and event sequences, which are observed during the operation of the production plant, are detected as potential faults. A case-based diagnosis is then carried out to identify fault types that are easy to interpret for humans. The proposed diagnostic procedure is evaluated using an industrial conveying system.

In this paper, a quantum computer-aided approach to job scheduling for automated storage and retrieval systems is introduced. The approach covers application cases, where various objects need to be transported between storage positions and the order of transport operations can be freely chosen. The objective of job scheduling is to arrange the transport operations in a sequence, where the cumulative costs of the transport operations and empty runs between subsequent transport operations are minimized. The scheduling problem is formulated as an asymmetric quadratic unconstrained binary optimization (QUBO) problem, in which the transport operations are modeled as nodes and empty runs are modeled as edges, with costs assigned to each node and each edge. An Quantum Approximate Optimization Algorithm (QAOA) is used to solve the QUBO. Evaluations of the quantum computer-aided job scheduling approach have been conducted on the IBM Q System One quantum computer in Ehningen. In particular, the running time for the solution of the QUBO has been investigated, as well as the scalability of the approach with respect to the required number of qubits.

Automated storage and retrieval systems (ASRS) are important in distribution centers and warehouses. To decrease cost or CO2 emissions it is natural to optimize various aspects of an ASRS. In this work, we provide a concept for a two-phase optimization combining two important optimization tasks in ASRS: Given multiple rearrangement jobs, we first sequence these jobs to minimize the total travelling distance of the cranes. We continue the optimization by computing optimal trajectories for the sequence to guarantee energy efficient driving of the cranes. We describe our algorithms for a complex ASRS architecture with two cranes on parallel rails in one aisle. Additionally, we describe how to use our results for parallelization of crane movements in the considered warehouse architecture.

This paper presents a novel fault detection approach for industrial batch processes. The batch processes under consideration are characterized by the interaction between discrete system modes and non-stationary continuous dynamics. Therefore, a stochastic hybrid process model (SHPM) is introduced, where process variables are modeled as time-variant Gaussian distributions, which depend on hidden system modes. Transitions between the system modes are assumed to be either autonomous or to be triggered by observable events such as on/off signals. The model parameters are determined from training data using expectation-maximization techniques. A new fault detection algorithm is proposed, which assesses the likelihoods of sensor signals on the basis of the stochastic hybrid process model. Evaluation of the proposed fault detection system has been conducted for a penicillin production process, with the results showing a significant improvement over the existing baseline methods.Note to Practitioners—Automatic fault detection makes it possible to limit the effects of faults by taking countermeasures at an early stage. In this work, a data-driven fault detection method for industrial batch processes is proposed, in which the underlying process model is learned from training data. The proposed fault detection system can be used for various industrial batch processes without the need for complex and error-prone manual configuration. In contrast to many other data-driven approaches such as neural networks, only a few process cycles are required to create a robust process model. It should be noted that in data-driven fault detection methods, the training data should cover a large part of the process states that occur during error-free process cycles. The developed method is therefore particularly suitable for cyclical processes, which, however, can have alternative process paths and variability between the process cycles.

Abstract Petri nets are a common modeling approach for parallel processes such as transport operations in conveying systems. In industrial applications, the Petri net models are usually created manually, which involves a lot of effort, especially if the modeled systems change frequently. This paper introduces a new learning method to automatically generate Petri nets from sensor data acquired in conveying systems. The underlying approach is to create prefix and postfix trees of possible event sequences and to merge them into a compact graph, which can be transformed into a deterministic Petri net model of the conveying system. Experimental results show that the proposed method produces realistic Petri net models even for conveying systems with ambiguous events.

Abstract In this paper, an LSTM autoencoder approach for online fault detection in industrial production systems is proposed. The presented LSTM autoencoder is used as a stochastic process model, which captures the normal behavior of a production system and allows to predict the probability distribution of sensor data. Discrepancies between the observed sensor data and the predicted probability density distribution are detected as potential faults. The approach combines the advantages of LSTMs and autoencoders: The correlations between individual sensor signals are exploited by an autoencoder, while the temporal dependencies are captured by LSTM neurons. A key challenge in training such a process model from historical data is to control the amount of information passed through the latent space of the autoencoder. Different regularization methods are investigated for this purpose. Fault detection with the proposed LSTM autoencoder has been evaluated on the use case of an industrial penicillin production, achieving significantly improved results in comparison to the baseline LSTM.

AI methods for fault handling in Cyber-Physical Production Systems (CPPS) such as production plants and tank systems are an emerging research topic. In the last years many methods for the detection of anomalies and faults, the diagnosis of the root cause and the automated repair have been developed. However, most of the methods are barely evaluated using a wide range of systems but applicability is shown using single use cases. In this paper, an ensemble of simulated benchmark systems is presented, which allows for a broad evaluation of AI methods for fault handling. The ensemble consists of seven different tank systems from process engineering with varying sizes and complexities and is made publicly available on Github. The suitability of the ensemble is shown using AI methods for fault handling such as anomaly detection, diagnosis and reconfiguration.

In this paper, a new information model for machine learning applications is introduced, which allows for a consistent acquisition and semantic annotation of process data, structural information and domain knowledge from industrial productions systems. The proposed information model is based on Industry 4.0 components and IEC 61360 component descriptions. To model sensor data, components of the OGC SensorThings model such as data streams and observations have been incorporated in this approach. Machine learning models can be integrated into the information model in terms of existing model serving frameworks like PMML or Tensorflowgraph. Based on the proposed information model, a tool chain for automatic knowledge extraction is introduced and the automatic classification of unstructured text is investigated as a particular application case for the proposed tool chain.

In this paper, a model-based routing approach for flexible manufacturing systems (FMS) with alternative routes for the work pieces is proposed. For each work piece, an individual task has to be accomplished, which consists of several processing steps. Each processing step can be executed on alternative working stations of the FMS. The proposed routing method employs a model of the conveying system to find energy efficient and fast routes for the respective work pieces. The conveying system model is based on a directed graph, where the individual conveyors are modeled as weighted edges. It can be straightforwardly applied to several types of FMS by adjusting the application-dependent parameters. Efficient computation of the fastest route through the conveying system is accomplished by means of dynamic programming, i e., by integration of Dijkstra's algorithm in a dynamic programming framework, which is based on the proposed conveying system model. Additional consideration of energy efficiency aspects leads to a Mixed Integer Quadratically Constraint Program (MIQCP), which is solved by substitution of Dijkstra's algorithm by a branch and bound method. Experimental results for an application scenario, where the energy efficient routing method is applied to route work pieces between the working stations of an FMS, lead to 20 % reduction of energy consumption on average.

Manufacturing industry is facing major challenges to meet customer requirements, which are constantly changing. Therefore, products have to be manufactured with efficient processes, minimal interruptions, and low resource consumptions. To achieve this goal, huge amounts of data generated by industrial equipment needs to be managed and analyzed by modern technologies. Since the big data era in manufacturing industry is still at an early stage, there is a need for a reference architecture that incorporates big data and machine learning technologies and aligns with the Industrie 4.0 standards and requirements. In this paper, requirements for designing a scalable analytics platform for industrial data are derived from Industrie 4.0 standards and literature. Based on these requirements, a reference big data architecture for industrial machine learning applications is proposed and compared to related works. Finally, the proposed architecture has been implemented in the Lab Big Data at the SmartFactoryOWL and its scalability and performance have been evaluated on parallel computation of an industrial PCA model. The results show that the proposed architecture is linearly scalable and adaptable to machine learning use cases and will help to improve the industrial automation processes in production systems.

This article addresses the automatic optimization of driving speeds in conveying systems. Electric drives in existing conveying systems are usually accelerated and decelerated according to predetermined movement profiles. Such an approach is inflexible for conveying applications with changing constraints and, in many cases, not optimal with respect to energy efficiency. In the present work, a method for automatic computation of energy efficient movement profiles is proposed. The proposed method is based on accurate models for electric drives and several types of conveying applications such as roll conveyors, belt conveyors and vertical conveyors. Furthermore, joint energy efficiency optimization for two drives, which are attached to an intermediate circuit, is investigated. Thereby, additional constraints on the energy flow between the drives are imposed in order to reduce load peaks and energy feedback into the grid. The resulting optimization problem is a mixed integer quadratic program (MIQP), which can be solved in a few milliseconds. Experimental results show that energy losses of electric drives are cut down by using the obtained non-trivial movement profiles instead of standard trapezoid movement profiles. The additional constraints on the energy flow between two drives lead to further significant improvements with respect to the overall energy losses.

The complexity of industrial production systems is steadily growing. Hence, the plant stuff has to search in an increasing number of documents within the daily work routine, e.g. in manuals, commissioning instructions, service notes, shift books, process data, repair instructions, data sheets, R/I flow charts, CAD drawings etc. To support the plant stuff, an intelligent search engine for industrial production environments is proposed in this paper. Characteristics of the developed search engine with respect to the domain of industrial production environments, e.g. tailored synonym replacements and document classifications, are outlined. Particularly, two methods for document classifications, a k-nearest-neighbor classifier and a Naive Bayes classifier, are evaluated with documents from industrial production environments.