Operating a complex and dynamic system, such as an agile manufacturing line, is a knowledge-intensive task. It imposes a steep learning curve on novice operators and prompts experienced operators to continuously discover new knowledge, share it, and retain it. In practice, training novices is resource-intensive, and the knowledge discovered by experts is not shared effectively. To tackle these challenges, we developed an AI-powered pervasive system that provides cognitive augmentation to users of complex systems. We present an AI cognitive assistant that provides on-the-job training to novices while acquiring and sharing (tacit) knowledge from experts. Cognitive support is provided as dialectic recommendations for standard work instructions, decision-making, training material, and knowledge acquisition. These recommendations are adjusted to the user and context to minimize interruption and maximize relevance. In this article, we describe how we implemented the cognitive assistant, how it interacts with users, its usage scenarios, and the challenges and opportunities.@en

Many industries face the challenge of capturing workers' knowledge to share it, particularly tacit knowledge. The operation of complex systems such as a manufacturing line is knowledge-intensive. Considering this knowledge's breadth and dynamic nature, existing knowledge-sharing solutions are inefficient and resource intensive. Conversational user interfaces are an efficient way to convey information that mimics how humans share knowledge; however, we know little about how to design them specifically for knowledge sharing, especially regarding tacit knowledge. In this work, we present an intelligent assistant that we have developed to support the elicitation of tacit knowledge from workers through systematic reflection. The system can interact with workers by voice or text and generate visualizations of shop floor data to support reflective prompts.

@en

Learning to operate a complex system, such as an agile production line, can be a daunting task. The high variability in products and frequent reconfigurations make it difficult to keep documentation up-to-date and share new knowledge amongst factory workers. We introduce CLAICA, a Continuously Learning AI Cognitive Assistant that supports workers in the aforementioned scenario. CLAICA learns from (experienced) workers, formalizes new knowledge, stores it in a knowledge base, along with contextual information, and shares it when relevant. We conducted a user study with 83 participants who performed eight knowledge exchange tasks with CLAICA, completed a survey, and provided qualitative feedback. Our results provide a deeper understanding of how prior training, context expertise, and interaction modality affect the user experience of cognitive assistants. We draw on our results to elicit design and evaluation guidelines for cognitive assistants that support knowledge exchange in fast-paced and demanding environments, such as an agile production line.

@en

Voice-enabled assistants, such as Alexa and Google Assistant, are among the fastest-growing information technology applications. Their technological foundation matured over the last years and reached a point where new application areas in challenging business environments become a certainty. Maintenance in manufacturing is one of these areas. This paper presents expectations, requirements, and a concept for a voice-enabled digital intelligent assistant that supports maintenance activities. We identified process monitoring, task execution, reporting, problem-solving, and maintenance planning as the key functional modules for an assistant. Realizing them depends on basic, utility, and maintenance functions. Our discussion states that all fundamental technologies and tools to realize an assistant for maintenance exist, but they have constraints. For instance, Speech-to-Text mechanisms lack transparent and performant solutions, and natural language understanding must rely on small datasets, which is challenging. We argue that continuous improvement and systematic evaluation of an assistant prototype is important to create high-quality training data. Trial-and-error is common because some technologies still mature, and conversation designers lack design patterns for the maintenance domain. Challenges for system adoption include providing an outstanding user experience, handling factory-specific jargon, and the limited availability of easy-to-use data exchange interfaces for machines and business applications. We conclude that further efforts on interoperability, technology stack management, AI-focused change management, and education programs are necessary. Furthermore, the accountability of AI systems is a cost factor for the assistant’s service providers and the client companies in manufacturing – AI insurance services, human-in-the-loop functions, user training, and professional education are actions to address this issue.@en

Cyber-physical systems (CPSs) have sophisticated control mechanisms that help achieve optimal system operations and services. These mechanisms, imply considering multiple signal inputs in parallel, to timely respond to varying working conditions. Despite the advantages that control mechanisms convey, they bring new challenges in terms of failure prevention. The compensatory action the control exerts cause a fault masking effect, hampering fault diagnosis. Likewise, the multiple information inputs CPSs have to process can affect the timely system response to faults. This article proposes a failure prognosis method, which combines time series-based forecasting methods with statistically based classification techniques in order to investigate system degradation and failure forming on system levels. This method utilizes a new approach based on the concept of the system operation mode (SOM) that offers a novel perspective for health management that allows monitoring the system behavior, through the frequency and duration of SOMs. Validation of this method was conducted by systematically injecting faults in a cyber-physical greenhouse testbed. The obtained results demonstrate that the degradation and fault forming process can be monitored by analyzing the changes of the frequency and duration of SOMs. These indicators made possible to estimate the time to failure caused by various failures in the conducted experiments.

@en

Malfunction or breakdown of certain mission critical systems (MCSs) may cause losses of life, damage the environments, and/or lead to high costs. Therefore, recognition of emerging failures and preventive maintenance are essential for reliable operation of MCSs. There is a practical approach for identifying and forecasting failures based on the indicators obtained from real life processes. We aim to develop means for performing active failure diagnosis and forecasting based on monitoring statistical changes of generic signal features in the specific operation modes of the system. In this paper, we present a new approach for identifying emerging failures based on their manifestations in system signals. Our approach benefits from the dynamic management of the system operation modes and from simultaneous processing and characterization of multiple heterogeneous signal sources. It improves the reliability of failure diagnosis and forecasting by investigating system performance in various operation modes, includes reasoning about failures and forming of failures using a failure indicator matrix which is composed of statistical deviation of signal characteristics between normal and failed operations, and implements a failure indicator concept that can be used as a plug and play failure diagnosis and failure forecasting feature of cyber-physical systems. We demonstrate that our method can automate failure diagnosis in the MCSs and lend the MCSs to the development of decision support systems for preventive maintenance.@en

Malfunction or breakdown of certain mission critical systems (MCSs) may cause losses of life, damage the environments, and/or lead to high costs. Therefore, recognition of emerging failures and preventive maintenance are essential for reliable operation of MCSs. There is a practical approach for identifying and forecasting failures based on the indicators obtained from real life processes. We aim to develop means for performing active failure diagnosis and forecasting based on monitoring statistical changes of generic signal features in the specific operation modes of the system. In this paper, we present a new approach for identifying emerging failures based on their manifestations in system signals. Our approach benefits from the dynamic management of the system operation modes and from simultaneous processing and characterization of multiple heterogeneous signal sources. It improves the reliability of failure diagnosis and forecasting by investigating system performance in various operation modes, includes reasoning about failures and forming of failures using a failure indicator matrix which is composed of statistical deviation of signal characteristics between normal and failed operations, and implements a failure indicator concept that can be used as a plug and play failure diagnosis and failure forecasting feature of cyber-physical systems. We demonstrate that our method can automate failure diagnosis in the MCSs and lend the MCSs to the development of decision support systems for preventive maintenance.

@en

Typically, emerging system failures have a strong impact on the performance of industrial systems as well as on the efficiency of their operational and servicing processes. Being aware of these, maintenance and repair researchers have developed multiple failure detection and diagnosis techniques that allow early recognition of system or component failures and maintaining continuous system operation in a cost-effective way. However, these techniques have many deficiencies in the case of self-tuning first generation cyber-physical systems (1G-CPSs). The reason is that these systems compensate for the effects of emerging system failures until their resources are exhausted, and the compensatory actions not only mask the failures, but also make their recognition difficult. Late recognition of failures is however in contrast with the principles of preventive maintenance. Therefore, the promotion research concentrated on the issue of recognizing and forecasting failures under dynamic and adaptive behavior of 1G-CPSs. CPSs are enabled to compensate for failure symptoms by changing their system operation modes (SOMs). It was also observed that transitions of SOMs reduce the reliability of a signal-based failure diagnosis. It was hypothesized that the frequency and the duration of the changes of the operational states of the 1G-CPS may be strong indicators of the failure emergence phenomenon and that investigation of SOMs facilitates early detection of failures. Therefore, the completed exploratory studies were aimed at exploring how the frequency and duration of transitions of SOMs can be brought into correlation with specific types of failures, and how they can be computed as measures of failure occurrence. The obtained results revealed that system failures tend to induce unusual system operation modes that can be used as basis for failure characterization, and even for failure forecasting. The empirical research made use of a cyber-physical greenhouse testbed to get experimental data and was completed by the development of computational model. A failure injection strategy was implemented in order to induce failure occurrence in a controlled manner. The proposed approach can be applied as a basis of forecasting system failures of 1G-CPSs, but additional research seems to be necessary.@en