Large language models (LLMs) have gained significant attention for their success in natural language generation and their widespread adoption across various business sectors. In process automation (PA) engineering, LLMs can be utilized to assist with tasks such as PLC programming, software development, and data processing. Despite their advantages, LLMs face challenges like biased responses, non-deterministic outputs and model hallucinations. Therefore, it is essential to validate the responses generated by LLMs before relying on them. This work presents a novel workflow to automatically validate unstructured outputs generated by LLMs used in question answering (Q&A) applications based on the user-specified documents in the PA domain. The Domain of Validity (DoV) of these LLMs is first estimated based on the user-specified documents. The proposed workflow uses the estimated DoV to evaluate the relevance of the outputs generated by the LLM with respect to the user documents. A regular expression approach is included in the workflow to check whether the mentioned tag names in the LLM output, if any, exist in the user-specified documents. Furthermore, an LLM-as-a-judge approach is used then to cross-check the LLM generated outputs with the ground truth information. The proposed workflow enables users to automatically detect invalid LLM outputs. This allows the user to focus on cases that truly require human judgment and act accordingly by re-prompting the LLM, adding documents, or switching models. A fictitious PA example is employed to demonstrate the versatility and benefits of the developed workflow.

Asset engineering process in energy industry domain is a complex multi-disciplinary and multi-organizational process including an iterative exchange of a large number of artifacts. Legacy document-centric exchange process of those artifacts is an error-prone semi-manual process negatively impacting overall project duration and quality. We present results of a collaborative study between ABB and Equinor exploring the potential of the digitalization of this process based on Industry 4.0 technologies, especially the Asset Administration Shell. An exemplary engineering workflow including OPC UA and AutomationML models embedded in Asset Administration Shells was implemented along with novel Asset Administration Shell infrastructure. Despite minor flaws in maturing Industry 4.0 technologies, no major roadblocks were identified.

Automation Engineering of a process automation system is still a very manual effort due to limited support for the interpretation and processing of process design specification documents. Even though standards for digital data exchange between process and automation engineering do exist, those formats are rarely used and consequently the immense automation potential in automation engineering cannot be lifted. This contribution presents an AI -based approach and prototype - using an ontology-enhanced LLM -based agent and a mixture-of-experts system - to structure and formalize multimodal unstructured process design information as in PDF, Excel, and Word formats and make it available for state-of-the-art engineering tools for the long-known “Automation of Automation”.

Looking at Process and Automation Engineering (P&AE) today, there are many different tools available to support the engineer in his work from translating engineering intentions, via describing modules, to defining entire process plant setups, for export to a control system. As of today, the tools to support these steps are rarely linked to each other and hence do not allow to communicate with each other or exchange knowledge or data. Therefore, we came up with the idea of the Semantic Facilitation & Integration Layer, SemFIL, an ontology-based software system and architecture, to overcome tool, knowledge, and data silos, and to facilitate integration of and interoperability between applications, especially in P&AE. In this work, we present and evaluate the concept of SemFIL, and how it allows to connect and integrate all P&AE tools into a powerful, comprehensive, integrated workflow. We conclude with a discussion of the proposed SemFIL, and list required efforts for as well as advantages from its implementation. Lastly, we give an outlook on the most relevant future research questions and directions.

Process plant operators use graphical user interfaces to supervise complex processes with tens of thousands of instruments. Such systems may require more than 1000 operator screens, but an operator can only view 3-4 screens simultaneously ("keyhole-effect"), thus prolonging root cause analysis and emergency reactions. This effect contributes to the billions of USD lost in the process industry every year due to unscheduled downtimes. Researchers have proposed alternative, zoomable 2D user interfaces (“HawkEye HMIs”) similar to Google Maps, to address the keyhole-effect, but manual efforts to create these HMIs proved to be cost-prohibitive. This paper proposes a novel method for HawkEye-HMI-Generation based on semi-automated page layouting and automated level-of-detail zooming, which can significantly lower graphics engineering costs. We tested a prototypical implementation on two cases from large production plants with up to 90 process graphics and more than 5000 graphical nodes, demonstrating high scalability, good usability, and a more than 70 percent cost savings potential.

The modularization of process plants and their modular automation are innovative concepts to meet increasing demands for flexibility in plant operation. However, previous approaches to support the engineering of such modular plants do not focus on the early phases of requirements elicitation and concept design, although these early phases are crucial for the economic efficiency of the later plant. This paper presents a method for the formalized description of intentions in the early engineering phases of modular process plants. According to this method, plant operators create a formalized model of their intentions and provide it to the module manufacturers in a modularized form. This allows a reduction of effort in the planning and development of the plant. The approach is demonstrated using an oil, gas and water separation process performed by a modular process plant.

Modular industrial process plants build a production system by integrating a set of predesigned modules supplied by different vendors. The plant process engineers do not have access to the internals of these predesigned modules. Even when a predesigned module is well-tested, the composition of a set of modules can be vulnerable to many unforeseen scenarios, resulting in system-level failures. Debugging the cause of failure becomes extremely difficult since the implementation details are not available to the engineers. This paper proposes a technique by which it models the abstract functionality of each module as a state machine. We model the plant execution as a set of communicating state machines, making it possible to perform error propagation analysis and generate a set of test cases during the engineering phase.

Modular plants build a production system by integrating a set of pre-designed modules. Integration of these modules, supplied by different vendors, is performed using various design tools during the engineering phase. The integration process must perform a rigorous validation of the correctness of a module specification (MTP). Otherwise, the integration process can fail without providing enough failure details. Consequently, Such a failure at the later stage can significantly impact implementation, testing, integration, and SAT. In this paper, we describe a validator tool, that allows a plant designer to define a set of invariants that must be satisfied so that an MTP can be deemed fit for integration. We have tested the validator on a set of MTPs and reported our findings. We expect that the use of such a validator can significantly reduce the possibility of introducing errors during the engineering phase.

Modular automation is an upcoming paradigm to improve flexibility and maintainability of industrial production systems. The lack of interoperability between equipment adhering to different modular automation specifications limits module sourcing and hampers coexistence of discrete and process modules in joint hybrid production applications. In this work, we present a reference software architecture for modular automation which is able to cover existing specifications and demonstrate how it can be used to assess alternatives of cross-standard orchestration. Based on this foundation, next steps regarding cross-module and cross-orchestrator communication are outlined.

The modularization of process plants represents an innovative approach to meet the increasing requirements regarding flexibility. The current research focuses on 100%modular process plants, which are characterized by high flexibility of the plant, but unlike monolithic process plants, they are not designed for a specific and optimized operating point. In addition, the flexibility provided by the modularity is not needed everywhere in the plant. Therefore, hybrid process plants, i.e. a combination of a monolithic plant and several modules, are expected to overcome this issue. This paper presents the key terms in the context of hybrid process plants as well as the requirements for the engineering and automation of such plants. Furthermore, a generic workflow model for the engineering of hybrid process plants is derived, with focus on the conceptual design.

There is growing interest in building and operating industrial plants from standardized modules described by MTP's (Module Type Package). But this new approach brings with it its own set of new challenges. While in the traditional approach of building an industrial plant, one would deal with individual sensors and actuators, this new approach requires, amongst other things, the management of large pools of modules in the engineering and operation phases of a plant. In this contribution, we analyze these new requirements and report on our approach on building a demonstrator to deal with these new issues.

Autonomy is today a widely discussed topic. Despite the already impressive technical possibilities, autonomy will still happen stepwise in industry. When it comes to paving the way toward autonomy in industry, it is important to define an appropriate taxonomy of autonomy levels from an industry perspective. Based on such a basic taxonomy, this paper suggests a process how to apply it to specific industrial systems. Exemplarily, an application to autonomous plants is sketched. Besides, enabling technologies required toward autonomy are discussed, as well as some implications of autonomy in industry.

Modularization of process plants is seen as one approach to face the increasing flexibility requirements in the process industry. Therefore, modules are combined to a modular process plant to achieve a higher flexibility of the production. Conventional distributed control systems do not support flexible production systems adequately. Hence, modules should encapsulate process functions as services. The services should be invoked and controlled by state-based control. This contribution presents an approach for the state-based control of services, following a defined state-model. The approach has been tested in two case studies and two different implementations: In a simulation environment based on Matlab Simulink Stateflow, and in an industrial implementation of a pump and a controller. This contribution provides an overview of the developed concepts and their capability for the use in industrial environments.

Actual trends, such as highly volatile markets and shorter product lifecycles, in the process industries require new approaches for the control of process plants. These changes are tackled by the approach of modular process plants. Modular process plants promise a shorter time-to-market and a high flexibility regarding the products produced herein. To achieve these advantages, beside the mechanical modularization also the control systems of such plants need to change. Current process control systems are not able to fulfill all requirements of modularization. Therefore, known approaches from information technology are adapted for the automation of modular process plants. Beside service-oriented architectures, the Microservice approach is of current interest in the information technology domain. This approach is presented in this contribution, and the suitability of Microservices as an architecture for modular process automation is analyzed. Furthermore, a Microservice-based approach for the control of modular process plants is presented.

This paper introduces an algorithm to construct a plant topology from analyzing correlations in operations data and iteratively combining pieces of information to the final result. Selected domain specific rules are given in detail and applied to a feeder unit example of a continuous process plant. The final result is good in accordance with the initial example P&ID. Benefits of the method and further research fields are discussed.

Most Cyber Physical Production Systems (CPPS) research is devoted to the manufacturing industry, but the need for flexibility is an increasing trend in the process industry, too. Today's process automation systems (PAS) are not well-suited for quick adaptation of the plant, i.e. the addition, removal or change of physical components, such as filters, mixers, or reaction units. The authors have developed a model-based approach, which allows reconfiguring the PAS based on a model, which acts as a representative of the physical component. Thus, the model and the physical component together form a CPPS. In the paper, the requirements, the model structure, and the model content will be described, and a practical example will be presented.

Evolution in global markets for chemical and pharmaceutical goods requires new approaches in process plant design. Modularization is commonly considered as a promising approach to face the challenges of changing production environment. In order to exploit the advantages of process modularization, new structures of communication and control are required for modular process automation. Service-oriented architecture has been introduced as a promising communication infrastructure combined with state-based control of services. This paper introduces requirements of modular process automation and general requirements of service-oriented architectures. Process functions are represented by encapsulated services in the control level of the automation pyramid. Interaction between a superior control system and intelligent modules is described. Furthermore, approaches of service-oriented architecture are introduced such as a reference model as well as specific implementations of services within modular automation. Subsequently, four different service-oriented architecture approaches are analyzed with respect to their fulfilment of the presented requirements. Finally, this paper discusses the result of the analysis and the general applicability of service-oriented architectures for modular process automation.

The design of conventional process plants is guided by the idea of structuring the process in functional units. Because of cost reduction this design is often neglected during the implementation phase. There is a huge potential in the re-modularization looking at the re-usability which directly effects the engineering efficiency when developing new plants. Furthermore, there is the possibility to create modularized KPIs. Those can be aggregated to plant-wide KPIs to evaluate the overall plant performance and status. In this paper the authors are presenting a concept for the detection of virtual modules in existing plant topologies. The plant topology is therefore transformed into a searchable graph. With the help of extended search algorithms and the definition of functional patterns it is now possible to identify all possible virtual module within a plant topology. Through a prototypical implementation, visualizing the identified modules in the graph, a proof of concept is shown.

This contribution presents a novel approach for the automatic generation of simulation models for existing plants. The aim is to focus on brown-field projects for which the automation system has to be renewed. Whereas most research approaches for simulation based control code testing focus on green-field projects, the authors developed a method that considers the type of planning, as well as engineering data, which is given in brown-field projects. In addition, a prototype has been implemented. The paper presents the validation results of a first field study on a North Sea oil production plant - one of the largest oil discoveries in the UK North Sea.

Effective use of integrated process and plant knowledge may significantly increase the accuracy and reliability of automated plant diagnostics. State-of-the-art diagnosis methods, however, do not exploit the whole richness of information found in process facilities mainly due to the difficulties entailed for its collection and timely retrieval. In an effort to contribute towards the exploitation of such knowledge, this paper presents concepts to classify, integrate, and facilitate the access to relevant data as a basis for automated plant diagnosis tasks. The proposed integration is based on the Formalized Process Description Guideline VDI/VDE 3682 and comprises four fundamental information sources, namely plant connectivity, plant dimension specific-, plant component specific-, and process specific-knowledge. The resulting model is described in the data format CAEX/AutomationML, which allows for seamless and effective information exchange among different diagnostic tools. Further concepts for data access and visualization are presented in this contribution.