Keywords

1 Introduction

1.1 Rationale

Industry 4.0 (I4.0) embraces the fusion of the physical with the virtual world while facilitating the automation of industrial value creation [1]. An important building block of I4.0 is to establish interoperability between manufacturing assets in complex global value networks [2, 3]. The realization of interoperability usually requires industry standards and their integration throughout different hierarchy levels of manufacturing systems. For this purpose, the standardized Asset Administration Shell (AAS) comes into play as the Industrial Digital Twin in I4.0. The development of a concept for exchanging digital services between manufacturing assets by utilizing the AAS’s functionalities seems particularly relevant for the future application of the AAS in manufacturing systems. Manufacturing equipment must be enabled to exchange services with other equipment or products. This will allow to automatically and decentralized match the required manufacturing process steps to manufacture a certain product [4].

1.2 Research Approach

The paper addresses a deductive, qualitative research approach, which demonstrates the results of ongoing I4.0 research at the University of Southern Denmark (SDU). In Sect. 2, a brief overview of the state-of-the-art based on a semi-structured literature review in the context of the AAS is elaborated. In Sect. 3, a conceptual model for exchanging services between assets based on the AAS is presented. In Sect. 4, a proof-of-concept demonstrates how the data link between a 3D printer and an Autonomous Mobile Robot (AMR), to enable service exchange, can be accomplished by using the AASX Package Explorer software. In Sect. 5, the research is summarized and an outlook for future research activities is given.

2 State-of-the-Art

2.1 Asset Administration Shell (AAS)

The application of the AAS in manufacturing systems is proposed by different industry associations as the key to achieving interoperability in I4.0 manufacturing systems [5,6,7]. The EU-based Industrial Digital Twin Association and the US-based Digital Twin Consortium just decided to collaboratively develop the AAS technology in the future by standardizing requirements and through discussions for harmonizing the technological standard [8]. The AAS aims at storing and collecting relevant data and data streams, such as technical data, operational data, identification data, etc., over the whole life cycle of an asset [9]. The data is clustered according to specific sub-models of the AAS (Table 1). The sub-models are defined by standardized semantic properties and contain sub-model elements e.g., functions, properties, and processes, which eventually determine the digital representation of the asset [10]. Industry-driven initiatives are contributing to coining the AAS standard [11]. For example, an industry-driven research project investigated and demonstrated how AAS technology can be used to digitalize the nameplate of products [12]. Current academic projects explore how the AAS can be designed and integrated into manufacturing systems [13]. Figure 1 gives an idea for integrating AASs throughout different hierarchy levels of manufacturing systems.

Table 1. Example of sub-models of the AAS [14].
Full size table
Fig. 1.
figure 1

Integration of AASs on different hierarchy levels of manufacturing systems [1]

Full size image

2.2 Service Exchange Between Assets

Technically, the exchange of information between assets can be defined as a minimum requirement for a service exchange [15]. For example, an AAS on the system level might request the process status of a physical asset on the equipment level, which then will provide the status by a boolean signal based on classifications of the data under the given property. In an AAS context, a service exchange is executed as a service request by one AAS and the response from another AAS within the manufacturing system. This automatic service exchange is utilized by connecting AASs using unique semantic IDs. These IDs provide an unambiguous identifier (or address/asset ID) for an AAS integrated into a network [12] and allow the implementation of self-managed assets. A self-managed asset is defined as an asset with its own semantic asset ID [14]. A foundation for exchanging services between manufacturing assets has been laid out by [16].

2.3 Research Gap and Objective

The academic and industry responses to the integration of AASs throughout manufacturing systems and the realization of service exchange between AASs are only beginning to take shape, see for example [1, 13, 16, 17]. A concept of how to link a system of different AASs to exchange services has not been sufficiently addressed in current industrial and academic work. The authors aim to contribute with their research to this discussion by demonstrating an approach for the realization of service exchange between manufacturing equipment based on integrated AASs. However, the research presented by the authors focuses more on the practitioner’s perspective on AASs.

Thus, the developed concept is intended to inspire manufacturing companies to realize a service exchange between assets.

3 Concept

The concept is created by facilitating expert opinion. This means that the conceptual model is iteratively developed and evaluated until a consensus was reached among the authors. Further, the concept incorporates the current state-of-the-art in the field of AAS. It is based on the sub-model concept of the AAS [10, 14] as well as on the so-called Reference Models of I4.0 components [11]. A requirement for linking AASs is that every AAS and sub-model must consist of its unique semantic ID within the network [18]. The concrete connection of AASs can be realized by linking the underlying AAS sub-models e.g., on the component or equipment level of an asset, to a concrete sub-model on the system level. The conceptual model for linking the AAS to achieve a service exchange between self-managed assets is depicted in Fig. 2.

Fig. 2.
figure 2

Conceptual model for exchanging services using the sub-model reference

Full size image

4 Proof-of-Concept

4.1 Case

The proof-of-concept aims at verifying and validating the conceptual model from Sect. 3. It covers a case for service exchange between a 3D printer and an Autonomous Mobile Robot (AMR) in SDU’s Industry 4.0 lab. The two assets are selected due to their relevance for Industry 4.0 manufacturing systems: 3D printers and AMRs are commonly used assets and thus provide a suitable and transferable case for exemplarily demonstrating the service exchange. Also, both assets are easily accessible in SDU’s Industry 4.0 lab. Both assets, the 3D printer and the AMR represent self-managed assets with their AASs.

An Arduino sound sensor is used to collect the vibration data of the 3D printer (Fig. 3a). This data allows deriving the operational state of the printer e.g., start, printing, or idling state. Collecting the vibration data using a sound sensor is suitable for the proof of concept. However, with an industrialized 3D printer, the operational state can often be collected directly from the asset without attaching additional sensors. The Arduino sensor is collecting analog data input which is saved in a local database. Typically, this data would be stored in a cloud database, from where the AAS can collect the information. Small statistical data analysis was performed in JMP 15 to demonstrate that the collected data can be used to derive the different operational states of the 3D printer as shown in Fig. 3b.

Fig. 3.
figure 3

(a) Printer-sensor setup; (b) Measured vibration (sound) data stream over time.

Full size image

The proof-of-concept setup is coined by the development of the AAS in the AASX Package Explorer software (AASX-PE), which is an open-source C# based editor [19]. The pre-defined sub-models from Table 1 have been implemented in the AAS on the equipment level. The sub-models on the system level have been selected based on [17]. A local database is hosting each AAS separately. The real-time linkage between AASs has not been fully implemented since the purpose of the implementation is to establish a sufficient conceptual demonstration of the service exchange. The linkage between the AASs is made by linking the semantic IDs from the AAS or relevant sub-model. The semantic IDs used are Internationalized Resource Identifier (IRI) links, but they could also be implemented by IRDI, FragmentID, ShortID, or custom ID depending on the use cases [19]. The IRI is named “https://example.com” and “https://companyX(orY).com”, to illustrate that the IRI is a unique worldwide link. Hence, it fosters interoperability since it can be used across companies. In Fig. 4, a linkage using the sub-model reference, by following the model from Fig. 2, is demonstrated. The unique semantic ID of the “OperationalData” sub-model from the 3D printer (“FLSUN3Dprinter”) has been directly linked to the “Printing” sub-model of the manufacturing cell’s AAS. Also, the operational data of the AMR (OperationalDATAMIR100) has been linked to the sub-model “Transportation” of the manufacturing cell. As a result, a logic for service exchange can be implemented as an additional sub-model. For example, within the “MESConnection” sub-model on the manufacturing cell level, the “PredictiveMaintenance” sub-model could now utilize the operational data from the AMR and the 3D printer to predict maintenance activities. Another example could be the implementation of a logic for scheduling the transportation tasks of the AMR, based on the operational data of the 3D printer: the ARM can be called automatically by the 3D printer for picking up the finished product.

Fig. 4.
figure 4

Linking AASs on different manufacturing system levels using AASX-PE

Full size image

4.2 Discussion and Limitations

The proposed conceptual model for exchanging services between manufacturing assets seems to be suitable for the available AAS technology and its open-source development tools. The functionality for exchanging service requests, i.e., manufacturing tasks, has been verified. The implementation of the logic for the service exchange was not part of the proof-of-concept and will be subject to future research. Only limited conclusions can be drawn in terms of the efficacy of the concept and its suitability for industrial application under real manufacturing conditions. Realizing a service exchange with the AAS technology might also be a future emphasis of the Industrial Digital Twin Association or other industry stakeholders. These actors might decide to realize the service exchange by using a different concept. However, the authors believe that the presented conceptual model demonstrates an initial idea of how the exchange of services can be implemented with the current state of standardized AAS technology.

5 Summary and Outlook

The research presented a conceptual model as well as a proof-of-concept for exchanging digital services within a system of Industrial Digital Twins based on the AAS technology. The service exchange is realized by integrating unique semantic IDs of AASs from manufacturing equipment into the AAS on the manufacturing system level and by defining a logic for the service exchange within the AAS of the manufacturing system. A proof-of-concept validated the basic functionality of the concept for exchanging services between two manufacturing assets. Future research will investigate the logic of service exchange within manufacturing systems. For example, it will be explored how a job shop scheduling approach can be implemented based on the AAS and the described service exchange concept.