A digital twin improves our use of a cyber–physical system and understanding of its emerging behavior. To this effect, a digital twin is to be developed and configured and potentially also operated by domain experts, who rarely have a professional software engineering background and for whom easy access and support, e.g., in form of low-code platforms are missing. In this paper, we report on an integrated method for the model-driven engineering of low-code development platforms for digital twins that enables domain experts to create and operate digital twins for cyber–physical systems using the most appropriate modeling languages. The foundation of this method is (1) a code generation infrastructure for information systems combined with (2) an extensible base architecture for self-adaptive digital twins and (3) reusable language components for their configuration. Using this method, software engineers first configure the information system with the required modeling languages to generate the low-code development platform for digital twins before domain experts leverage the generated platform to create digital twins. This two-step method facilitates creating tailored low-code development platforms as well as creating and operating customized digital twins for a variety of applications. 

Generating customized low-code development platforms for digital twins

In dynamic production environments, bringing flexibility to production planning and control (PPC) against continuously changing requirements is key to fulfill production objectives. Such environments often involve various information systems that make the integration of knowledge on the current state of production more challenging. In this paper, we present a conceptual approach of Digital Shadows that allow for a holistic data view on PPC-specific tasks based on an ontology, serving as a knowledge base of semantic-enriched data and production constraints. An exemplary real-world application for PPC in injection molding demonstrates the benefits of the concept. • Stakeholders in PPC are overwhelmed with information from less interlinked IT system. • Digital Shadows are a promising approach to support decisions in PPC. • Concept of Digital Shadows for Production Planning and Control. • Decision support provided through purpose driven information generation. • Exemplary concept application for minimizing rejection rates in injection molding. 

Towards digital shadows for production planning and control in injection molding

The engineering of digital twins and their user interaction parts with explicated processes, namely process-aware digital twin cockpits (PADTCs), is challenging due to the complexity of the systems and the need for information from different disciplines within the engineering process. Therefore, it is interesting to investigate how to facilitate their engineering by using already existing data, namely event logs, and reducing the number of manual steps for their engineering. Current research lacks systematic, automated approaches to derive process-aware digital twin cockpits even though some helpful techniques already exist in the areas of process mining and software engineering. Within this paper, we present a low-code development approach that reduces the amount of hand-written code needed and uses process mining techniques to derive PADTCs. We describe what models could be derived from event log data, which generative steps are needed for the engineering of PADTCs, and how process mining could be incorporated into the resulting application. This process is evaluated using the MIMIC III dataset for the creation of a PADTC prototype for an automated hospital transportation system. This approach can be used for early prototyping of PADTCs as it needs no hand-written code in the first place, but it still allows for the iterative evolvement of the application. This empowers domain experts to create their PADTC prototypes. 

Process-aware digital twin cockpit synthesis from event logs

Cyber-Physical Production Systems (CPPSs) are envisioned as next-generation adaptive production systems combining modern production techniques with the latest information technology. A CPPS creates a complex environment between different domains (mechanical, electrical, software engineering), requiring multidisciplinary solutions to tackle growing complexity issues and reduce (maintenance) effort. Software plays an increasingly important role in assuring an effective and efficient operation of CPPSs. However, software engineering methods applied for CPPSs seem to lag behind modern software engineering methods, where tremendous progress has been made in the last years. We initiated the Software Engineering in Cyber-Physical Production Systems Workshop (SECPPS-WS) to analyze and overcome this gap. After two instances with mostly academic participants, we conducted a full-day workshop with nine industry representatives from eight companies that develop and maintain CPPSs. Each industry representative presented their current work and challenges. We collected these challenges and condensed a categorized list of challenges backed by industry statements and literature. This paper presents the resulting list and pointers to (partial) solutions to offer guidance for academia and identify promising research opportunities in this area.

Industry Voices on Software Engineering Challenges in Cyber-Physical Production Systems Engineering

For planning and designing production and work systems, a holistic approach is necessary that considers both levels of factory planning and workplace design. Currently, separate digital tools are mostly used for the design of factories and the detailed planning of work systems. That leads to workers being considered inadequately or too late in the planning process of production. The consequence can be a time-consuming and costly replanning to solve problems in existing production and work processes. Using the example of an assembly of washing machines, an iterative approach is presented for a combined digital planning on factory and workplace level. A holistic design of the assembly line is carried out using the ema Software Suite, consisting of the ema Plant Designer (emaPD) and ema Work Designer (emaWD). In the case study, emaPD is used to optimize production elements such as operating resources, layout, and logistics by considering the material flow, throughput times, and production costs. These results are applied for detailed planning and design at the workstation level with emaWD, which uses an algorithmic approach for self-initiated motion generation based on objective task descriptions. The generated simulations are examined and optimized based on production time estimation (MTM-UAS) and ergonomic risk assessments (EAWS, NIOSH, reach and vision analysis) as well as workers’ abilities (age, anthropometry). As a result, an efficient factory with an optimized material flow could be planned while minimizing the manufacturing costs and throughput times while complying with the space specifications and ergonomics. The takeover of ergonomically unfavorable processes by robots as hybrid workstations enables, among other things, an improvement in ergonomics. The digital planning approach of combined factory (emaPD) and workplace design (emaWD) also enable early, coordinated, efficient planning of economical and ergonomic production.

Digital production planning and human simulation of manual and hybrid work processes using the ema Software Suite

Smart manufacturing demands to process data in domain-specific real-time. Engineering models created for constructing, commissioning, planning, or simulating manufacturing systems can facilitate aggregating and abstracting the wealth of manufacturing data to faster processable data structures for more timely decision making. Current research lacks conceptual foundations for how data and engineering models can be exploited in an integrated way to achieve this. Such research demands expertise from different smart manufacturing domains to harmonize the notion space. We propose a conceptual model to describe digital shadows, data structures tailored to exploit models and data in smart manufacturing, through a metamodel and its notion space. This conceptual model was established through interdisciplinary research in the German excellence cluster “Internet of Production” and evaluated in various real-world manufacturing scenarios. This foundation for an understanding helps to manage complexity, automated analyses, and syntheses, and, ultimately, facilitates cross-domain collaboration.

A Conceptual Model for Digital Shadows in Industry and Its Application.

In order to develop and establish a functional, integrated system for automated layout design that can be applied in practice, certain fields of action still need to be identified and solved. For this purpose, a systematic literature review is used to identify fields of action on the way to automated factory layout design and to derive recommendations for action. As a result, the need for action was identified in the areas of multicriteria optimization, the layout evaluation and selection process, the existing implemented algorithms and in the integration of the human planner. To make qualitative target criteria machine-interpretable was identified as the most important next step.

Fields of action towards automated facility layout design and optimization in factory planning – A systematic literature review

On the way to automated layout planning numerous challenges must be faced. One of these challenges is the automated layout evaluation, which can be considered as the core of automated layout planning. The research field of factory planning differs from related research topics such as facility layout planning (FLP) from operations research (OR) by focusing for example on qualitative criteria in the evaluation. However, findings on how qualitative criteria can be interpreted in a quantifiable and measurable way are still rare. In this article, we address these questions with a focus on changeability, which can be regarded as the most important qualitative evaluation criterion in layout planning. For this purpose, a survey was conducted that aims to determine changeability from the understanding of industry experts and to define it consistently with the understanding from literature. Based on this understanding, performance metrics are derived to measure the layout changeability.

/pdf/uncovering-the-human-evaluation-of-changeability-for-512ajxc1tn.pdf

Uncovering the human evaluation of changeability for automated factory layout planning: an expert survey

Abstract Manufacturing companies are facing a turbulent market environment. Challenges for these companies lie in balancing efficient, economical action with maintaining agility, responsiveness and competitiveness. Thus, it is becoming even more important that as much potential as possible is leveraged as early as the planning phase of a factory. Automated layout design with focus on changeability could help to increase the agility and responsiveness for the factories. Approaches for automated design and optimization of layout variants are already prevalent. Ultimately, the major challenge remains in the automation of the layout evaluation taking qualitative criteria in consideration. Burggräf et al. provide a first approach for the quantification of the qualitative layout evaluation criterion changeability through specific metrics. Within the scope of this work those metrics are examined with regard to their reliability. Anomalies or deviations in the evaluation results shall be identified and suggestions for improvement will be proposed based on the obtained insights. 

/pdf/sensitivity-analysis-of-specific-metrics-for-automated-3nmyixjd.pdf

Sensitivity analysis of specific metrics for automated evaluation of layout changeability

Manufacturing companies are confronted with the challenge of adapting to ever-changing requirements of markets in order to remain competitive. Besides the rising number of product variants, increasingly frequent product changes require a continuous adaptation of assembly processes including its work instructions. Adaptive and highly connected agile assembly systems are designed to meet these challenges by enabling the interaction of various assets in assembly. A successful implementation of such Industry 4.0 (I4.0) solutions requires the development of a semantic oriented adaptive framework, which connects the physical with the virtual world. It enables interactive and situation-aware solutions such as Augmented Reality applications to adapt to worker capabilities and to improve worker satisfaction by providing information, based on individual experience, skills and personal preferences. A central part of the adaptive framework is the semantic representation of tangible and intangible assets through a Virtual Asset Representation containing all relevant asset information for adaptive assembly. This paper shows a three levels structure for adaptive assembly implementation, consisting of the adaptive framework level, the Virtual Asset Representation (VAR) ontology level and the use case level. The implementation of an adaptive assembly system is shown in the use case of a rear light assembly process of an electric vehicle in the context of the EU funded project A4BLUE. Based on the gained experiences a critical reflection on target fulfilment and user-friendliness of the VAR is given.

Viviane Hahn

Papers

A Conceptual Model for Digital Shadows in Industry and Its Application.

Fields of action towards automated facility layout design and optimization in factory planning – A systematic literature review

Uncovering the human evaluation of changeability for automated factory layout planning: an expert survey

Sensitivity analysis of specific metrics for automated evaluation of layout changeability

Virtual Asset Representation for enabling Adaptive Assembly at the Example of Electric Vehicle Production