Multi-level production process modeling language

https://doi.org/10.1016/j.cola.2021.101053Get rights and content

Abstract

The fourth industrial revolution introduces changes in traditional manufacturing systems and creates basis for a lot-size-one production. The complexity of production processes is significantly increased, alongside the need to enable efficient process simulation, execution, monitoring, real-time decision making and control. The main goal of our research is to define a methodological approach and a software solution in which the Model-Driven (MD) principles and Domain-Specific Modeling Languages (DSMLs) are used to create a framework for the formal description and automatic execution of production processes. In that way production process models are used as central artifacts to manage the production. In this paper, we analyze production process modeling domain and present a DSML which can be used to create production process models suitable for automatic generation of executable code. The generated code is used for automatic execution of production processes within a simulation or a shop floor. The language can be used to specify errors that may occur during the process execution and to specify error handling and corrective steps, too. The DSML is evaluated by different groups of users and the evaluation results are presented. Both the DSML and the accompanying modeling tool are still in the prototype phase, as they are created and evaluated in use cases covering just the assembly of goods. To enable wider application of the language and the tool, it is required to have additional use cases from different manufacturing domains.

Introduction

Advanced technologies in the form of smart resources and smart products are the basis for the fourth industrial revolution, as they enable changes in factories and production. Industry 4.0 introduces primarily IT-driven changes in existing production systems in order to enable production of individualized products while preserving all beneficial economic characteristics of mass production [1], [2].

Producing highly individualized products in traditional production facilities often requires multiple production lines, or in case of a single production line, stopping the production to allow reconfiguration of machines which causes additional costs. To enable a flexible, individualized, lot-size-one production that is economically viable, the production needs to be carried out without stopping a production line for machine reconfiguration [3]. Therefore, it is necessary to solve a problem of tedious machine adaptation to frequent production changes, which is common in the context of Industry 4.0. Additionally, there is a problem of frequent position changes and relocations of human workers in a factory [4]. Due to a decreasing number of workers and an increasing level of automation in factories, factory workers are required to perform multiple different tasks. Frequently changing worker’s tasks leads to increased production dynamics and requires fine coordination of workers in a factory, so their work can be optimized and production downtime avoided. As workers often switch between the tasks, fast knowledge transfer is required so they do not lose time when changing workplaces.

To enable the production of individualized products at a lower cost, a solution for production orchestration at a higher abstraction level can be utilized [5]. This solution would require a formal method to specify production processes and create process models suitable for automatic generation of instructions executed on smart resources. A smart resource represents a machine or a human worker able to receive generated instructions and execute them on materials and products.

In this context, it is possible to apply a Model-Driven (MD) approach in which a centralized representation of knowledge would exist in a form of production process models. Therefore, in our previous work [6], we proposed a novel MD approach for production process modeling and automatic production process execution. The MD approach aims to reduce the gap between individual customer’s needs and the ability to produce required products. The main goals of the proposed MD approach are:

  • enable easier adaptation of machines to dynamic changes of production processes;

  • improve coordination of human workers and machines in factories; and

  • enable automatic execution of production processes.

A formal specification of a production process is the crucial part of the proposed approach. Existing process modeling languages are not tailored to model production processes. Currently, production processes are specified using different models like Bill of Materials (BOM) [7], Flow Process Chart (FPC) [8] and Failure Mode and Effects Analysis (FMEA) [9]. As these models have different syntaxes and semantics, it is hard to extract semantics and reason about production details to enable automatic execution of production processes.

As there is a requirement for a formal method to specify production processes and integrate production details from different models, a formal language to specify production processes needs to be designed and implemented. We compared different process modeling languages in order to determine which production process concept can be modeled [10]. To the best of our knowledge, there is no unified formal language required for an automatic execution aimed at modeling all production process aspects. For that reason, we decided to create a new Domain-Specific Modeling Language (DSML) directed at production process modeling. Our MD approach, described in Section 2, would enable flexible manufacturing with the help of an Orchestrator software system that manages production processes by using a knowledge base and models created with the DSML. The Orchestrator software system is running on a cluster of industrial computers and it enables orchestration [11], [12], detection and configuration of new and existing smart resources [13].

In this paper, we present a DSML named Multi-level Production process modeling Language (MultiProLan), which is created based on conclusions from our previous research [6]. MultiProLan currently allows process engineers, quality engineers and plant managers to collaborate on specifying a production process by using a common language. In this paper, we denote process engineers, quality engineers and plant managers together as process designers. A process designer is a person in charge of transforming a valuable idea or an experiment into an industrial process in a way to fulfill not only originality, efficiency, quality and sustainability criteria, but to consider a large number of often contradictory constraints. A production process designer can specify a production process model independent of any particular production system. Such a production process model expressed by the modeling concepts of MultiProLan is denoted as a Master-Level (MasL) model. In addition to modeling concepts aimed at the specification of MasL models, MultiProLan also contains modeling concepts aimed at specifying process execution details. A MasL model enriched with the execution details, which are dependent on a specific production system that will execute the process, we denote as a Detail-Level (DetL) model of a production process. In our approach, a DetL model can be created manually, by a production process designer using the custom-built Modeling Tool or automatically, from a MasL model, by utilizing the Orchestrator software system. Multiple DetL models can be created based on the same MasL model. Each of them represents one possible way of executing the MasL process model within a different production system setup.

In that way, MultiProLan enables modeling of production processes suitable for automatic execution. It can be used in a flexible and orchestrated production to facilitate the lot-size-one production. The MultiProLan language is currently being used to model assembly production processes but can be extended to model different kinds of production processes. MultiProLan and its Modeling Tool are still in the prototype phase, as they are created and evaluated in use cases covering the assembly of goods. To enable wider application of the language and the tool, it is required to have additional use cases from different manufacturing domains. Thus, we plan to extend the language and the Modeling Tool with new modeling concepts and functionalities that will make them applicable in broader spectrum of manufacturing domains. Supported by MultiProLan, our MD approach should increase the degree of factory automation by enabling easier adaptation of machines to dynamic production changes and by increasing coordination of resources.

This paper is the extended version of our conference paper [14]. In addition to the original content, the following extensions are made: (i) the architecture of the system for automatic production orchestration, process execution and document generation is presented in more details; (ii) a discussion on how MultiProLan is used by different kinds of users is provided; (iii) analysis of the production process modeling domain is described; (iv) extensions of MultiProLan’s syntax which enable specification of error handling, sub-processes and unordered process steps are presented; (v) process model examples are modified to include the aforementioned extensions; and (vi) the results of the language evaluation are given and discussed.

Following this Introduction, this paper is structured as follows. An overview of the MD approach for modeling and automatic execution of production processes and the MultiProLan’s basic concepts is presented in Section 2. Usage of MultiProLan by different users is described in Section 3. Analysis of the production process modeling domain is presented in Section 4. Both abstract and concrete syntaxes of MultiProLan are described in Section 5. The evaluation of MultiProLan is discussed in Section 6. Related work that includes different modeling languages and approaches is summarized in Section 7. Conclusions and future work are presented in Section 8.

Section snippets

An overview of the MD approach for modeling and automatic execution of production processes

In the MD paradigm, models represent a central artifact at all stages of system development. Although models have been used extensively in the software development process for decades, often they were used just for documentation purposes. Problems of growing complexity, number of different platforms and interoperability in system engineering in general, and in software engineering in particular, have motivated a paradigm shift. A system developed by following the MD paradigm includes models

The usage of MultiProLan

MultiProLan is a production process modeling language used in the domain of hardware assembly. Currently, it cannot be used to model any kind of production processes, as it is specific for the assembly domain, but it can be easily extended to support the modeling in different domains in future. The language unifies different production process aspects and therefore, it also unifies the work from different user groups. At this point of the MultiProLan development, these user groups are process

Production process modeling domain analysis

Before creating the MultiProLan DSML, a domain of production process modeling has been analyzed. Domain knowledge is gathered from research papers, technical documentation and domain experts. Feature-Oriented Domain Analysis (FODA) is used as a domain analysis method [34].

This section is divided into two subsections. First, a brief overview of FODA modeling concepts is presented, after the FODA model of MultiProLan is described.

Abstract and concrete syntaxes of MultiProLan

In this section we present the abstract and concrete syntaxes of MultiProLan for modeling production processes suitable for automatic code generation and execution. During the design of MultiProLan, a few domain experts — process engineers were involved and provided us with different model examples specified as ASME FPCs, BOMs and textual descriptions. We iteratively developed the language using their help to identify domain concepts and validated the modeling concepts of our language.

We used

Evaluation of the MultiProLan tool

To systematically evaluate the tool, various types of users need to be included in the evaluation process to gather wider scope of feedback. According to Salman et al. [46] when applying a new technology or a new approach in a software engineering experiment, there is no significant difference whether students or professionals are involved. Thus, we included researchers and students from the academic community as well as process engineers and software developers from the industry in the

Related work

Production processes should be digitally supported in Industry 4.0 [53] so they can be integrated within a smart factory. Modeling production processes is very important in industrial informatics [54], but it is not enough to document processes and store them in a factory database. Production processes should be modeled to lead the production and process models should be ready for automatic production, but also not too complex for a human to comprehend. In this section, different production

Conclusion and future work

In this paper we presented the DSML — MultiProLan for modeling hardware production processes suitable for automatic execution. The goal of the language is to support the modeling of all production details required for automatic execution, but not to be too complex for a human to comprehend. To achieve this goal, two levels of detail are implemented so that production processes could be modeled in a generic way. By creating two levels of detail, production process models become independent from

CRediT authorship contribution statement

Marko Vještica: Conceptualization, Methodology, Software, Validation, Investigation, Writing – original draft. Vladimir Dimitrieski: Conceptualization, Methodology, Writing – review & editing, Supervision, Project administration. Milan Pisarić: Investigation, Resources, Project administration, Funding acquisition. Slavica Kordić: Formal analysis, Visualization. Sonja Ristić: Methodology, Formal analysis, Resources, Writing – review & editing. Ivan Luković: Formal analysis, Writing – review &

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

The research in this paper is supported by KEBA AG Linz, Austria; and the Ministry of Education, Science and Technological Development of Republic of Serbia [grant number 451-03-68/2020-14/200156].

We would like to thank Milica Todorović and Maksim Lalić for their help in preparing and analyzing the questionnaire, as well as all the evaluation participants for their time and useful feedback.

References (92)

  • DorofeevK. et al.

    Agile operational behavior for the control-level devices in plug & produce production environments

  • GoreckyD. et al.

    Human-machine-interaction in the industry 4.0 era

  • S.M. Fallah, S. Wolny, M. Wimmer, Towards model-integrated service-oriented manufacturing execution system, in: Proc....
  • VješticaM. et al.

    Towards a formal description and automatic execution of production processes

  • JiaoJ. et al.

    Generic bill-of-materials-and-operations for high-variety production management

    Concurr. Eng.

    (2000)
  • ASME Standard: Operation and Flow Process Charts

    (1947)
  • StamatisD.H.

    Failure Mode and Effect Analysis: FMEA from Theory To Execution

    (2003)
  • VješticaM. et al.

    Towards a formal specification of production processes suitable for automatic execution

    Open Comput. Sci.

    (2021)
  • LoskyllM. et al.

    Semantic service discovery and orchestration for manufacturing processes

  • KeddisN.

    Capability-Based System-Aware Planning and Scheduling of Workflows for Adaptable Manufacturing Systems

    (2016)
  • M. Pisarić, V. Dimitrieski, M. Babić, S. Veselinović, F. Dušić, Towards a plug-and-play architecture in industry 4.0,...
  • VješticaM. et al.

    The syntax of a multi-level production process modeling language

  • BrambillaM. et al.

    Model-Driven Software Engineering in Practice

    (2017)
  • VallespirB. et al.

    Enterprise modelling: from early languages to models transformation

    Int. J. Prod. Res.

    (2018)
  • ZacharewiczG. et al.

    Model driven interoperability for system engineering

    Modelling

    (2020)
  • MernikM. et al.

    When and how to develop domain-specific languages

    ACM Comput. Surv.

    (2005)
  • DimitrieskiV.

    Model-Driven Technical Space Integration Based on a Mapping Approach

    (2017)
  • QiQ. et al.

    Digital twin and big data towards smart manufacturing and industry 4.0: 360 degree comparison

    IEEE Access

    (2018)
  • PisarićM. et al.

    Towards a non-disruptive system for dynamic orchestration of the shop floor

  • QuT. et al.

    IoT-Based real-time production logistics synchronization system under smart cloud manufacturing

    Int. J. Adv. Manuf. Technol.

    (2016)
  • WitschM. et al.

    Towards a formal specification framework for manufacturing execution systems

    IEEE Trans. Ind. Inform.

    (2012)
  • TodorovićM. et al.

    An automatic generation of production documentation from MultiProLan models

  • S. Vaidya, P. Ambad, S. Bhosle, Industry 4.0 – A Glimpse, in: Procedia Manuf., Maharashtra, India, 2018: pp. 233–238....
  • J. Wan, H. Cai, K. Zhou, Industrie 4.0: Enabling technologies, in: Proc. 2015 Int. Conf. Intell. Comput. Internet...
  • FarooquiA. et al.

    Error handling within highly automated automotive industry: Current practice and research needs

  • GaoZ. et al.

    A survey of fault diagnosis and Fault-Tolerant techniques—Part I: Fault diagnosis with model-based and signal-based approaches

    IEEE Trans. Ind. Electron.

    (2015)
  • A. García-Domínguez, M. Marcos-Bárcena, I. Medina-Bulo, A Comparison of BPMN 2.0 with Other Notations for Manufacturing...
  • KangK.C. et al.

    Feature-Oriented Domain Analysis (FODA) Feasibility Study

    (1990)
  • CzarneckiK. et al.

    Formalizing cardinality-based feature models and their specialization

    Softw. Process Improv. Pract.

    (2005)
  • PiklJ. et al.

    Yet Another Feature Modeling Tool (YAFMT)

    (2019)
  • ZVEI, Details of the Asset Administration Shell. Part 1 - the Exchange of Information Between Partners in the Value Chain of Industrie 4.0 (Version 3.0RC01)

    (2020)
  • BrunoeT.D. et al.

    Integrated product-process modelling for platform-based co-development

    Int. J. Prod. Res.

    (2020)
  • SteinbergD. et al.

    EMF: Eclipse Modeling Framework

    (2008)
  • Eclipse sirius documentation

    (2020)
  • VujovićV. et al.

    Sirius: A rapid development of DSM graphical editor

  • I. Dejanovic, M. Tumbas, G. Milosavljevic, B. Perisic, Comparison of Textual and Visual Notations of DOMMLite...
  • View full text