459931 Process Systems Lifecycle Management Using a Model Based Engineering Approach

Monday, November 14, 2016
Grand Ballroom B (Hilton San Francisco Union Square)
Manuel Rodriguez Hernandez1, Ismael Diaz2, Carlos Hernandez Corbato3, Ricardo Sanz Bravo3 and Julia Bermejo3, (1)Chemical Engineering, Technical University of Madrid (UPM), Madrid, Spain, (2)Chemical Engineering, Technical University of Madrid, Madrid, Spain, (3)Technical University of Madrid, Madrid, Spain

The motivation of this work is the constant evolution in industry. Nowadays we are in what is called the fourth industrial revolution (called Industry 4.0 or Connected Enterprise or Industrial Internet depending on the country or consortium). This revolution is being fostered in many countries to get a more competitive industry. Industry 4.0 target is to make more efficient and flexible plants, reduce times and costs of projects and products lifecycle.

This revolution tries to converge the physical and the virtual plant, this is achieved by means of developing a digital model of the plant before constructing it. Industrial Internet tackles the current challenges of the industry: increased functionality in products, more connectivity, high interdisciplinarity, design for X, more product/process complexity, very software intensive, digitalization, needs of simulation and visualization, compliance with more demanding regulations, etc. It is an integrated engineering approach were different software tools are interconnected during the whole lifetime of the project, reducing inconsistencies and developing times.

To fulfil the new requirements, traceability throughout the entire system lifecycle is needed, traceability from the initial requirements to the final physical product

The concept of Model Based Systems Engineering provides methods to guide the cross-disciplinary, virtual product development process and to achieve the required traceability. Model Based Systems Engineering (MBSE) is a multi-disciplinary engineering paradigm propagating the use of models instead of documents to support analysis, specification, design and verification of the system being developed. Using models instead of documents, a discipline-neutral view of the system specification is created. The resulting coherent system model helps to understand and to overview the complexity of the developed system.

System models are created by application of the Systems Modelling Language (SysML). This is a general-purpose graphical modelling language for specifying, analysing, designing, and verifying complex systems that may include hardware, software, information, personnel, procedures, and facilities. In particular, the language provides graphical representations with a semantic foundation for modelling system requirements, behaviour, structure, and parametrics, which is used to integrate with other engineering analysis models.

Under this framework models appear as a core component in every new development. Using a systems engineering methodology the developed model will be the one that guarantees the consistency and derives the different applications needed in every stage of the lifecycle, from simulation, to risk assessment or even documentation maintenance.

The objective of our work is to develop a model of a process plant using SysML. This model will follow a systems engineering approach, starting from the requirements and will cover the whole lifecycle of the project. This model will be the core of the project allowing for the different applications.

In this paper, we present first a Model Based Systems Engineering architecture for process systems and its relationship with the different stages of the lifecycle process. The core of the architecture is the neutral model of the system. This model is generated once the requirements are clearly stated. The generation of the general system model starts with the development of the functional model that refines the requirements to more specific functions that have to be fulfilled by the system. Once the functional model is built the behavioural and structural models are developed. These models reify the functions to physical equipment and to a topology and defines the behaviour of the components that achieve the functions. This general model has many parts that can be reusable and a model library is also constructed to store them. The final step is to build the specific models, these are models related to different views of the process and related with specific disciplines. The generic model is developed using SysML but as for specific models more efficient languages and tools exist it is necessary to do some model transformations from and back to SysML. Different applications are needed for the design and operation phases of the project (steady state simulation for process design, risk assessment tools, fault diagnosis, alarm management, etc.)

After the architecture has been introduced we create a SysML model of a process plant (the production of ethylbenzene) starting with the requirements (package diagram). This model includes the behavioural (block definition and activity diagrams) and structural (internal and block definition diagrams) aspects of the model. Once the model is created an automatic transformation (bidirectional) from it to some applications, where specific and more efficient environments for those purposes exist, is illustrated. The applications addressed in this work are:

-Process design. Transformation to a steady state model used in a process simulation environment, in this case Aspen Plus. This model is used for plant design calculating process operating conditions, equipment design and different production routes.

-Risk assessment. Transformation to a functional model suited to perform risk analysis (HAZOP like analysis). In this case the transformation is made to D-higraphs, a graphical functional language.

-Process optimization. Automatic generation of the optimization objective and constraints. In this case we do it using Matlab. The model used in the optimization will be the one generated for Aspen Plus that can be embedded in Matlab.

In summary, this work shows some of the potential of having models at design time. Models that are neutral and that can be used for different applications. Current industry trends look for models that generate virtual plants that can be “operated” and that allow to discover design mistakes in this early design phase. Thus providing, when constructed, more robust, consistent and efficient plants.


Eigner,M., Dickopf,T., Apostolov,H., Schaefer,P., Faißt,K.-G., Keßler,A.,2014. System lifecycle management: Initial approach for a sustainable product development process based on methods of model based systems engineering. In: Product Lifecycle Management for a Global Market. Springer, pp. 287–300.

Friedenthal, S., Moore, A., Steiner, R., 2014. A Practical Guide to SysML, 3rd Edition. Morgan Kaufmann.

Kagermann, H., Wahlster, W., 2013. Recommendations for implementing the strategic initiative INDUSTRIE 4.0, Final report. Tech. rep., Industrie 4.0 Working Group

Evans, P. C., Anninziata, M., 2012. Industrial internet: Pushing the boundaries of minds and machines. Tech. rep., General Electric

Paredis, C., 2012. Why model-based systems engineering? benefits and payoffs. In: 4th PLM Future. Tagung, Mannheim.

Rodriguez, M., de la Mata, J. L., 2012. Automating hazop studies using d-higraphs. CACE 45, 102 – 113.

Extended Abstract: File Not Uploaded