ADOPTING MBSE AND SYSML FOR DESIGN OF ELECTRIC VEHICLE
Kiran Jacob1, Archana B. S.2
1
Dassault Systemes India Pvt. Ltd
Kiran.Jacob@3ds.com
2
Dassault Systemes India Pvt. Ltd
ABS13@3ds.com
ABSTRACT
Automobiles have become an important part of today’s transportation needs. With the increase in consumer needs
and regulations, automobiles have seen huge spike in complexity involved in design, development and production of
vehicles. Major factors contributing to this complexity is the usage of software and electronics in design related to
powertrain, safety, ride quality, infotainment and environment constraints. In addition, with new concepts like
autonomous vehicles, connected cars, inter grated traffic management will only to the chaos existing complexity.
In order to manage this complexity, traditional approach of disintegrated process involving multiple disconnected
tools for product development will not be sufficient to manage this complexity. Disintegrated process, tools and lack
of collaboration between development teams makes it difficult to automobile manufactures to identify mistakes
early in development cycle and by the time issues are caught it might be too late to rectify. So having proper MBSE
process involving SysML becomes very critical for developing precision products.
Information provided as part of storyline depicted in this paper, illustrates how a traditional workflow involving
product development can be improved with introduction of MBSE and SysML for design of electric vehicle. The
paper also introduces a common MBSE methodology used as part of product design called Cyber MagicGrid
methodology to develop an electric vehicle.
INTRODUCTION
It is common notion to get confused with SE for MBSE or vise versa. However there is difference between the two.
To put it in nutshell, MBSE was an answer to the problem faced in executing traditional document based SE
process. Traditional documents based SE process involved usage of documents for capturing information such as
specifications, architecture, design, interfaces, test results, etc. Since the information is contained in documents,
issues related to synchronization, maintenance and accessibility existed with it. Also concepts like traceability and
impact change analysis was cease to exist [1].
Model Based Systems Engineering (MBSE) is essential to avoid unexpected system failures during critical phases of
product development cycle like validation or utilization. One of the most popular graphical programing language
which implements MBSE perfectly is the Systems Modeling Language (SysML).
MBSE is a disciplined, rigorous approach to systems engineering that uses visual modeling and simulation to
support requirements, analysis, and design, through conceptual, logical, and later phases of the system life cycle
[2]. There are multitude of modeling techniques and approaches that fall within the preview of MBSE and include
techniques like structural analysis and design, data flow modeling, state transition design, FEM, CAD, behavior
modeling (Modelica), process modeling, etc. [2]. One of the most popular MBSE practices is the graphical
programing language called SysML for developing complex product design.
SysML became popular with aerospace and defense establishments to develop mission critical or complex
technology driven solutions. System model developed using SysML covers all the key aspects of MBSE like
capturing system requirements, systems design and analysis, trade off studies, test planning and interface definition
[2]. Object Management Group's (OMG) SysML is a general purpose graphical modeling language for representing
systems that may include combination of hardware and equipment, software, data, people, facilities and natural
object [4].
MBSE BENEFITS
Apart from above mentioned benefits, MBSE do provide additional benefits. The below graph quantifies, the
advantage of usage of MBSE over traditional SE process for product development. The blue line in the graph
indicates investment for MBSE activities for a project and the red line indicates investment for traditional SE
process. From the graphs, one can interpret that though MBSE has initial spike in terms of investment due to
methodology adoption, training, etc. during initial life cycle phases like concept and preliminary design, however
has considerable gains towards the other phases compared to traditional SE process.
Figure: MBSE v/s Traditional SE [8]
THE CYBER MAGICGRID METHODOLOGY
The Cyber MagicGrid approach is based on a framework consisting of viewpoints and aspects organized in a grid
view. Each cell of the grid is an artifact to deliver in the systems engineering process. The method is based on
existing studies in the field and real-life findings in managing models for organizations from different systems
engineering domains.
The approach includes the definition of the problem, solution, and implementation domains in the system model. It
alignment with the processes defined by ISO/IEC/IEEE 15288 as follows: problem domain with the Stakeholder
Needs Development process, solution domain with the Architecture Definition process, and implementation domain
with the Design Definition process. Each domain is represented as a separate row of the MagicGrid framework. The
row that represents the problem domain splits into two, to convey that the problem domain should be defined by
looking at the system of interest (SoI) from the black-box and then from the white-box perspective. The row that
represents the solution domain splits into numerous inner rows to convey that the solution architecture of the system
can be specified in multiple levels of detail.
The row that represents the implementation domain doesn’t split and is not fully highlighted like the upper rows, to
convey that everything except the requirements specification in this domain is not a part of MBSE and therefore
appears outside the scope of the MagicGrid approach.
Each domain definition includes four different aspects of the system to be considered and captured in the model.
These aspects match the four pillars of the SysML, that is, requirements, behavior, structure, and parameters (also
known as parametrics). They are represented as columns of the matrix.
Figure: Cyber MagicGrid Methodology
ELECTRIC VEHICLE
Electric Vehicle (EV) is an automobile, which is driven by one or many electric motors, and these electric motors
are powered by electricity from batteries or fuel cells. EV’s powered by batteries are called Battery Electrical
Vehicles (BEV).
BEV is similar to traditional combustion-based vehicles in architecture, except for the fact that electric motor
replaces combustion engine and fuel tank by batteries. Controllers are still used in EV in order to control the current
supplied to motors from batteries as per load demands. Sometimes BEV’s contain combustion engine and together
with electrical generator to charge the batteries is called range extender. BEV’s also contain other systems like
DCDC converters, capacitors, cooler packs, air conditioning systems and steering systems, which act as load to
batteries. Below is the view of the architecture or topology used in BEV’s.
Figure: BEV - Topology
REFERENCE
[1] INCOSE, 2015, System Engineering Handbook, Chapter -9.2, pp- 189.
[2] INCOSE UK, 2015, Z9: What is Model Based Systems Engineering,
https://incoseuk.org/Normal_Files/Publications/zGuides, Retrieved May 2020.
[3] Sanford Friedenthal, Alan Moore, Rick Steiner 2008, OMG Systems Modeling Language Tutorial, pp-
6.
[4] Sanford Friedenthal, Alan Moore, Rick Steiner, 2015, A Practical Guide to SysML, Chapter -1, pp3.
[5] Wikipedia, SyML Partners, https://en.wikipedia.org/wiki/SysML_Partners, retrieved May 2020.
[6] Functional Mock-up Interface, FMI, https://fmi-standard.org/, May 2020.
[7] OSLC, https://open-services.net/, May 2020.
[8] Azad M. Madni and Shatad Purohit, February 2019, Economic Analysis of Model-Based Systems
Engineering, pp-13.
