Towards Automatic Model Completion from Requirements to SysML State Machines Maria Stella de Biase

2025-05-06 0 0 377.44KB 4 页 10玖币

侵权投诉

Towards Automatic Model Completion: from

Requirements to SysML State Machines

Maria Stella de Biase

Dip. di Matematica e Fisica

Universit`

a della Campania “L. Vanvitelli”

Caserta, Italy

mariastella.debiase@unicampania.it

Stefano Marrone

Dip. di Matematica e Fisica

Universit`

a della Campania “L. Vanvitelli”

Caserta, Italy

stefano.marrone@unicampania.it

Angelo Palladino

Aerospace Business Unit

Kineton srl

Napoli, Italy

angelo.palladino@kineton.it

Abstract—Even if model-driven techniques have been enabled

the centrality of the models in automated development processes,

the majority of the industrial settings does not embrace such a

paradigm due to the procedural complexity of managing model

life cycle. This paper proposes a semi-automatic approach for the

completion of high-level models of critical systems. The proposal

suggests a speciﬁcation guidelines that starts from a partial

SysML (Systems Modeling Language) model of a system and on

a set of requirements, expressed in the well-known Behaviour-

Driven Design paradigm. On the base of such requirements,

the approach enables the automatic generation of SysML state

machines fragments. Once completed, the approach also enables

the modeller to check the results improving the quality of

the model and avoiding errors both coming from the mis-

interpretation of the tool and from the modeller himself/herself.

An example taken from the railway domain shows the approach.

Index Terms—Behavior Driven Development, Requirement En-

gineering, Natural Language Processing, SysML, Critical System

Design

I. INTRODUCTION

The increasing complexity of critical systems requires a

greater and greater level of software dependability while

market pressure pushes down products time-to-market. To this

aim, it is crucial to shorten software development periods

without sacriﬁcing quality. The best way to reduce fatal

errors is to take preventive measures. Since the deﬁnition of

requirements is the ﬁrst fundamental step in the development

process, according to the software engineering V-Model [3]

(mainly used in safety-critical systems development [11]),

it is necessary to sift through the clarity, correctness and

consistency of the requirements in order to eliminate certain

types of errors in advance.

In this research, we follow the well-known Model-Based

Development (MBD) [10], which explored different aspects

of the software development and of the above-mentioned V-

model. This notwithstanding, the MBD approaches mainly

focused in the past years on the “lower-level” activities (e.g.,

code generation, testing). On the other hand, industries do

not prefer big changes in their assessed processes due to the

necessity of managing multi-years ongoing projects; instead,

they prefer small improvements that easily adapt to existing

practices.

To this aim, this paper describes the ﬁrst idea of an ongoing

research whose primary aim is mixing the contribution of

both humans and automatic tools in a requirement engineering

process. The goal is to build a tool that, starting from a

partial SysML model of a system and from a set of textual

requirements, is able to complete the model by creating some

SysML model fragments to integrate in the global one. We

are aware that this goal is very ambitious and then there are

some working hypotheses for the research:

•The requirements are deﬁned according the well-known

Behavioural Driven Development (BDD) [2] principle

and using the Given-When-Then paradigm;

•The requirements are also structured according to guide-

lines considering different ways (requirement templates)

of writing a BDD requirement;

•SysML model is itself built according to a speciﬁc

guideline where components and their main states are

partially deﬁned.

Under these preliminary conditions, the ﬁrst goal is to

build an automatable process able to: (1) read requirements

and choose the most suitable model fragments for them; (2)

generate state machines’ transitions capturing the behaviour

described by the requirement; (3) annotate the requirement

with a set of SysML annotations to improve traceability and

to ease the modeller in verifying the ﬁnal result of the model.

The fulﬁllment of this concrete objective will be possible by

enhancing traditional writing processes with formal grammars

and using Artiﬁcial Intelligence (AI) and Natural Processing

Language (NLP) algorithms. SysML proﬁling methods will

also be followed, too.

The industrial impact of this research is high since one of

the most critical obstacles in the adoption of a full model-

driven process in industrial settings is constituted by the phys-

iological disalignment between models and other “runnable”

artefacts. In this way, the model would be at the center of the

system development process since its early phases; automation

would enable the model as a “living object” of the process,

and the traceability information enables explainability and

interpretability that are crucial for critical systems, mitigating

the traditional disinclination of international standards as ISO

26262, DO-178B, and DO-178C to AI and NLP adoption. One

arXiv:2210.03388v1 [cs.SE] 7 Oct 2022

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

10 玖币 0人已下载

立即下载

摘要：

TowardsAutomaticModelCompletion:fromRequirementstoSysMLStateMachinesMariaStelladeBiaseDip.diMatematicaeFisicaUniversitadellaCampaniaL.VanvitelliCaserta,Italymariastella.debiase@unicampania.itStefanoMarroneDip.diMatematicaeFisicaUniversitadellaCampaniaL.VanvitelliCaserta,Italystefano.marrone@un...

展开>> 收起<<

Towards Automatic Model Completion from Requirements to SysML State Machines Maria Stella de Biase.pdf

共4页,预览1页

还剩页未读，继续阅读

声明：本站为文档C2C交易模式，即用户上传的文档直接被用户下载，本站只是中间服务平台，本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间，仅对用户上传内容的表现方式做保护处理，对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私，请立即通知玖贝云文库，我们立即给予删除！

Towards Automatic Model Completion from Requirements to SysML State Machines Maria Stella de Biase

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

热门标签

举报选择: