Why feature dependencies challenge the requirements engineering of automotive systems: An empirical study

Vogelsanag, Andreas; Fuhrmann, Steffen

doi:10.1109/re.2013.6636728

Cited by 31 publications

(12 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this case, both functions are independent from each other, but can be used in coordination. In an automation plant with a higher degree of automation, for example, if the function "balance load" may automatically start or stop additional beach wells by activating the function "toggle beach well", it is necessary to add internal channels that model the communication between functions [11], [12].…”

Section: Coupling With the Executable Automation Software Behaviormentioning

confidence: 99%

Early Validation of Automation Plant Control Software using Simulation Based on Assumption Modeling and Validation Use Cases

Brandstetter

Froese

Tenbergen

et al. 2015

CSIMQ

Self Cite

View full text Add to dashboard Cite

Abstract. In automation plants, technical processes must be conducted in a way that products, substances, or services are produced reliably, with sufficient quality and with minimal strain on resources. A key driver in conducting these processes is the automation plant's control software, which controls the technical plant components and thereby affects the physical, chemical, and mechanical processes that take place in automation plants. To this end, the control software of an automation plant must adhere to strict process requirements arising from the technical processes, and from the physical plant design. Currently, the validation of the control software often starts late in the engineering process in many cases -once the automation plant is almost completely constructed. However, as widely acknowledged, the later the control software of the automation plant is validated, the higher the effort for correcting revealed defects is, which can lead to serious budget overruns and project delays. In this article we propose an approach that allows the early validation of automation control software against the technical plant processes and assumptions about the physical plant design by means of simulation. We demonstrate the application of our approach on the example of an actual plant project from the automation industry and present it's technical implementation.

show abstract

Section: Coupling With the Executable Automation Software Behaviormentioning

confidence: 99%

Early Validation of Automation Plant Control Software using Simulation Based on Assumption Modeling and Validation Use Cases

Brandstetter

Froese

Tenbergen

et al. 2015

CSIMQ

Self Cite

View full text Add to dashboard Cite

show abstract

“…While there is an important study of dependencies in real-world automotive systems [15], the FI problem in realworld automotive systems has just been touched yet. Static analyses for finding (structural) dependencies along the lines of [15] will be useful for locating potential (behavioral) FI, but for really finding unknown FI, much more work will have to be done.…”

Section: Background and Related Workmentioning

confidence: 99%

“…Static analyses for finding (structural) dependencies along the lines of [15] will be useful for locating potential (behavioral) FI, but for really finding unknown FI, much more work will have to be done.…”

Section: Background and Related Workmentioning

confidence: 99%

Investigating and Coordinating Safety-critical Feature Interactions in Automotive Systems Using Simulation

Luckeneder

Rathmair

Kaindl

2017

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

Automotive systems are safety-critical cyber-physical systems. In particular, undesired feature interaction can lead to safety-critical behavior. In order to address this problem, we investigate physical feature interaction in this context using simulation (with more than one physical variable). This allows us to visualize both the behavior of features in isolation and their interaction. Our major result is a new insight about feature coordination. In such a cyberphysical context, it can be insufficient to coordinate as usual by giving one feature priority over another one. Instead, coordinating based on a physical variable involved in the feature interaction appears to be both necessary and sufficient. In summary, we present our investigation of safetycritical feature interactions and their coordination in automotive systems using simulation, and its results.

show abstract

“…Integration is performed by describing the required properties of vehicle features and mapping those requirements onto the properties of the component network. This enables continuous evaluation in early design stages which avoids problems like unknown feature interaction [8]. Based on that, we can perform functional integrity verification as well as the validation of required properties like Safety Integrity Levels (SIL) [9] and timing constraints.…”

Section: Contributionsmentioning

confidence: 99%

“…This validation is performed by evaluating each feature isolated from other features. A huge issue in automotive architectures though, as stated in [8] is the discovery of unintentional interactions between features.…”

Section: Discovery Of Feature Interactionmentioning

confidence: 99%

Component-based modeling and integration of automotive application architectures

Schorp

Sommer

2014

2014 IEEE International Electric Vehicle Conference (IEVC)

View full text Add to dashboard Cite

Abstract-The introduction of new propulsion technologies such as electric or hybrid drives imposes fundamental changes to the overall structure of the vehicle's electric and electronic system architecture. It also increases the need for cross-domain functionality, such as centralized energy management or the orchestration of mechanical braking and electric energy recuperation during deceleration. This leads to new challenges with respect to architecture development as interconnections between features are introduced that are not yet fully understood. The vehicle's system architecture evolves from towards a distributed multi-functional control system. Component oriented, model based approaches with multiple viewpoints have already proven being suitable in other domains to manage the dependencies between functionality by decomposing a system into a network of functional entities encapsulated in components.In this paper, we present a domain-specific component model to describe functional interdependencies as well as non-functional requirements needed to enable safe integration of software components in a centralized automotive ICT architecture. The model enables the composition of high-level functions and the definition of compatibility constraints. The approach is then applied to unveil feature interaction in a component architecture. This forms the foundation of a sound development and integration process for heavily interconnected functions. It also enables online product validation mechanisms to ensure functional integrity and safety as well as meeting of deployment constraints and timing requirements.

show abstract

Why feature dependencies challenge the requirements engineering of automotive systems: An empirical study

Cited by 31 publications

References 14 publications

Early Validation of Automation Plant Control Software using Simulation Based on Assumption Modeling and Validation Use Cases

Early Validation of Automation Plant Control Software using Simulation Based on Assumption Modeling and Validation Use Cases

Investigating and Coordinating Safety-critical Feature Interactions in Automotive Systems Using Simulation

Component-based modeling and integration of automotive application architectures

Contact Info

Product

Resources

About