Addressing the state explosion problem when visualizing off-nominal behaviors in a set of reactive requirements

Reactive systems with a large degree of human interaction can be vulnerable to off-nominal behaviors (ONBs) that arise from the human operator’s unpredictability. In prior research, we have addressed the ONB problems by developing an approach to translating a set of reactive requirements into the ru...

Full description

Saved in:
Bibliographic Details
Published in:Requirements engineering Vol. 24; no. 2; pp. 161 - 180
Main Authors: Aceituna, Daniel, Do, Hyunsook
Format: Journal Article
Language:English
Published: London Springer London 01-06-2019
Springer Nature B.V
Subjects:
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Reactive systems with a large degree of human interaction can be vulnerable to off-nominal behaviors (ONBs) that arise from the human operator’s unpredictability. In prior research, we have addressed the ONB problems by developing an approach to translating a set of reactive requirements into the rule-based causal component model (CCM). The CCM’s analysis involved expanding the CCM rules into a larger set of rules that encompass the system’s entire state space, displaying both nominal and off-nominal behaviors as transition paths. However, a major limitation of CCM is the potential for state explosion, which grows as a function of the system’s component states. In this paper, we introduce the causal scenario model (CSM), which uses the same rule-based approach, while addressing the state explosion problem associated with CCM. The CSM grows as a function of system components and provides a visually concise alternative to CCM, while still providing information useful in the exposing and addressing of ONBs during the requirements analysis phase. We introduce CSM and demonstrate the effectiveness of CSM, using a case study that would be more difficult to visualize using CCM, and most other state-based modeling techniques.
ISSN:0947-3602
1432-010X
DOI:10.1007/s00766-017-0281-y