Literature Review to Collect Conceptual Variables of Scenario Methods for Establishing a Conceptual Scenario Framework

TL;DR

This paper conducts a semi-systematic literature review to identify and conceptualize core variables of scenario methods, establishing a shared framework to improve understanding and comparison of scenario-based engineering approaches.

Contribution

It introduces a set of scenario variables (SVs) based on a conceptual meta-model, aiding in the systematic analysis and comparison of scenario methods across domains.

Findings

The conceptual model effectively supports understanding of scenario methods.

The scenario variables facilitate comparison between different scenario approaches.

The model is validated through analysis of five scenario methods in the ADS domain.

Abstract

Over recent decades, scenarios and scenario-based software/system engineering have been actively employed as essential tools to handle intricate problems, validate requirements, and support stakeholders' communication. However, despite the widespread use of scenarios, there have been several challenges for engineers to more willingly utilize scenario-based engineering approaches (i.e., scenario methods) in their projects. First, the term scenario has numerous published definitions, thus lacking in a well-established shared understanding of scenarios and scenario methods. Second, the conceptual basis for engineers developing or employing scenarios is missing. To establish shared understanding and to find common denominators of scenario methods, this study leverages well-defined metamodeling and conceptualization that systematically investigate the concepts under analysis and define core entities and their relations. By conducting a semi-systematic literature review, conceptual variables are collected and conceptualized as a conceptual meta-model. As a result, this study introduces scenario variables (SVs) that represent constructs/semantics of scenario descriptions, according to 4 levels of constructs of a scenario method. To evaluate the comprehensibility and applicability of the defined variables, we analyze five existing scenario methods and their instances in automated driving system (ADS) domains. The results showed that our conceptual model and its constituent scenario variables adequately support the understanding of a scenario method and provide a means for comparative analysis between different scenario methods.