Abstract The development and maintenance of video games present unique challenges that differentiate them from Classic Software Engineering (CSE) such as the increased difficulty in locating bugs within video games. This distinction has given rise to Game Software Engineering (GSE), a subfield that intersects software engineering and video games. Our work proposes a novel way for bug localization in video games by evolving simulations via an evolutionary algorithm, which helps to explore the large number of possible simulations. Simulations generate data (i.e., traces) from the behavior of non-player characters (NPCs). NPCs are not controlled by the player and are key components of video games. We hypothesize that such traces can be instrumental in locating bugs. Our approach automatically locates potential buggy model elements from traces. Furthermore, we propose a novel way of applying genetic operations to evolve simulations by selectively combining their components, rather than combining all components as a whole. We evaluate our approach in the commercial video game Kromaia, and the results indicate that evolving simulations using our novel component-specific genetic operations boosts bug localization. Specifically, our approach improved the F-measure for all bug categories over randomly combining all components, the baseline (which focuses on CSE and utilizes bug reports), and Random Search by 7.93%, 27.17%, and 46.34%, respectively. This work opens a new research direction for further exploration in bug localization within GSE and potentially in CSE as well. Moreover, it encourages other researchers to explore alternative genetic operations rather than selecting them by default.

Search-Based Software Engineering (SBSE) offers solutions that efficiently explore large complex problem spaces. To obtain more favorable solutions, human participation in the search process is needed. However, humans cannot handle the same number of solutions as an algorithm. We propose the first hybrid fitness function that combines human effort with human simulations. Human effort refers to human participation for providing evaluations of candidate solutions during the search process, whereas human simulations refer to recreations of a scenario in a specific situation for automatically obtaining the evaluation of candidate solutions. We also propose three variants for the hybrid fitness function that vary in the distribution of human effort in order to study whether the variants influence the performance in terms of solution quality. Specifically, we leverage our hybrid fitness function to locate bugs in software models for the video games of game software engineering. Video games are a fertile domain for these hybrid functions because simulated players are naturally developed as part of the video games (e.g., bots in First-Person Shooters). Our evaluation is at the scale of industrial settings with a commercial video game (Play Station 4 and Steam) and 29 professional video game developers. Hybridizing the fitness function outperforms the results of the best baseline by 33.46% in F-measure. A focus group confirms the acceptance of the hybrid fitness function. Hybridizing the fitness function significantly improves the bug localization process by reducing the amount of tedious manual work and by minimizing the number of bugs that go unnoticed. Furthermore, the variant that obtains the best results is a counter-intuitive result that was under the radar of the interactive SBSE community. These results can help not only video game developers to locate bugs, but they can also inspire SBSE researchers to bring hybrid fitness functions to other software engineering tasks.

Video games have characteristics that differentiate their development and maintenance from classic software development and maintenance. These differences have led to the coining of the term Game Software Engineering to name the emerging subfield that intersects Software Engineering and video games. One of these differences is that video game developers perceive more difficulties than other non-game developers when it comes to locating bugs. Our work proposes a novel way to locate bugs in video games by means of evolving simulations. As the baseline, we have chosen BLiMEA, which targets classic software engineering and uses bug reports and the defect localization principle to locate bugs. We also include Random Search as a sanity check in the evaluation. We evaluate the approaches in a commercial video game (Kromaia). The results for F-measure range from 46.80%. to 70.28% for five types of bugs. Our approach improved the results of the baseline by 20.29% in F-measure. To the best of our knowledge, this is the first approach that is designed specifically for bug localization in video games. A focus group with professional video game developers has confirmed the acceptance of our approach. Our approach opens a new research direction for bug localization for both game software engineering and possibly classic software engineering.