Bayesian propensity score matching in automotive embedded software engineering
This provides a practical solution for automotive software engineers to conduct causal inference when ethical or practical constraints prevent randomized testing, though it is an incremental adaptation of an existing method to a specific domain.
The paper tackles the problem of evaluating new software in automotive embedded systems where randomized experiments are not feasible, by applying Bayesian propensity score matching to observational data, resulting in more balanced groups for causal inference even with a small treatment sample (N_t=38).
Randomised field experiments, such as A/B testing, have long been the gold standard for evaluating the value that new software brings to customers. However, running randomised field experiments is not always desired, possible or even ethical in the development of automotive embedded software. In the face of such restrictions, we propose the use of the Bayesian propensity score matching technique for causal inference of observational studies in the automotive domain. In this paper, we present a method based on the Bayesian propensity score matching framework, applied in the unique setting of automotive software engineering. This method is used to generate balanced control and treatment groups from an observational online evaluation and estimate causal treatment effects from the software changes, even with limited samples in the treatment group. We exemplify the method with a proof-of-concept in the automotive domain. In the example, we have a larger control ($N_c=1100$) fleet of cars using the current software and a small treatment fleet ($N_t=38$), in which we introduce a new software variant. We demonstrate a scenario that shipping of a new software to all users is restricted, as a result, a fully randomised experiment could not be conducted. Therefore, we utilised the Bayesian propensity score matching method with 14 observed covariates as inputs. The results show more balanced groups, suitable for estimating causal treatment effects from the collected observational data. We describe the method in detail and share our configuration. Furthermore, we discuss how can such a method be used for online evaluation of new software utilising small groups of samples.