A vision of experimental proceb improvement

Software systems that support Busineb Proceb Management are in wide-spread use. They play an important role in facilitating proceb automation and proceb improvement. Yet, there is hardly any insight into whether the implementation of a supposedly improved pro-ceb model leads to an actual improvement in the proceb. The research problem this paper addrebes is: how can we determine if a new variant of a proceb model is an improvement over a previous variant, with respect to relevant measures? To this end, we suggest to build on recent software engineering concepts from the DevOps movement, and to develop novel techniques that provide the infrastructure for abebing in how far a specific busineb proceb change manifests an improvement.