The goal of our service science and engineering research is to advance IT-based services technology for service inno- vation, composition and delivery. In this project, we study the service accountability framework to detect, diagnose, and defuse the real problem in a service process. This is impor- tant when a service process has external service providers. We need an efficient mechanism to ensure that external ser- vices do provide an acceptable level of performance and sta- bility, in order to meet the overall quality of service (QoS). Our design adopts Bayesian Networks to look for the most likely causes in a service process and inspects those ser- vices individually to identify the problem. We implement the LLAMA accountability service bus, a dynamic and effi- cient service infrastructure to support the monitoring, anal- ysis, and reconfiguration of service processes. It includes accountability agents to continuously monitor services and an accountability authority to identify faulty ones.