Improving business process decision making based on past experience

Business processes entail a large number of decisions that affect their business performance. The criteria used in these decisions are not always formally specified and optimized. The paper develops a semi-automated approach that improves the business performance of processes by deriving decision criteria from the experience gained through past process executions. The premise that drives the approach is that it is possible to identify a process path that would yield best performance at a given context. The approach uses data mining techniques to identify the relationships between context, path decisions, and process outcomes, and derives decision rules from these relationships. It is evaluated using a simulation of a manufacturing process, whose results demonstrate the potential of improving the business performance through the rules generated by the approach.