Decision-making support for input data in business processes according to former instances

Business Processes facilitate the execution of a set of activities to achieve the strategic plans of a company. During the execution of a business process model, several decisions can be made that frequently involve the values of the input data of certain activities. The decision regarding the value of these input data concerns not only the correct execution of the business process in terms of consistency, but also the compliance with the strategic plans of the company. Smart decision-support systems provide information by analyzing the process model and the business rules to be satisfied, but other elements, such as the previous temporal variation of the data during the former executed instances of similar processes, can also be employed to guide the input data decisions at instantiation time. Our proposal consists of learning the evolution patterns of the temporal variation of the data values in a process model extracted from previous process instances by ap plying Constraint Programming techniques. The knowledge obtained is applied in a Decision Support System (DSS) which helps in the maintenance of the alignment of the process execution with the organizational strategic plans, through a framework and a methodology. Finally, to present a proof of concept, the proposal has been applied to a complete case study.