Preferences in Temporal Relational Databases

Despite the huge amount of work devoted to the treatment of time within the relational context, some relevant phenomena remain to be fully addressed. We focus on one of them, i.e., temporal indeterminacy with preferences. In several domains (e.g., workflows, guidelines) and tasks (e.g., planning, scheduling), the exact time of occurrence of facts is not known: only an interval of possible values for their starting time, and a range of possible durations is available. Additionally, preferences can be assigned to the different temporal possibilities. We propose the first relational temporal database approach coping with such issues. We introduce a new data model to cope with indeterminate time with preferences, considering a family of preference functions, and we propose new definitions of relational algebraic operators to query the new data model. We also ascertain the properties of the new model and algebra, with emphasis on reducibility, and on the correctness of the algebraic operators.