An efficient and scalable method for aggregate nearest neighbor queries on time-dependent road networks

We study the k aggregate nearest neighbor queries on time-dependent road networks (TDRNs), assuming that the travel time to traverse an edge depends on the time it is initiated. Given a set of query points, a set of points of interest (POIs), an aggregate function (e.g., max) and a number k, the query returns k POIs p in increasing order of the values of the aggregate travel times from the query points to p. We investigate two forms of queries. The single departure time query requires all the query points to simultaneously start at a fixed departure time given as a query condition. In addition, we propose a more flexible form of query, the time interval query that allows the query points to independently depart at any arbitrary time in a given departure time interval.