Time-variant decision support systems


Intelligent decision systems in time-invariant situations must be able to manipulate time-variant data and functions. The authors present an algorithm for the following dynamic computational problem: given a set of continuous functions, keep track efficiently of their order in their domain. The algorithm has time complexity of O(nlog/sub 2/n+K min(log/sub 2/n, log/sub 2/K)), where n is the number of functions and K is the total number of intersections between the functions. The solution provides for a convenient object-oriented view of the time-variant priority queue. Contrary to the abstract data type consisting of the array (data structure) and the associated utilities (push, pop), the time-variant priority queue maintains a private state as the data change.< >
Published in: Conference Proceedings., IEEE International Conference on Systems, Man and Cybernetics
Date of Conference: 14-17 November 1989
Date Added to IEEE Xplore06 August 2002
DOI: 10.1109/ICSMC.1989.71290
Conference Location: Cambridge, MA, USA