Now I've goofed up. Anish, I MEANT to say that I have NOT heard of an algorithm that fits your needs. Sorry.
It appears from your description Anish, that the proper distance between any two scheduled time slots should be STDistance = Calender Depth / Number of Time Slots required by the queue. This does nothing to address which queue should go into the first/second/any Calender Table slot, however. Anish's description leaves us in the dark on that point - which appears important but may not be so. If only the distance is important, then the STDistance should be calculated for each queue, and then the problem can be looked at two ways: 1) As a "bin packing or knapsack" problem, where the queue with the largest number of slots is assigned ideal distances in the schedule table, followed by the assignment of the next largest queue to their assignments, etc. This has been shown to be no more than 78% correct, (and generally down into the 60's), however. It's very fast though. 2) All combinations could be examined if the number of possible combinations is not too large. From that examination, the best schedule table combination could be found. So first, we calculate the ideal STD, using the above formula for each que. Then either a) begin scheduling the largest, first. Working down to the smallest queue's or b) begin examing which combination is really the best, although it may take a while, depending on the total # of combinations that must be generated, and scored. For scoring, I think we should take the total # of scheduled time slots that are schedulaed correctly and divide that by the total # of scheduled time slots, (a total number). So in baseball terms, a correct time assignment equals a hit, and the total time slots represents a batter's at-bat attempts. Any other idea's? Your thoughts, Anish? Adak --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Algorithm Geeks" group. To post to this group, send email to algogeeks@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/algogeeks -~----------~----~----~----~------~----~------~--~---
