Class InferQuickModel

java.lang.Object
jline.api.infer.InferQuickModel

public final class InferQuickModel extends Object
Convenience factory for the single-layer networks the inference estimators fit.

The model is named quickModel and has one Queue QueueStation<i> per entry of stations, with that discipline and servers[i] servers, and one class Class<c> per row of demands, served at station i by Exp.fitMean(demands[c][i]).

OPEN: a Source mySource and a Sink mySink, every class routed serially Source, QueueStation1, ..., QueueStationM, Sink. No arrival process is set: the estimators supply it. CLOSED: class c has jobs[c] jobs referenced at QueueStation1 and follows routing[c] (an M x M station-to-station matrix) when given, and the cycle 1, 2, ..., M, 1 otherwise.

Port of MATLAB infer_quick_model.m and the Python-native infer_quick_model.

  • Method Details

    • infer_quick_model

      public static Network infer_quick_model(boolean isOpen, SchedStrategy[] stations, double[][] demands)
      Serial routing, one server per station and one job per closed class.
    • infer_quick_model

      public static Network infer_quick_model(boolean isOpen, SchedStrategy[] stations, double[][] demands, int[] servers, int[] jobs, double[][][] routing)
      Parameters:
      isOpen - true for an open network, false for a closed one
      stations - (M) scheduling discipline of each queue
      demands - (K x M) mean service demand of class c at station i
      servers - (M) server count per station, or null for one each
      jobs - (K) population per closed class, or null for one each
      routing - (K x M x M) per-class station routing of a closed model, or null for serial
      Returns:
      the linked network