piwik-script

Intern
    Chair of Computer Science II - Software Engineering

    QPME Features

    If you have a suggestion for how to improve or enhance QPME in directions not covered by the features below, please file a feature request in our feature request tracker.

    Currently Supported Features (v2.00)

    Model Analysis Methods

    • Discrete Event Simulation (SimQPN)
      • Method of non-overlapping batch means
      • Method of independent replications (replication/deletion approach)
      • Method of Welch for estimating the length of the initial transient
    • Support for simulating hierarchical QPNs (HQPNs)

    Reported Metrics

    • Token throughput (arrival and departure rates)
    • Minimum/maximum number of tokens
    • Average number of tokens
    • Token color occupancy
    • Queue utilization
    • Minimum/maximum observed token residence time
    • Mean and standard deviation of observed token residence times
    • Confidence interval for the steady state mean token residence time
    • Custom probes for observing token residence times across multiple places
    • Histogram of observed token residence times
    • Interface to the statistical tool R for further statistical analysis

    Results Visualization

    • Tabular data
    • Bar charts
    • Pie charts
    • Histograms

    Supported Queueing Disciplines

    • First-Come-First-Served (FCFS)
    • Processor-Sharing (PS)
    • Infinite Server (IS)

    Supported Service-time Distributions for Queues

    • Beta
    • BreitWigner
    • BreitWignerMeanSquare
    • ChiSquare
    • Gamma
    • Hyperbolic
    • Exponential
    • ExponentialPower
    • Logarithmic
    • Normal
    • StudentT
    • Uniform
    • VonMises
    • Empirical
    • Deterministic

    Supported Extensions to Standard QPNs

    • Global QPN-wide list of token colors
    • Distinction of firing weights at the transition level from the firing mode level
    • First-In-First-Out (FIFO) token departure discipline for places
    • Possibility to share queues among multiple queueing places

    Features Under Development

    Model Analysis Methods

    • Further techniques for simulation output data analysis
      • Spectral analysis
      • Standardized time series
    • Further variance reduction techniques
    • Further methods for estimating the length of the initial transient (warm-up period)

    Planned Future Extensions

    Model Expressiveness

    • Load-dependent service times (resource demands)
    • Further scheduling strategies for queues (e.g., GPS, Priority)
    • Transition priorities and inhibitor arcs
    • Timed transitions and immediate queueing places

    Model Analysis Methods

    • Support for parallel/distributed simulation
    • Support for analytical solution techniques
      • Structured analysis techniques (e.g., structured SOR, structured JOR)
      • Product-form solution techniques
      • Approximation techniques
    • Support for tolerance limits

    Change Logs