Discrete-event simulation

From Wikipedia

Jump to: navigation, search

Contents

Methods

Problem definition

This section describes the problem to be solved by the model, including how the baseline model can be used to predict the system performance through alternative models.

Rationale for model choice

The model should imitate the conditions of the real world and it is applicable mainly when the execution of an experiment in the real world is expensive or impossible. The choice of model (e.g., system dynamics, discrete event, agent based) is described along with a brief explanation of what that model entails. See the paper From System Dynamics and Discrete Evento to Practical Agent Based Modeling:Reasons, Techniques, Tools as a reference[1].


System Dynamics


Image:Simulation_System_Dinamics.jpg



Discrete Event

Image:Simulation_Discrete_Event.jpg


Agent Based

Image:Simulation_Agent_Based.jpg


Model elements

Actor definition

This section describes each of the actors involved in the baseline and alternative models.

Task definition

This section describes each of the tasks involved in the baseline and alternative models.

Other elements

This section describes any other ancillary elements such as meetings and how they impact processes such as noise levels and/or delays.


Processes

Parallel activities

This section describes the role of parallel activities and how they can speed up the execution of the entire project.

Delays

This section describes delays related to activities being performed by, for example, less experienced actors.

Noise

This section describes how noise increases the number of time a process might be repeated


Baseline model

Data source used to build baseline model

This section specifies the data source used to build the baseline model.

Measures of model validation

This section determines how baseline model was compared to data source and which criteria were used to consider the baseline model adequately reproducing the system being modeled.

Model description

Graphic describing model along with any pertinent details regarding actors and processes.


Alternative models

Data source used to build alternative model

This section specifies the data source used to build the alternative models. As a general rule, multiple models should be built so that multiple scenarios can be compared to shed light on the original problem.

Measures of model validation

This section determines how alternative models were compared to the data source and which criteria were used to consider the alternative model as providing an adequate prediction.

Model description

Graphic describing model along with any pertinent details regarding actors and processes. In this section it is important to validate a system resulting from the interaction of multiple simultaneous variables. Evaluating the interaction across multiple variables is actually the very reason for the creation of such a model.


Model performance metrics

The simulation tool must supply the user with some kind of charts to measure the results.

Examples:

Summary Statistics

Image:Summary Statistics.jpg

Schedulle Growth

Image:Schedulle_Growth.jpg

Person Backlog

Image:Person Backlog.jpg

Software

List software packages and programming languages used in the modeling process.


Results

Baseline model

Graphics and tables describing model performance metrics at baseline.

The simulation tool must supply the user with some kind of charts to measure the results.

Examples

Work Breakdown

Image:Work Breakdown.jpg

Project Finance

Image:Finance.jpg

The user can execute some experiences and show the results as a table like this [2]

Alternative models

Graphics and tables comparing multiple alternative models, emphasizing in which situations each of the alternative models might be better than others.