Chapter 1 - What is Simulation
1.1Different Kinds of Simulations
Static and Dynamic
Static: Time doesn’t play a natural role E.X.: Buffon needle problem
Dynamic: On the other hand
Continuous and Discrete
Deterministic and Stochastic
Deterministic: Models that have no random input
Stochastic: at least some inputs’ are random
Chapter 2 - Fundamental Simulation Concepts
2.1 Goals of the Study
output performance measures
1)The average waiting time in the queue
2)The maximum waiting time in the queue
3)The total production
4)The time-average number of parts waiting in the queue
Letting Q ( t ) be the number of parts in the queue at any time instant t,
this time-average queue length is the total area under the Q ( t ) curve,
divided by the length of the run.
5)The utilization of the drill press
2.2 Analysis Options
2.2.1 Queueing Theory
the simplest and most popular object of queueing theory M/M/1
- the first M: Arrival process is Markovian
Interarrival times are independent and identically distributed “draws” from an exponential probability distribution - The second M: service-time distribution
and here it’s also exponential. - 1: ”indicates that there’s just a single server.
Better yet, most of our output performance measures can be expressed as simple formulas.
For instance, the average waiting time in the queue (expected from a long run) is
problems:
- estimates are not exact, and errors in the result
- assumption of interarrival and service time distribution may not be satisfied
- the formula is for long-run appearance
- just a first-cut approximation
2.3 Pieces of a Simulation Model
Chapter 3 Simulation with Arena
3.1 Basic
3.1.1 Process moduel
1、Resources Consumption:
2、Change the capacity of Resources:
3、Monitor the variable
3.1.2 Run Setup
Run Speed: Larger and faster
Number of Replications: # times of Replication ( day )
Replication Length: # times of time units ( hour )
3.2 Advance
3.2.1 Have more than one entities
e.g. We have two kinds of customers: men and women
3.2.2 ‘Decide’ Module
Can use the boolean expression to simplify:
Attention:
1. the last choice is NOT necessary (e.g. others)
2. choices must be complement to each other
3.2.3 Adding a plot
3.2.4 Adding a textbox
3.2.5 Separating Seize, Delay, and Release
the reason to separate:
some services require different types of resources at different stages.
e.g.
Stage1: resource1 +resource2, delay for NORM(1.83,0.35)
Stage2: just resource2, delay for NORM(1.83,0.35)
which will be:
Seize Delay 0: means seize R1R2
Delay: NORM(1.83,0.35), deal with stage 1
Delay 0 and Release: Release R1
Delay: NORM(1.83,0.35), deal with stage 2
Delay 0 and Release: Release R2
3.2.5 Traffic Intensity
3.2.6 Assigning Priorities to entities
1)Assign an Attribute to entities
2)change the Type in the Queue module and select the Attribute name
3.2.7 Assign and Record Module
target: we want to calculate the average circle time of all entities, not for each type.
1) Use the Assign module to set “Start Time” to each entity
2) Use the Record module to calculate the “Time Interval”
Time interval: calculates and records the difference between a specified attribute’s value and the current simulation time.
Time Between: tracks and records the time between entities entering the module.
3.2.8 Resource Schedules
Background:
Make the following improvements on the original basis:
1)“ComputerDesk”: use one of the three computers there to sign up for the service to be done in the “Service” module
subject to failure :exponential distribution with mean 8 hours
time to fix :exponential distribution with mean 30 minutes
2)Service module: two shifts on each day
(0-8am) only requires 1 server,
(8-16pm) needs 3 servers.
What we need to do:
1)set failure:
2)Connect the failure with resources
Resource panel
3)Set shifts in Service module
Schedule panel
two ways to enter the schedule:
(1) “Edit via Spreadsheet” and
(2) “Edit via Dialog.” To use the spreadsheet, simply click the button at the end of the row and then click on the spreadsheet window to define the schedule
4)Attach Schedule to Servers
Resource: Type: Base on Schedule