Simulation Software DiscreteEvent System Simulation 5 th Edition

  • Slides: 47
Download presentation
Simulation Software Discrete-Event System Simulation 5 th Edition Chapter 4 1

Simulation Software Discrete-Event System Simulation 5 th Edition Chapter 4 1

World Views of Simulation Model n Event-Scheduling View As with our project 1 q

World Views of Simulation Model n Event-Scheduling View As with our project 1 q Focus on processing each event q n Process-interaction View model as a set of processes through which an entity “flows” q Life-cycle approach – time-sequenced list of events, activities, & delays q Common in simulation environments q 2

World Views of Simulation Model n Activity Scanning Approach Focus on activities & conditions

World Views of Simulation Model n Activity Scanning Approach Focus on activities & conditions that allow it to begin q At each clock advance, scan conditions to start any activity that can begin q Approach is simple, but scan is slow q New 3 -phase approach includes some event scheduling – somewhat more complex but more efficient q 3

Categories of Simulation Software n General Purpose Languages q n Simulation Languages q n

Categories of Simulation Software n General Purpose Languages q n Simulation Languages q n C, C++, Java GPSS, SIMAN, SLAM, SSF Simulation Environments q Enterprise Dynamics, Arena, SIMUL 8 4

Features of Simulation Languages Some focus on a single type of application n Built

Features of Simulation Languages Some focus on a single type of application n Built in features include n Statistics collection q Time management q Queue management q Event generation q 5

Features of Simulation Environments Some focus on one type of application n Icon based

Features of Simulation Environments Some focus on one type of application n Icon based n Analysis of I/O n Advanced Statistics n Optimization n Support for Experimentation n 6

History of Simulation Software (Nance 1995) n 1955 -60 n n n Period of

History of Simulation Software (Nance 1995) n 1955 -60 n n n Period of Search 1961 -65 Advent 1966 -70 Formative Period 1971 -78 Expansive Period 1979 -86 Period of Consolidation & Regeneration 1987 - 2008 Period of Integrated Environments 2009 + The Future 7

Simulation Languages 1981 – 137 Simulation languages reported n More have be developed since

Simulation Languages 1981 – 137 Simulation languages reported n More have be developed since n Now Simulation Environments n 8

The Search: : 1955 - 60 FORTRAN – one of a few languages n

The Search: : 1955 - 60 FORTRAN – one of a few languages n Focus on unifying concepts & reusable functions n General Simulation Program – first effort at “language” which as a set of functions n 9

The Advent: : 1961 -65 n GPSS – 1961 @ IBM q q q

The Advent: : 1961 -65 n GPSS – 1961 @ IBM q q q n Based on block diagrams Well-suited for queuing models Expensive at first SIMSCRIPT – 1963 – Rand Corp. q q q US Air Force – government is biggest user FORTRAN influence Owned by CACI in CA. 10

The Advent: : 1961 -65 (continued) n GASP – 1961 Based on Algol, then

The Advent: : 1961 -65 (continued) n GASP – 1961 Based on Algol, then Fortran q Collection of Fortran functions q n SIMULA – extension of Algol q n Widely used in Europe CSL (Control & Simulation Language) 11

Formative Period: : 1966 -70 Concepts caused major revisions of languages n Languages gained

Formative Period: : 1966 -70 Concepts caused major revisions of languages n Languages gained wider usage n GPSS (several variations) n Simscript II – English-like n ECSL – Europe n SIMULA – added classes & inheritance n 12

The Expansion Period: : 197178 GPSS/H – 1977 n GASP IV – 1974 –

The Expansion Period: : 197178 GPSS/H – 1977 n GASP IV – 1974 – Purdue n SIMULA n Attempt to simplify the modeling process q Program generators – severe limitations q 13

Consolidation & Regeneration: : n Movement to mini and PC computers 1979 -1986 n

Consolidation & Regeneration: : n Movement to mini and PC computers 1979 -1986 n SLAM II (descendant of GASP) q 3 world views q n Event, Network, Continuous SIMAN (descendant of GASP) q q q General Modeling + Block Diagrams 1 st first major language - PC & MS-DOS Fortran functions w/ Fortran programming 14

Integrated Environments: : 1987 - 2008 n Growth on PC’s n Simulation Environments GUI

Integrated Environments: : 1987 - 2008 n Growth on PC’s n Simulation Environments GUI q Animation q Data analyzers q 15

The Future : : 2009 - 2011 What can we expect in the future?

The Future : : 2009 - 2011 What can we expect in the future? (2008) n Virtual Reality n Improved Interfaces n Better Animation n Agent-based Modeling 16

Agent-Based Software Any. Logic n Ascape n MASON n Net. Logo n Star. Logo

Agent-Based Software Any. Logic n Ascape n MASON n Net. Logo n Star. Logo n Swarm n Re. Past n 17

Evaluating Software n Consider multiple issues q n Speed of execution q n Ease

Evaluating Software n Consider multiple issues q n Speed of execution q n Ease of use, support, applicability Experimental runs – Debugging Beware of demos & advertising Will focus on strengths only q Ask for demo of YOUR problem q 18

Evaluating Software Carefully consider comparison checklists with yes/no answers n Can software link to

Evaluating Software Carefully consider comparison checklists with yes/no answers n Can software link to external languages n Carefully consider trade-off between graphical model building & simulation programming language n Costs – one-time vs. licensing n 19

Simulation Software Features See the following tables in text: n Model-building features q n

Simulation Software Features See the following tables in text: n Model-building features q n Runtime Environment q n P. 123 – Table 4. 1 P. 124 – Table 4. 2 Animation & Layout features q P. 124 – Table 4. 3 20

Simulation Software Features n Output features q n P. 125 – Table 4. 4

Simulation Software Features n Output features q n P. 125 – Table 4. 4 Vendor Support - Documentation q P. 125 – Table 4. 5 21

Example Simulation Checkout Counter – Single Server Queue Consider at standard checkout counter environment

Example Simulation Checkout Counter – Single Server Queue Consider at standard checkout counter environment with on clerk and one queue. Interarrival times are exponentially distributed with mean 4. 5 minutes; service times normally distributed with mean 3. 2 and standard deviation 0. 6 minutes. Simulate for 1000 customers. 22

Java Model n Section 4. 4 – p. 126 n Note similarity to our

Java Model n Section 4. 4 – p. 126 n Note similarity to our process in project one 23

GPSS General Purpose Simulation System Highly Structured n Process Approach n Queuing Systems n

GPSS General Purpose Simulation System Highly Structured n Process Approach n Queuing Systems n Block Diagrams n 40 standard blocks q Block corresponds to a statement q n Transactions FLOW through the system 24

GPSS Block Diagram for Example Figure 4. 10 – p. 138 n Each entity

GPSS Block Diagram for Example Figure 4. 10 – p. 138 n Each entity has a name n q Name each queue, server, etc. In rectangle, parameters (as necessary) n Right attachment, name of entity n Far right column – GPSS Command n 25

GPSS Syntax Assembly-like Label Op. Code Subfields ; comment n Label: col. 1, <=

GPSS Syntax Assembly-like Label Op. Code Subfields ; comment n Label: col. 1, <= 9 alphanumeric, alpha start n Op. Code: 4+ characters of command n Subfields: as necessary, separated by commas n Comment: after ; or with * in column 1 26

GPSS Program Figure 4. 11 – p. 139 n Declaration Section n Customized vs.

GPSS Program Figure 4. 11 – p. 139 n Declaration Section n Customized vs. Standard Output n Code Section n 27

Generate Queue Seize Depart Advance Release Depart Test_GE Blet Terminate Start rvexpo (1, &IAT)

Generate Queue Seize Depart Advance Release Depart Test_GE Blet Terminate Start rvexpo (1, &IAT) Systime Line Checkout Line rvnorm(1, &mean, &stdev) Checkout Systime M 1, 4, Term &Count = &Count +1 1 &Limit 28

GPSS Output n Customized q n Figure 4. 12 – P. 141 Standard q

GPSS Output n Customized q n Figure 4. 12 – P. 141 Standard q Figure 4. 13 – P. 142 29

Other Simulation Software n SSF – Scalable Simulation Framework Application Program Interface (API) q

Other Simulation Software n SSF – Scalable Simulation Framework Application Program Interface (API) q Object-oriented, process view q 5 Base Classes q n Process, Entity, Event, In. Channel, Out. Channel Designed for high-performance computers q Bridges pure Java & simulation languages q Figures 4. 14 & 4. 15 q 30

Simulation Environments ~~ Common Features GUI n Animation n Automatic statistics n Output (tables,

Simulation Environments ~~ Common Features GUI n Animation n Automatic statistics n Output (tables, graphs, custom) n Analysis n Process world view n 31

Common Features (# 2) Some allow q Event Scheduling q Mixed continuous-discrete models n

Common Features (# 2) Some allow q Event Scheduling q Mixed continuous-discrete models n Animations – 2 D & 3 D n Business Graphics n 32

Simulation Environments n Any. Logic n Extend. Sim n Arena n Flexsim n Auto.

Simulation Environments n Any. Logic n Extend. Sim n Arena n Flexsim n Auto. Mod n Pro. Model n Enterprise n SIMUL 8 Dynamics 33

Any. Logic Supports: discrete event, agent-based, system dynamics (& combination) n Hybrid: discrete &

Any. Logic Supports: discrete event, agent-based, system dynamics (& combination) n Hybrid: discrete & continuous n Object library n Java models, publish as applets n Animation, Statistics, optimization, debugger n 34

Arena Discrete & Continuous systems n Object-based; GUI n 2 D, 3 D Animation

Arena Discrete & Continuous systems n Object-based; GUI n 2 D, 3 D Animation n Business & Manufacturing processes n Supports Analysis n Opt. Quest for optimization n Based on SIMAN; embedded Visual Basic n 35

Auto. Mod Manufacturing & Materials handling n Detailed large models for planning, decision support,

Auto. Mod Manufacturing & Materials handling n Detailed large models for planning, decision support, control systems n Auto. Stat - Experimentation & analysis n Auto. View - Make movies of 3 D animations n Full simulation language included n 36

Object oriented n Discrete Events n Open GL 3 D visualization engine n 4

Object oriented n Discrete Events n Open GL 3 D visualization engine n 4 D Script programming language n Interfaces with databases n Opt. Quest optimization n 37

Extend. Sim Block-diagram approach n Versions for mixed and for continuous only n Includes

Extend. Sim Block-diagram approach n Versions for mixed and for continuous only n Includes C-like programming language n Supports linking to external languages n 38

Flexsim Dynamic-flow systems - manufacturing n Discrete-event, Object-oriented simulator; developed in C++ using Open

Flexsim Dynamic-flow systems - manufacturing n Discrete-event, Object-oriented simulator; developed in C++ using Open GL n Animation: 2 D, 3 D, Virtual reality n Drag & Drop n 39

Pro. Model Manufacturing Systems n Simulation & Animation (2 D & 3 D) n

Pro. Model Manufacturing Systems n Simulation & Animation (2 D & 3 D) n Output viewer – graphs, tables n Sim. Runner – optimizer based on evolutionary algorithm technique n Opt. Quest is also available n Med. Model, Service. Model n 40

SIMUL 8 Service industries, transaction processing n Drop & Drag model development n Saves

SIMUL 8 Service industries, transaction processing n Drop & Drag model development n Saves in XML format n Pre-built templates for common applications n 3 D virtual reality graphics n Links to database n 41

Experimentation & Statistical Analysis Tools Included in most all simulation systems n Add-ons also

Experimentation & Statistical Analysis Tools Included in most all simulation systems n Add-ons also available n Features n q Optimization – define fitness or cost function 42

Arena Output & Process Analyzer Confidence intervals n Comparison of systems n Warm-up determinations

Arena Output & Process Analyzer Confidence intervals n Comparison of systems n Warm-up determinations n Graphs (all types) – 2 D & 3 D n Scenario definition n 43

Auto. Stat (from Auto. Mod) Warm-up determination n Steady state determination n Confidence intervals

Auto. Stat (from Auto. Mod) Warm-up determination n Steady state determination n Confidence intervals n Sensitivity analysis n Optimization via evolutionary strategy n 44

Opt. Quest Based on scatter search, tabu search, linear-integer programming, data mining, neural nets

Opt. Quest Based on scatter search, tabu search, linear-integer programming, data mining, neural nets (evolutionary) n Uncertainty problems n Global optimums n Handles non-linear and discontinuous relationships n 45

Sim. Runner (from Pro. Model) Based evolutionary models & genetic algorithms n Optimizations n

Sim. Runner (from Pro. Model) Based evolutionary models & genetic algorithms n Optimizations n 3 D graphics n Warm-up (steady state) determination n 46

Conclusion Many simulation software environments available n Many do have trial versions to download

Conclusion Many simulation software environments available n Many do have trial versions to download for trying n Before deciding, consider the features and the add-ons available that will suit your particular environment n 47