Date post: | 10-Apr-2018 |
Category: |
Documents |
Upload: | rarautis-mademoisselle |
View: | 221 times |
Download: | 0 times |
8/8/2019 02 Introduction to Simulation
http://slidepdf.com/reader/full/02-introduction-to-simulation 1/11
1
Lecture 1
INTRODUCTION TOSIMULATION
Lecture 1
INTRODUCTION TOSIMULATION
Simulasi [email protected] 1
Winda Nur Cahyo, ST., MT.Faculty of Industrial TechnologyIslamic University of [email protected]
[email protected](YM, FB)Website: www.teknik-industri.com
What is Simulation?• The Oxford American Dictionary (1980):
– Simulation is a way “ to reproduce the conditions of asituation, as by means of a model, for study ortesting or training, etc. ”
• For our ur oses we are interested in
Simulasi [email protected] 2
reproducing the operational behavior of dynamicsystems.
• The model that will be using is a computermodel.
• Simulation can be defined as the imitation of adynamic system using a computer model.
What simulation …
• Schriber (1987) – Simulation is “ the modeling of a process or
system in such a way that the model mimicsthe response of the actual system to events
Simulasi [email protected] 3
that take place over time. ”
• By studying the behavior of the model,insight about the behavior of the actualsystem can be gained.
• In practice, – Simulation is performed using commercial
simulation software. – Performance statistics are gathered during
the simulation – Modern simulation software provides a
realistic, graphical animation of the system
Simulasi [email protected] 4
being modeled. – During the simulation, the user can
interactively adjust the animation speed andchange model parameter values to do “ what-if” analysis on the fly.
– State-of-the art simulation technologyprovides optimization capability
• This lecture focuses primarily on discrete-event simulation, which models the effects ofthe events in a system as they occur overtime.
• Discrete-event simulation employs statisticalmethods for generating random behavior and
Simulasi [email protected] 5
es ma ng mo e per ormance.• These methods are sometimes referred to as
Monte Carlo methods because of theirsimilarity to the probabilistic outcomes foundin games of chances.
Why Simulate?• Simulation provides a way to validate whether or
not the best decisions are being made.• Simulation avoid the expensive, time-
consuming, and disrupted nature of traditionaltrial-and-error techniques.
• The power of simulation lies in the fact that it
Simulasi [email protected] 6
prov es a me o o ana ys s a s no on yformal and predictive, but is capable ofaccurately predicting the performance of asystem.
• By using a computer to model a system before itis built or to test operating policies before theyare actually implemented, many of the pitfallscan be avoided
8/8/2019 02 Introduction to Simulation
http://slidepdf.com/reader/full/02-introduction-to-simulation 2/11
2
• The characteristics of simulation that make it such apowerful planning and decision-making tool: – captures system interdependence – accounts for variability in the system – is versatile enough to model any system – shows behavior over time – is less costly, time-consuming, and disruptive that
Simulasi [email protected] 7
– provides information on multiple performance measures – is visually appealing and engages people ’ s interest – provides results that are easy to understand and
communicate – runs in compressed, real, or even delayed time – forces attention to detail in a design
Doing Simulation• Simulation is nearly always performed as a part
of a larger process of system design or processimprovement.
• Alternative solutions are generated andevaluated, and the best solution is selected and
Simulasi [email protected] 8
implemented.• Simulation comes into play during the evaluation
phase.• Simulation is an experimentation tool in which
a computer model of a new or existing system iscreated for the purpose of conductingexperiments.
Doing Simulation …
Simulation provides a virtual method for doingsystem experimentation
Simulasi [email protected] 9
Doing Simulation … .
• Doing simulation talks about “ the processof designing a model of a real system andconducting experiments with this model ” .
Simulasi [email protected] 10
•reduces the time, cost, and disruption of
experimenting on the actual system. – Simulation is a virtual prototyping tool for
demonstrating proof of concept.
Doing Simulation…
..• The procedure for doing simulation follows
the scientific method of: – formulating a hypothesis,
Simulasi [email protected] 11
– , – testing the hypothesis through
experimentation – drawing conclusions about the validity of the
hypothesis.
The process ofsimulation experimentation
START FORMULATE A HYPOTHESIS
DEVELOP A SIMULATION
Simulasi [email protected] 12
RESUME A SIMULATIONEXPERIMENT
HYPOTHESISCORRECT?
STOPYESYES NONO
8/8/2019 02 Introduction to Simulation
http://slidepdf.com/reader/full/02-introduction-to-simulation 3/11
3
Use of Simulation
• Simulation began to be used incommercial applications in 1960s. – Initial models were usually programmed in
Simulasi [email protected] 13
.
• Only in the last couple of decades hassimulation gained popularity as a decision-making tool in manufacturing and serviceindustries
Use of Simulation ………
• The surge in popularity of computersimulation: – Increased awareness and understanding of
Simulasi [email protected] 14
. – Increased availability, capability, and ease of
use of simulation software. – Increased computer memory, processing
speeds, especially of PCs. – Declining computer hardware and software
costs.
Typical Applications ofSimulation
• Work-flow planning• Capacity planning• Cycle time reduction• Staff and resource
• Throughput analysis• Productivity improvement• Layout analysis• Line balancing
Simulasi [email protected] 15
planning• Work prioritization
• Bottleneck analysis• Quality improvement• Cost reduction• Inventory reduction
• Batch size optimization• Production scheduling
• Resource scheduling• Maintenance scheduling• Control system design
When Simulation isAppropriate
• Not all system problems that could besolved with the aid of simulation should besolved using simulation,
Simulasi [email protected] 16
•task.
• Simulation has certain limitations of whichone should be aware before making adecision to apply it to a given situation.
When Simulation is Appropriate… ..
• As a general guideline, simulation isappropriate if – An operational (logical or quantitative) decision is
being made.
Simulasi [email protected] 17
– The process being analyzed is well defined andrepetitive.
– Activities and events are interdependent and variable. – The cost impact of the decision is greater than the
cost of doing the simulation. – The cost of experiment on the actual system is
greater than the cost of simulation.
Qualification for DoingSimulation
• Participants in the simulation project includemodeler, decision maker, and process owner.
• A certain degree of knowledge and skill: – Project management – Communication
Simulasi [email protected] 18
– System engineering – Statistical analysis and design of experiments – Modeling principles and concepts – Basic programming and computer skills – Training on one or more simulation products – Familiarity with the system being investigated
8/8/2019 02 Introduction to Simulation
http://slidepdf.com/reader/full/02-introduction-to-simulation 4/11
4
Economic Justification ofSimulation
• Cost is always important issues whenconsidering the use of any software tool andsimulation is no exception.
• Simulation should not use if the cost exceeds
Simulasi [email protected] 19
.• This means that both the costs and the benefits
should be carefully assessed.• The use of simulation is often prematurely
dismissed due to the failure to recognize thepotential benefits and savings it can produce.
• Savings from simulation are realized byidentifying and eliminating problems and
inefficiencies.• Cost is reduced by eliminating overdesign and
removing excessive safety factors.• One of the difficulties in develo in an economic
Simulasi [email protected] 20
justification for simulation is the fact that it isusually not known in advance how much savingswill be realized.
• One way to assess in advance the economicbenefit of simulation is to assess the risk ofmaking poor design and operational decisions.
Economic Justification of Simulation……
• The real savings from a simulation comefrom allowing to make mistake and workout design errors on the model rather than
Simulasi [email protected] 21
.• Simulation helps avoid many of the
downstream costs associated with poordecision that are made up front.
Cost of making changes at subsequent stagesof system development
Simulasi [email protected] 22
Comparison of cumulative system costs withand without simulation
Simulasi [email protected] 23
System Approach
Simulasi [email protected] 24
Minggu ke 1 bagian 2
8/8/2019 02 Introduction to Simulation
http://slidepdf.com/reader/full/02-introduction-to-simulation 5/11
5
System Definition
• A system is defined as a collection ofelements that function together to achievea desired goal.
Simulasi [email protected] 25
• – A system consists of multiple elements. – These elements are interrelated and work in
cooperation. – A system exists for the purpose of achieving
specific objectives.
• Examples of systems: – Traffic systems
– Political systems – Economic systems – Manufacturing systems – Service systems
Simulasi [email protected] 26
• Main focus of our subject – manufacturing and service systems that
process materials, information, and people.
Manufacturing System
• Manufacturing systems: – Small job shops – Machining cells
Simulasi [email protected] 27
– arge pro uc on ac es – Assembly lines
– Warehousing – Distribution – Supply chain systems
Service System
• Service systems: – Health care facilities – Call centers
Simulasi [email protected] 28
– musemen par s – Public transportation systems
– Restaurant – Bank – etc
• Both manufacturing and service systemsmay be termed processing systems . – They process items through a series of
Processing Systems
Simulasi [email protected] 29
..
• Processing systems: – Artificial (human-made) – Dynamic (elements interact overtime) – Usually stochastic (they exhibit random
behavior)
System Elements• From a simulation perspective, a system
consists of entities , activities , resources ,controls .
Simulasi [email protected] 30
, , ,when , and how of entity processing.
8/8/2019 02 Introduction to Simulation
http://slidepdf.com/reader/full/02-introduction-to-simulation 6/11
6
System : Input-Output Box
Simulasi [email protected] 31
• items processed through the system such as
products, customers, and documents.• divided into:
• human or animate (customers, patients, etc.)• inanimate (parts, documents, bins, etc.)
Entities
Simulasi [email protected] 32
• intangible (calls, electronic mail, etc.)
• [For most manufacturing and service systems]discrete items.
• [For some production systems: continuoussystems] nondiscrete substance
• Example: oil refineries, paper mills
Activities• the tasks performed in the system (directly or
indirectly) in the processing of entities.• Servicing a customer• cutting a part on machine• repairing a piece of equipment
• consume time and often involve the use of
Simulasi [email protected] 33
resources• classified as
• entity processing (check-in, treatment, inspection, fabrication,etc.)
• entity and resource movement (forklift travel, riding in anelevator, etc.)
• resource adjustments, maintenance, and repairs (machinesetups, copy machine repair, etc.)
Resources
– the means by which activities are performed. – provide the supporting facilities, equipment,
and personnel for carrying out activities. – can constrain rocessin b limitin the rate
Simulasi [email protected] 34
at which processing can take place.
– have characteristics, e.g. capacity, speed,cycle time, and reliability.
Resources
– can be categorized as:• Human or animate (operators, doctors,
maintenance personnel, etc.)• Inanimate (equipment, tooling, floor space, etc.)
Simulasi [email protected] 35
• Intangible (information, electrical power, etc.)
– also can be classified as• dedicated or shared• permanent or consumable• mobile or stationary
Controls – dictate how, when, and where activities are
performed. – impose order on the system. – [at the highest level] consists of schedules,
lan and olicies
Simulasi [email protected] 36
, . – [at the lowest level] take the forms of written
procedures and machine control logic. – [at all levels] provide the information and
decision logic for how the system shouldoperate.
8/8/2019 02 Introduction to Simulation
http://slidepdf.com/reader/full/02-introduction-to-simulation 7/11
7
Controls
– Examples:• Routing sequences• Production plans•
Simulasi [email protected] 37
• Task prioritization• Control software• Instruction sheets
System Complexity• Elements of a system operate with one another
in ways that often result in complex interactions.• Unaided human intuition is not very good at
analyzing and understanding complex systems.• Inability of the human mind to grasp real-world
complexity is called as “ the principle of bounded”
Simulasi [email protected] 38
.• This principle states that “ the capacity of human
mind for formulating and solving complexproblem is very small compared with the size ofproblem whose solution is required forobjectively rational behavior in the real world, oreven for a reasonable approximation to suchobjective rationality (Simon, 1957).
System Complexity
• is a primary function of two factors: – Interdependencies between elements so that
each element affects other elements. –
Simulasi [email protected] 39
uncertainty.
System ComplexityThe degree of analytical difficulty increasesexponentially as the number of interdependenciesand random variables increase.
Simulasi [email protected] 40
System Performance Metrics• Metrics are measures used to assess the
performance of a system.• At the highest level of an organization or
business, metrics measure overall performancein terms of profits, revenues, cost relative tobudget, return on assets, and so on.
Simulasi [email protected] 41
– Such metrics are inherently lagging, disguise low-level performance, and are reported only periodically
• From an operational standpoint, it is morebeneficial to track such factors as time, quality,quantity, efficiency, and utilization. – These operational metrics reflect immediate activity
and are directly controllable – They drive the higher financially related metrics.
Key operational metrics• Describe the effectiveness and efficiency of
manufacturing and service systems: – Flow time – Utilization
Simulasi [email protected] 42
– - – Waiting time – Flow rate – Inventory or queue levels – Yield – Customer responsiveness – Variance
8/8/2019 02 Introduction to Simulation
http://slidepdf.com/reader/full/02-introduction-to-simulation 8/11
8
System Variables• Designing a new system or improving an existing system
requires more than simply identifying the elements andperformance goals of the system.
• It requires an understanding of how system elementsaffect each other and overall performance objectives.
• Three types of system variable must be understand:
Simulasi [email protected] 43
– Decision variables – Response variables – State variables
System Variables
• Decision variables – called as input factors or independent
variables – ’
Simulasi [email protected] 44
independent variables affects the behavior ofthe system
– controllable or uncontrollable – controllable variable decision variables
Response Variables• Response variables
– called as performance or output variables – measure performance of the system in
response to particular decision variable
Simulasi [email protected] 45
settings. – In an experiment, the response variable is the
dependent variable. – The goal in system planning is to find the right
values or settings of decision variables thatgive the desired response value.
State Variables
• State variables – State variables are the status of the system at
any specific point in time. –
Simulasi [email protected] 46
state variable changes over time.
– State variables are dependent variables.
System Optimization• Optimization is finding the right setting for
decision variables that best meetsperformance objectives.
• O timization seeks the best combination
Simulasi [email protected] 47
of decision variable values that eitherminimizes or maximizes some objectivefunctions such as costs or profits.
• An objective function is a responsevariable of the system.
System Optimization• A typical objective in an optimization
problem for a manufacturing or servicesystems:
Simulasi [email protected] 48
– – maximizing flow rate
• Optimization problems may includeconstraints that limits the values ofdecision variables.
8/8/2019 02 Introduction to Simulation
http://slidepdf.com/reader/full/02-introduction-to-simulation 9/11
9
System Optimization
• In some instances, there are problems ofconflicting objectives.
Simulasi [email protected] 49
System Approach
• Due to departmentalization andspecialization, decisions in the real worldoften made without regard to overall
Simulasi [email protected] 50
.• Approaching system design with overall
objectives in mind and considering howeach element relates to each other and tothe whole is called a systems or holisticapproach to system design.
System ApproachFour-step iterative approach to systems improvement
Simulasi [email protected] 51
System Approach
• Identifying problems and opportunities – Developing a solution starts by understanding the
problem, identifying key variables, and describingimportant relationships.
– This hel s identif ossible areas of focus and
Simulasi [email protected] 52
leverage points for applying a solution. – Techniques such as cause-and-effect analysis and
pareto analysis are useful. – Performance standards must be set high in order to
look for the greatest improvement opportunities. – Setting high standards pushes people to think
creatively and often results in breakthroughimprovement
System Approach• Developing alternative solutions
– Once a problem or opportunity has beenidentified and key decision variables isolated,alternative solution can be explored.
Simulasi [email protected] 53
– This is where most of the design andengineering expertise comes into play.
– Generating alternative solutions requirescreatively as well as organizational andengineering skills.
– Simulation is particularly helpful in that itencourages thinking in radical new ways.
System Approach• Evaluating the solutions
– Alternative solutions should be evaluated based on their ability tomeet the criteria established for the evaluation.
– These criteria include performance goals, cost of implementation,impact on the socio-technical infrastructure, and consistencywith organizational strategies.
Simulasi [email protected] 54
– Many of these criteria are difficult to measure in absolute terms. – After narrowing the list to two or three of the most promising
solutions using common sense and rough-cut analysis, moreprecise evaluation techniques may need to be used.
– This is where simulation and other formal analysis tools comeinto play.
8/8/2019 02 Introduction to Simulation
http://slidepdf.com/reader/full/02-introduction-to-simulation 10/11
10
System Approach• Selecting and implementing the best solution
– Often the final selection of what solution to implementis not left to the analyst, but rather a managementdecision.
– The analyst ’ s role is to present his/her evaluation inthe clearest wa ossible so that an informed
Simulasi [email protected] 55
decision can be made.
– Even after a solution is selected, additional modelingand analysis are often needed for fine-tuning thesolution.
– Implementers should then be careful to make surethat the system is implemented as designed,documenting reasons for any modifications.
System Analysis Techniques• While simulation is perhaps the most versatile
and powerful system analysis tool, otheravailable techniques also can be useful inplanning.
• These alternative techniques are usually
Simulasi [email protected] 56
compu a ona me o s a wor we or s mp esystems with little interdependency andvariability.
• For more complex systems, these techniquesstill can provide rough estimates but fall short inproducing the insights and accurate answersthat simulation provides.
System Analysis TechniquesSimulation improves performance predictability
Simulasi [email protected] 57
System Analysis Techniques
• In addition to simulation, system analysistools include: – Hand calculations
Simulasi [email protected] 58
– – Operations Research techniques
System Analysis Techniques• Hand calculations
– Quick-and-dirty, pencil-and-paper sketches andcalculations can be remarkably helpful inunderstanding basic requirements for a system
Simulasi [email protected] 59
– ome ecs ons may e so asc a a qu c men acalculation yields the needed results.
– Most of these calculations involve simple algebra. – The obvious drawback is the inability to manually
perform complex calculations or to take into accounttens or potentially even hundreds of complexrelationship simultaneously.
System Analysis Techniques• Spreadsheets
– What-if experiments can be run onspreadsheet based on expected values andsimple interactions.
Simulasi [email protected] 60
– Spreadsheet simulation can be very usefulfor getting rough performance estimates.
– Weaknesses of spreadsheet modeling:• Some potential problems are not readily apparent• All behavior is assumed to be period-driven rather
than event-driven
8/8/2019 02 Introduction to Simulation
http://slidepdf.com/reader/full/02-introduction-to-simulation 11/11
11
• OResearch Techniques – Traditional OR techniques utilize mathematical
models to solve problems. – These mathematical models include both
deterministic models and probabilistic models (e.g.
System Analysis Techniques
Simulasi [email protected] 61
queuing and decision trees). – These OR techniques provide quick, quantitative
answers without going through the guessworkprocess of trial and error.
– OR techniques can be divided into: prescriptive anddescriptive
System Analysis Techniques
• Prescriptive techniques – an optimum solution to a problem – linear programming, dynamic programming – do not allow random variables – conditions are constant over the eriod of stud
Simulasi [email protected] 62
• Descriptive techniques
– static analysis techniques such as queuing theory thatprovide good estimates for basic problems
– limited to only one or two metrics – give only average performance measures rather than
a complete picture of performance over time
QUEUING SYSTEM
Simulasi [email protected] 63
QUEUING SYSTEM• Elements of queuing system
– Input source (calling population)• Size• Arrival distribution
– Queue
Simulasi [email protected] 64
• engt
– Queuing disciplines
• FIFO, LIFO, priority – Service mechanism
• Number of service facilities• Structure of service facilities• Service distribution
QUEUING SYSTEM• Kendall ’ s notation
(a/b/c ) : (d/e/f )
Simulasi [email protected] 65
a : Interarrival time distributionb : Service time distributionc : Number of parallel serversd : Service disciplinee : Maximum number of queue
f : Number of calling population
QUEUING SYSTEMFor a and b
M : Exponential distribution D : Degenerate distribution E k : Erlang distribution
Simulasi [email protected] 66
G : General distribution for service time
Untuk disiplin pelayanan
FIFO : First-in, first out LIFO : Last-in, first outSIRO : Service in random order