Upload
marianna-potter
View
240
Download
1
Tags:
Embed Size (px)
Citation preview
© 2008 Prentice Hall, Inc. F – 1
Operations ManagementOperations ManagementModule F – Module F – SimulationSimulation
PowerPoint presentation to accompany PowerPoint presentation to accompany Heizer/Render Heizer/Render Principles of Operations Management, 7ePrinciples of Operations Management, 7eOperations Management, 9e Operations Management, 9e
© 2008 Prentice Hall, Inc. F – 2
OutlineOutline
What Is Simulation?What Is Simulation?
Advantages and Disadvantages of Advantages and Disadvantages of SimulationSimulation
Monte Carlo SimulationMonte Carlo Simulation
Simulation of A Queuing ProblemSimulation of A Queuing Problem
Simulation and Inventory AnalysisSimulation and Inventory Analysis
© 2008 Prentice Hall, Inc. F – 3
Learning ObjectivesLearning Objectives
When you complete this module you When you complete this module you should be able to:should be able to:
List the advantages and disadvantages List the advantages and disadvantages of modeling with simulationof modeling with simulation
Perform the five steps in a Monte Carlo Perform the five steps in a Monte Carlo simulationsimulation
Simulate a queuing problemSimulate a queuing problem
Simulate an inventory problemSimulate an inventory problem
Use Excel spreadsheets to create a Use Excel spreadsheets to create a simulationsimulation
© 2008 Prentice Hall, Inc. F – 4
What is Simulation?What is Simulation?
An attempt to duplicate the An attempt to duplicate the features, appearance, and features, appearance, and characteristics of a real systemcharacteristics of a real system
1.1. To imitate a real-world situation To imitate a real-world situation mathematicallymathematically
2.2. To study its properties and To study its properties and operating characteristicsoperating characteristics
3.3. To draw conclusions and make To draw conclusions and make action decisions based on the action decisions based on the results of the simulationresults of the simulation
© 2008 Prentice Hall, Inc. F – 5
Computer AnalysisComputer Analysis
© 2008 Prentice Hall, Inc. F – 6
Simulation ApplicationsSimulation Applications
Ambulance location and Ambulance location and dispatchingdispatching
Assembly-line balancingAssembly-line balancing
Parking lot and harbor designParking lot and harbor design
Distribution system designDistribution system design
Scheduling aircraftScheduling aircraft
Labor-hiring decisionsLabor-hiring decisions
Personnel schedulingPersonnel scheduling
Traffic-light timingTraffic-light timing
Voting pattern predictionVoting pattern prediction
Bus schedulingBus scheduling
Design of library operationsDesign of library operations
Taxi, truck, and railroad Taxi, truck, and railroad dispatchingdispatching
Production facility schedulingProduction facility scheduling
Plant layoutPlant layout
Capital investmentsCapital investments
Production schedulingProduction scheduling
Sales forecastingSales forecasting
Inventory planning and controlInventory planning and control
Table F.1Table F.1
© 2008 Prentice Hall, Inc. F – 7
Select best course
Examine results
Conduct simulation
Specify valuesof variables
Construct model
Introduce variables
The The Process of Process of SimulationSimulation
Figure F.1Figure F.1
Define problem
© 2008 Prentice Hall, Inc. F – 8
Advantages of SimulationAdvantages of Simulation
1.1. Relatively straightforward and flexibleRelatively straightforward and flexible
2.2. Can be used to analyze large and Can be used to analyze large and complex real-world situations that complex real-world situations that cannot be solved by conventional cannot be solved by conventional modelsmodels
3.3. Real-world complications can be Real-world complications can be included that most OM models cannot included that most OM models cannot permitpermit
4.4. ““Time compression” is possibleTime compression” is possible
© 2008 Prentice Hall, Inc. F – 9
Advantages of SimulationAdvantages of Simulation
5.5. Allows “what-if” types of questionsAllows “what-if” types of questions
6.6. Does not interfere with real-world Does not interfere with real-world systemssystems
7.7. Can study the interactive effects of Can study the interactive effects of individual components or variables in individual components or variables in order to determine which ones are order to determine which ones are importantimportant
© 2008 Prentice Hall, Inc. F – 10
Disadvantages of SimulationDisadvantages of Simulation
1.1. Can be very expensive and may take Can be very expensive and may take months to developmonths to develop
2.2. It is a trial-and-error approach that may It is a trial-and-error approach that may produce different solutions in repeated produce different solutions in repeated runsruns
3.3. Managers must generate all of the Managers must generate all of the conditions and constraints for conditions and constraints for solutions they want to examinesolutions they want to examine
4.4. Each simulation model is uniqueEach simulation model is unique
© 2008 Prentice Hall, Inc. F – 11
Monte Carlo SimulationMonte Carlo Simulation
The Monte Carlo method may be used The Monte Carlo method may be used when the model contains elements thatwhen the model contains elements thatexhibit chance in their behaviorexhibit chance in their behavior
1.1. Set up probability distributions for important Set up probability distributions for important variablesvariables
2.2. Build a cumulative probability distribution for Build a cumulative probability distribution for each variableeach variable
3.3. Establish an interval of random numbers for Establish an interval of random numbers for each variableeach variable
4.4. Generate random numbersGenerate random numbers
5.5. Simulate a series of trialsSimulate a series of trials
© 2008 Prentice Hall, Inc. F – 12
Probability of DemandProbability of Demand(1)(1) (2)(2) (3)(3) (4)(4)
Demand Demand for Tiresfor Tires FrequencyFrequency
Probability of Probability of OccurrenceOccurrence
Cumulative Cumulative ProbabilityProbability
00 1010 10/200 = .0510/200 = .05 .05.05
11 2020 20/200 = .1020/200 = .10 .15.15
22 4040 40/200 = .2040/200 = .20 .35.35
33 6060 60/200 = .3060/200 = .30 .65.65
44 4040 40/200 = .2040/200 = .20 .85.85
55 3030 30/ 200 = .1530/ 200 = .15 1.001.00
200 days200 days 200/200 = 1.00200/200 = 1.00
Table F.2Table F.2
© 2008 Prentice Hall, Inc. F – 13
Assignment of Random Assignment of Random NumbersNumbers
Daily Daily DemandDemand ProbabilityProbability
Cumulative Cumulative ProbabilityProbability
Interval of Interval of Random Random NumbersNumbers
00 .05.05 .05.05 01 01 throughthrough 05 05
11 .10.10 .15.15 06 06 throughthrough 15 15
22 .20.20 .35.35 16 16 throughthrough 35 35
33 .30.30 .65.65 36 36 throughthrough 65 65
44 .20.20 .85.85 66 66 throughthrough 85 85
55 .15.15 1.001.00 86 86 throughthrough 00 00
Table F.3Table F.3
© 2008 Prentice Hall, Inc. F – 14
Table of Random NumbersTable of Random Numbers
5252 5050 6060 5252 0505
3737 2727 8080 6969 3434
8282 4545 5353 3333 5555
6969 8181 6969 3232 0909
9898 6666 3737 3030 7777
9696 7474 0606 4848 0808
3333 3030 6363 8888 4545
5050 5959 5757 1414 8484
8888 6767 0202 0202 8484
9090 6060 9494 8383 7777
Table F.4Table F.4
© 2008 Prentice Hall, Inc. F – 15
Simulation Example 1Simulation Example 1
Select random numbers from
Table F.3
DayDayNumberNumber
RandomRandomNumberNumber
Simulated Simulated Daily DemandDaily Demand
11 5252 33
22 3737 33
33 8282 44
44 6969 44
55 9898 55
66 9696 55
77 3333 22
88 5050 33
99 8888 55
1010 9090 55
3939 TotalTotal
3.93.9 Average Average
© 2008 Prentice Hall, Inc. F – 16
Simulation Example 1Simulation Example 1
DayDayNumberNumber
RandomRandomNumberNumber
Simulated Simulated Daily DemandDaily Demand
11 5252 33
22 3737 33
33 8282 44
44 6969 44
55 9898 55
66 9696 55
77 3333 22
88 5050 33
99 8888 55
1010 9090 55
3939 TotalTotal
3.93.9 Average Average
Expecteddemand = ∑ (probability of i units) x
(demand of i units)
= (.05)(0) + (.10)(1) + (.20)(2) + (.30)(3) + (.20)(4) + (.15)(5)
= 0 + .1 + .4 + .9 + .8 + .75
= 2.95 tires
5
i =1
© 2008 Prentice Hall, Inc. F – 17
Queuing SimulationQueuing Simulation
Number Number of Arrivalsof Arrivals ProbabilityProbability
Cumulative Cumulative ProbabilityProbability
Random-NumberRandom-NumberIntervalInterval
00 .13.13 .13.13 01 01 throughthrough 13 13
11 .17.17 .30.30 14 14 throughthrough 30 30
22 .15.15 .45.45 31 31 throughthrough 45 45
33 .25.25 .70.70 46 46 throughthrough 70 70
44 .20.20 .90.90 71 71 throughthrough 90 90
55 .10.10 1.001.00 91 91 throughthrough 00 00
1.001.00
Overnight barge arrival rates Overnight barge arrival rates Table F.5Table F.5
© 2008 Prentice Hall, Inc. F – 18
Queuing SimulationQueuing Simulation
Daily Daily Unloading Unloading
RatesRates ProbabilityProbabilityCumulative Cumulative ProbabilityProbability
Random-NumberRandom-NumberIntervalInterval
11 .05.05 .05.05 01 01 throughthrough 05 05
22 .15.15 .20.20 06 06 throughthrough 20 20
33 .50.50 .70.70 21 21 throughthrough 70 70
44 .20.20 .90.90 71 71 throughthrough 90 90
55 .10.10 1.001.00 91 91 throughthrough 00 00
1.001.00
Barge unloading rates Barge unloading rates Table F.6Table F.6
© 2008 Prentice Hall, Inc. F – 19
Queuing SimulationQueuing Simulation(1)(1)
DayDay
(2)(2)NumberNumber
Delayed fromDelayed fromPrevious DayPrevious Day
(3)(3)
Random Random NumberNumber
(4)(4)NumberNumber
of Nightlyof NightlyArrivalsArrivals
(5)(5)TotalTotalto Beto Be
UnloadedUnloaded
(6)(6)
Random Random NumberNumber
(7)(7)
Number Number UnloadedUnloaded
11 00 5252 33 33 3737 33
22 00 0606 00 00 6363 00
33 00 5050 33 33 2828 33
44 00 8888 44 44 0202 11
55 33 5353 33 66 7474 44
66 22 3030 11 33 3535 33
77 00 1010 00 00 2424 00
88 00 4747 33 33 0303 11
99 22 9999 55 77 2929 33
1010 44 3737 22 66 6060 33
1111 33 6666 33 66 7474 44
1212 22 9191 55 77 8585 44
1313 33 3535 22 55 9090 44
1414 11 3232 22 33 7373 33
1515 00 0000 55 55 5959 33
2020 4141 3939
© 2008 Prentice Hall, Inc. F – 20
Queuing SimulationQueuing Simulation
Average number of bargesAverage number of bargesdelayed to the next daydelayed to the next day ==
= 1.33= 1.33 barges delayed per day barges delayed per day
20 20 delaysdelays1515 days days
Average number of Average number of nightly arrivalsnightly arrivals ==
= 2.73= 2.73 arrivals per night arrivals per night
41 41 arrivalsarrivals1515 days days
Average number of bargesAverage number of bargesunloaded each dayunloaded each day ==
= 2.60= 2.60 unloadings per day unloadings per day
39 39 unloadingsunloadings1515 days days
© 2008 Prentice Hall, Inc. F – 21
Inventory SimulationInventory Simulation
(1)(1)Demand forDemand for
Ace DrillAce Drill
(2)(2)
FrequencyFrequency
(3)(3)
ProbabilityProbability
(4)(4)CumulativeCumulativeProbabilityProbability
(5)(5)Interval ofInterval of
Random NumbersRandom Numbers
00 1515 .05.05 .05.05 01 01 throughthrough 05 05
11 3030 .10.10 .15.15 06 06 throughthrough 15 15
22 6060 .20.20 .35.35 16 16 throughthrough 35 35
33 120120 .40.40 .75.75 36 36 throughthrough 75 75
44 4545 .15.15 .90.90 76 76 throughthrough 90 90
55 3030 .10.10 1.001.00 91 91 throughthrough 00 00
300300 1.001.00
Table F.8Table F.8
Daily demand for Ace DrillDaily demand for Ace Drill
© 2008 Prentice Hall, Inc. F – 22
Inventory SimulationInventory Simulation
(1)(1)Demand forDemand for
Ace DrillAce Drill
(2)(2)
FrequencyFrequency
(3)(3)
ProbabilityProbability
(4)(4)CumulativeCumulativeProbabilityProbability
(5)(5)Interval ofInterval of
Random NumbersRandom Numbers
11 1010 .20.20 .20.20 01 01 throughthrough 20 20
22 2525 .50.50 .70.70 21 21 throughthrough 70 70
33 1515 .30.30 1.001.00 71 71 throughthrough 00 00
5050 1.001.00
Table F.9Table F.9
Reorder lead timeReorder lead time
© 2008 Prentice Hall, Inc. F – 23
Inventory SimulationInventory Simulation
1.1. Begin each simulation day by checking to see if Begin each simulation day by checking to see if ordered inventory has arrived. If if has, increase ordered inventory has arrived. If if has, increase current inventory by the quantity ordered.current inventory by the quantity ordered.
2.2. Generate daily demand using probability Generate daily demand using probability distribution and random numbers.distribution and random numbers.
3.3. Compute ending inventory. If on-hand is Compute ending inventory. If on-hand is insufficient to meet demand, satisfy as much as insufficient to meet demand, satisfy as much as possible and note lost sales.possible and note lost sales.
4.4. Determine whether the day's ending inventory has Determine whether the day's ending inventory has reached the reorder point. If it has, and there are reached the reorder point. If it has, and there are no outstanding orders, place an order. Choose no outstanding orders, place an order. Choose lead time using probability distribution and lead time using probability distribution and random numbers.random numbers.
© 2008 Prentice Hall, Inc. F – 24
Inventory SimulationInventory Simulation
(1)(1)
DayDay
(2)(2)UnitsUnits
ReceivedReceived
(3)(3)Beginning Beginning InventoryInventory
(4)(4)Random Random NumberNumber
(5)(5)
DemandDemand
(6)(6)Ending Ending
InventoryInventory
(7)(7)LostLostSalesSales
(8)(8)
Order?Order?
(9)(9)RandomRandomNumberNumber
(10)(10)LeadLeadTimeTime
11 1010 0606 11 99 00 NoNo
22 00 99 6363 33 66 00 NoNo
33 00 66 5757 33 33 00 YesYes 0202 11
44 00 33 9494 55 00 22 NoNo
55 1010 1010 5252 33 77 00 NoNo
66 00 77 6969 33 44 00 YesYes 3333 22
77 00 44 3232 22 22 00 NoNo
88 00 22 3030 22 00 00 NoNo
99 1010 1010 4848 33 77 00 NoNo
1010 00 77 8888 44 33 00 YesYes 1414 11
4141 22
Table F.10Table F.10Order quantity = Order quantity = 10 10 units Reorder point = units Reorder point = 55 units units
© 2008 Prentice Hall, Inc. F – 25
Inventory SimulationInventory Simulation
Average ending inventory Average ending inventory = = 4.1= = 4.1 units/day units/day4141 total units total units1010 days days
Average lost sales Average lost sales = = .2= = .2 unit/day unit/day22 sales lost sales lost1010 days days
= = .3= = .3 order/day order/day3 3 ordersorders1010 days days
Average number Average number of orders placedof orders placed
© 2008 Prentice Hall, Inc. F – 26
Inventory SimulationInventory SimulationDaily order costDaily order cost == ((cost of placingcost of placing 1 1 orderorder) x ) x ((number of orders placed per daynumber of orders placed per day))
== $10 $10 per orderper order xx .3 .3 order per dayorder per day = $3 = $3Daily holding costDaily holding cost == ((cost of holdingcost of holding 1 1 unit forunit for 1 1 dayday) x ) x ((average ending inventoryaverage ending inventory))
== 50¢ 50¢ per unit per per unit per dayday xx 4.1 units 4.1 units per day per day
== $2.05$2.05Daily stockout costDaily stockout cost== ((cost per lost cost per lost salesale) x ) x ((averageaverage number of lost sales per daynumber of lost sales per day))
== $8 $8 per lost saleper lost sale xx .2 .2 lost sales per daylost sales per day
== $1.60$1.60Total daily inventory costTotal daily inventory cost== DailyDaily order costorder cost + D + Daily holding cost + aily holding cost + Daily stockout costDaily stockout cost
==$6.65$6.65
© 2008 Prentice Hall, Inc. F – 27
Using Software in SimulationUsing Software in Simulation
Computers are critical in simulating Computers are critical in simulating complex taskscomplex tasks
General-purpose languages - BASIC, C++General-purpose languages - BASIC, C++
Special-purpose simulation languages - Special-purpose simulation languages - GPSS, SIMSCRIPTGPSS, SIMSCRIPT
1.1. Require less programming time for large Require less programming time for large simulationssimulations
2.2. Usually more efficient and easier to check Usually more efficient and easier to check for errorsfor errors
3.3. Random-number generators are built inRandom-number generators are built in
© 2008 Prentice Hall, Inc. F – 28
Using Software in SimulationUsing Software in Simulation
Commercial simulation programs are Commercial simulation programs are available for many applications - Extend, available for many applications - Extend, Modsim, Witness, MAP/1, Enterprise Modsim, Witness, MAP/1, Enterprise Dynamics, Simfactory, ProModel, Micro Dynamics, Simfactory, ProModel, Micro Saint, ARENASaint, ARENA
Spreadsheets such as Excel can be used Spreadsheets such as Excel can be used to develop some simulationsto develop some simulations
© 2008 Prentice Hall, Inc. F – 29
Using Software in SimulationUsing Software in Simulation