Decision
Making
Decision
Making
Module based on
Module based on
Operation Management, 9e
Operation Management, 9e
PowerPoint presentation to accompany PowerPoint presentation to accompany Heizer
Heizer/Render /Render
Lecturer: F. Priyo Suprobo, ST, MT Lecturer: F. Priyo Suprobo, ST, MT
Product Design of Product Design of ITATS ITATS
Permasalahan
Permasalahan
Konsultan
Konsultan desaindesain HCIDHCID--ITATSITATS bekerjabekerja untukuntuk HealthyHealthy PillowPillow Company
Company sedangsedang mengusulkanmengusulkan rancanganrancangan AlasAlas TidurTidur KesehatanKesehatan yang
yang mutakhirmutakhir dengandengan beberapabeberapa pilihanpilihan.. BekerjasamaBekerjasama dengandengan tenaga
tenaga pemasaranpemasaran HealthyHealthy PillowPillow dirumuskanlahdirumuskanlah beberapabeberapa alternatif
alternatif berikutberikut peluangpeluang keberhasilannyakeberhasilannya sebagaisebagai berikutberikut::
Selanjutnya, terhadap alternatif yang ada, apakah Selanjutnya, terhadap alternatif yang ada, apakah saran Anda sebagai staf HCID
saran Anda sebagai staf HCID--ITATS untuk Healthy ITATS untuk Healthy Pillow ini?
Outline
Outline
;
; Proses Keputusan
Proses Keputusan
;
; Dasar
Dasar--Dasar Pengambilan
Dasar Pengambilan
Keputusan
Keputusan
;
Outline
Outline –
– Continued
Continued
;
; Tipe Pengambilan Keputusan
Tipe Pengambilan Keputusan
;
; Pengambilan Keputusan dalam Pengambilan Keputusan dalam Ketidakpastian
Ketidakpastian
;
; Pengambilan Keputusan dengan Pengambilan Keputusan dengan Resiko
Resiko
;
; Pengambilan Keputusan dalam Pengambilan Keputusan dalam Kepastian
Kepastian
;
; Expected Value of Perfect Expected Value of Perfect Information (EVPI)
Outline
Outline –
– Continued
Continued
;
; Pohon Keputusan
Pohon Keputusan
;
; Pohon Keputusan SederhanaPohon Keputusan Sederhana
;
; Pohon Keputusan yang lebih Pohon Keputusan yang lebih Kompleks
Learning Objectives
Learning Objectives
When you complete this module you
When you complete this module you
should be able to:
should be able to:
1.
1. Membuat sebuah pohon keputusan Membuat sebuah pohon keputusan sederhana
sederhana 2.
2. Membangun tabel keputusanMembangun tabel keputusan 3.
3. Menjelaskan kapan menggunakan Menjelaskan kapan menggunakan salah satu tipe dalam pengambilan salah satu tipe dalam pengambilan keputusan
keputusan 4.
4. MenghitungMenghitung expected monetary value expected monetary value (EMV)
Learning Objectives
Learning Objectives
When you complete this module you
When you complete this module you
should be able to:
should be able to:
5.
5. MenghitungMenghitung expected value of perfect expected value of perfect information (EVPI)
information (EVPI) 6.
6. Mengevaluasi titikMengevaluasi titik--titik dalam Pohon titik dalam Pohon Keputusan
Keputusan 7.
7. Membuat Pohon Keputusan dengan Membuat Pohon Keputusan dengan penyelesaian berurutan
The Decision Process in
The Decision Process in
Operations
Operations
1.
1. Clearly define the problems and the Clearly define the problems and the factors that influence it
factors that influence it 2.
2. Develop specific and measurable Develop specific and measurable objectives
objectives 3.
3. Develop a modelDevelop a model 4.
4. Evaluate each alternative solutionEvaluate each alternative solution 5.
5. Select the best alternativeSelect the best alternative 6.
6. Implement the decision and set a Implement the decision and set a timetable for completion
Fundamentals of
Fundamentals of
Decision Making
Decision Making
1.
1. Terminologi/Istilah
Terminologi/Istilah::
a.a. AlternativeAlternative –– Sebuah tindakan atau Sebuah tindakan atau strategi yang dapat dipilih oleh
strategi yang dapat dipilih oleh pengambil keputusan
pengambil keputusan b.
b. State of State of naturenature/Kondisi Alami/Kondisi Alami –– Sebuah kejadian atau kondisi Sebuah kejadian atau kondisi dimana pengambil keputusan dimana pengambil keputusan
hanya punya sedikit kendali atau hanya punya sedikit kendali atau tidak sama sekali
Fundamentals of
Fundamentals of
Decision Making
Decision Making
2.
2. Symbols
Symbols dalam Pohon Keputusan
dalam Pohon Keputusan::
a.
a. –– Sebuah titik keputusan dimana Sebuah titik keputusan dimana terdapat satu atau lebih alternatif terdapat satu atau lebih alternatif yang dapat dipilih
yang dapat dipilih
b.
b. {{ –– sebuah simbol titik kondisi sebuah simbol titik kondisi alami yang mungkin terjadi
Decision Tree Example
Decision Tree Example
Pasar sesuai harapan Pasar sesuai harapan
Pasar tidak sesuai harapan Pasar tidak sesuai harapan
Pasar sesuai harapan Pasar sesuai harapan
Pasar tidak sesuai harapan Pasar tidak sesuai harapan Desain TPC
Desain TPC Titik Keputusan
Titik Keputusan Titik Kondisi AlamiTitik Kondisi Alami
Figure A.1 Figure A.1
Decision Table Example
Decision Table Example
Kondisi Alami Kondisi Alami Alternatives
Alternatives Pasar sesuaiPasar sesuai Pasar TidakSesuaiPasar TidakSesuai Desain UMPC Desain UMPC $200,000$200,000 ––$180,000$180,000 Desain Tablet PC Desain Tablet PC $100,000$100,000 ––$ 20,000$ 20,000 Do nothing Do nothing $ 0$ 0 $ 0$ 0 Table A.1 Table A.1
Decision
Decision--Making
Making
Environments
Environments
;
; Pengambilan Keputusan dalam KetidakpastianPengambilan Keputusan dalam Ketidakpastian
;
; Kondisi alami tidak dapat diperkirakanKondisi alami tidak dapat diperkirakan
;
; Pengambilan Keputusan dengan ResikoPengambilan Keputusan dengan Resiko
;
; Beberapa kondisi alami mungkin terjadiBeberapa kondisi alami mungkin terjadi
;
; Tetapi masingTetapi masing--masing pilihan tetap berpeluangmasing pilihan tetap berpeluang
;
; Pengambilan Keputusan dalam KepastianPengambilan Keputusan dalam Kepastian
;
Ketidakpastian
Ketidakpastian
1.
1. Maximax
Maximax
;
; Find the alternative that maximizes Find the alternative that maximizes the maximum outcome for every
the maximum outcome for every alternative
alternative
;
; Pick the outcome with the maximum Pick the outcome with the maximum number
number
;
; Highest possible gainHighest possible gain
;
; This is viewed as an optimistic This is viewed as an optimistic approach
Ketidakpastian
Ketidakpastian
2.
2. Maximin
Maximin
;
; Find the alternative that maximizes Find the alternative that maximizes the minimum outcome for every
the minimum outcome for every alternative
alternative
;
; Pick the outcome with the minimum Pick the outcome with the minimum number
number
;
; Least possible lossLeast possible loss
;
; This is viewed as a pessimistic This is viewed as a pessimistic approach
Ketidakpastian
Ketidakpastian
3.
3. Equally
Equally likely
likely (Sama rata)
(Sama rata)
;
; Find the alternative with the highest Find the alternative with the highest average outcome
average outcome
;
; Pick the outcome with the maximum Pick the outcome with the maximum number
number
;
; Assumes each state of nature is Assumes each state of nature is equally likely to occur
Uncertainty Example
Uncertainty Example
Kondisi alamiah Kondisi alamiah Pasar sesuai
Pasar sesuai Pasar tidakPasar tidak MaximumMaximum MinimumMinimum RowRow Alternatives
Alternatives HarapanHarapan sesuaisesuai in Rowin Row in Rowin Row AverageAverage Desain Desain UMPC UMPC $200,000$200,000 --$180,000$180,000 $200,000$200,000 --$180,000$180,000 $10,000$10,000 Desain Desain Tablet PC Tablet PC $100,000$100,000 --$20,000$20,000 $100,000$100,000 --$20,000$20,000 $40,000$40,000 Do nothing Do nothing $0$0 $0$0 $0$0 $0$0 $0$0 1.
1. MaximaxMaximax choice is to construct a choice is to construct a UMPC DesignUMPC Design 2.
2. MaximinMaximin choice is to do nothingchoice is to do nothing 3.
3. Equally likely choice is to construct a Equally likely choice is to construct a Tablet PCTablet PC
Maximax
Maximax MaximinMaximin Equally Equally likely likely
Beresiko
Beresiko
;
; Each possible state of nature has an Each possible state of nature has an assumed probability
assumed probability
;
; States of nature are mutually exclusiveStates of nature are mutually exclusive
;
; Probabilities must sum to 1Probabilities must sum to 1
;
; Determine the expected monetary value Determine the expected monetary value (EMV) for each alternative
Expected Monetary Value
Expected Monetary Value
EMV (Alternative i) =
EMV (Alternative i) = (Payoff of 1(Payoff of 1stst state of state of
nature) x (Probability of 1 nature) x (Probability of 1stst
state of nature) state of nature) +
+ (Payoff of 2(Payoff of 2ndnd state of state of
nature) x (Probability of 2 nature) x (Probability of 2ndnd
state of nature) state of nature) +…+
+…+ (Payoff of last state of (Payoff of last state of nature) x (Probability of nature) x (Probability of last state of nature)
EMV Example
EMV Example
1. 1. EMV(EMV(AA11) = (.5)($200,000) + (.5)() = (.5)($200,000) + (.5)(--$180,000) = $10,000$180,000) = $10,000 2. 2. EMV(EMV(AA22) = (.5)($100,000) + (.5)() = (.5)($100,000) + (.5)(--$20,000) = $40,000$20,000) = $40,000 3. 3. EMV(EMV(AA33) = (.5)($0) + (.5)($0) = $0) = (.5)($0) + (.5)($0) = $0 Kondisi Alamiah Kondisi Alamiah Pasar sesuaiPasar sesuai Pasar tidakPasar tidak Alternatives
Alternatives HarapanHarapan sesuai harapansesuai harapan Desain UMPC
Desain UMPC (A1)(A1) $200,000$200,000 --$180,000$180,000 Desain Tablet PC
Desain Tablet PC (A2)(A2) $100,000$100,000 --$20,000$20,000 Do nothing (A3) Do nothing (A3) $0$0 $0$0 Probabilities Probabilities .50.50 .50.50 Table A.3 Table A.3
EMV Example
EMV Example
1. 1. EMV(EMV(AA11) = (.5)($200,000) + (.5)() = (.5)($200,000) + (.5)(--$180,000) = $10,000$180,000) = $10,000 2. 2. EMV(EMV(AA22) = (.5)($100,000) + (.5)() = (.5)($100,000) + (.5)(--$20,000) = $40,000$20,000) = $40,000 3. 3. EMV(EMV(AA33) = (.5)($0) + (.5)($0) = $0) = (.5)($0) + (.5)($0) = $0 Kondisi Alamiah Kondisi Alamiah PasarPasar Pasar tidakPasar tidak Alternatives
Alternatives Sesuai Sesuai Harapan Sesuai HarapanHarapan Sesuai Harapan Desain UMPC
Desain UMPC (A1)(A1) $200,000$200,000 --$180,000$180,000 Desain Tablet PC
Desain Tablet PC (A2)(A2) $100,000$100,000 --$20,000$20,000 Do nothing (A3) Do nothing (A3) $0$0 $0$0 Probabilities Probabilities .50.50 .50.50 Best Option Table A.3 Table A.3
Kepastian
Kepastian
;
; Is the cost of perfect information
Is the cost of perfect information
worth it?
worth it?
;
; Determine the expected value of
Determine the expected value of
perfect information (EVPI)
Expected Value of
Expected Value of
Perfect Information
Perfect Information
EVPI is the difference between the payoff EVPI is the difference between the payoff under certainty and the payoff under risk under certainty and the payoff under risk
EVPI = EVPI = –– Expected value Expected value with perfect with perfect information information Maximum Maximum EMV EMV
Expected value with Expected value with perfect information perfect information (EVwPI)
(EVwPI)
=
= (Best outcome or consequence for 1(Best outcome or consequence for 1stst state state
of nature) x (Probability of 1
of nature) x (Probability of 1stst state of nature)state of nature)
+
+ Best outcome for 2Best outcome for 2ndnd state of nature) state of nature)
x (Probability of 2
x (Probability of 2ndnd state of nature)state of nature)
+
+ … + Best outcome for last state of nature) … + Best outcome for last state of nature) x (Probability of last state of nature)
EVPI Example
EVPI Example
1.
1. Hasil terbaik untuk kondisi alamiah Pasar Hasil terbaik untuk kondisi alamiah Pasar yang sesuai Harapan adalah Desain
yang sesuai Harapan adalah Desain UMPC dengan
UMPC dengan payoff payoff of of $200,000$200,000. . Hasil Hasil terbaik untuk Pasar yang Tidak sesuai terbaik untuk Pasar yang Tidak sesuai Harapan adalah “
Harapan adalah “do do nothing” nothing” dengandengan payoff of payoff of $0$0.. Expected value Expected value with perfect with perfect information information (EVwPI) (EVwPI) = ($200,000)(.50) + ($0)(.50) = $100,000 = ($200,000)(.50) + ($0)(.50) = $100,000
EVPI Example
EVPI Example
2.
2. MMaximumaximum EMV is EMV is $40,000$40,000, , yang adalah yang adalah hasil harapan terbaik tanpa informasi hasil harapan terbaik tanpa informasi sempurna. Sehingga
sempurna. Sehingga::
= $100,000
= $100,000 –– $40,000 = $60,000$40,000 = $60,000
EVPI = EVwPI
EVPI = EVwPI –– Maximum Maximum EMV
EMV
The most the company should pay for The most the company should pay for
perfect information is
Pohon Keputusan
Pohon Keputusan
;
; Information in decision tables can be Information in decision tables can be displayed as decision trees
displayed as decision trees
;
; A decision tree is a graphic display of the A decision tree is a graphic display of the decision process that indicates decision decision process that indicates decision alternatives, states of nature and their alternatives, states of nature and their respective probabilities, and payoffs for respective probabilities, and payoffs for each combination of decision alternative each combination of decision alternative and state of nature
and state of nature
;
; Appropriate for showing sequential Appropriate for showing sequential decisions
Decision Trees
Decision Trees
Pohon Keputusan
Pohon Keputusan
1.
1. Mendefinisikan MasalahMendefinisikan Masalah 2.
2. Menggambar Pohon KeputusanMenggambar Pohon Keputusan 3.
3. Menentukan Peluang bagi Kondisi Menentukan Peluang bagi Kondisi Alamiah
Alamiah 4.
4. Memperkirakan imbalan bagi setiap Memperkirakan imbalan bagi setiap kombinasi alternatif keputusan dan kombinasi alternatif keputusan dan kondisi alamiah yang mungkin
kondisi alamiah yang mungkin 5.
5. Menyelesaikan permasalahan dengan Menyelesaikan permasalahan dengan mengerjakan dari belakang ke depan mengerjakan dari belakang ke depan melalui perhitungan EMV untuk masing melalui perhitungan EMV untuk masing--masing titik kondisi alamiah.
Decision Tree Example
Decision Tree Example
= (.5)($200,000) + (.5)(
= (.5)($200,000) + (.5)(--$180,000)$180,000)
EMV for node 1
= $10,000
EMV for node 2
= $40,000 = (.5)($100,000) + (.5)(= (.5)($100,000) + (.5)(--$20,000)$20,000) Payoffs Payoffs $200,000 $200,000 --$180,000$180,000 $100,000 $100,000 --$20,000$20,000 $0 $0 Desain Desain Tablet PC Tablet PC
Pasar sesuai harapan Pasar sesuai harapan (.5)(.5)
Pasar tidak sesuai Pasar tidak sesuai (.5)(.5)
1
Pasar sesuai harapan Pasar sesuai harapan (.5)(.5)
Pasar tidak sesuai Pasar tidak sesuai (.5)(.5)
2
Figure A.2 Figure A.2
Complex
Complex
Decision
Decision
Tree
Tree
Example
Example
Figure A.3 Figure A.3Complex Example
Complex Example
1.
1. Given favorable survey resultsGiven favorable survey results
EMV(2) = (.78)($190,000) + (.22)(
EMV(2) = (.78)($190,000) + (.22)(--$190,000) = $106,400$190,000) = $106,400 EMV(3) = (.78)($90,000) + (.22)(
EMV(3) = (.78)($90,000) + (.22)(--$30,000) = $63,600$30,000) = $63,600
The EMV for no plant
The EMV for no plant = = --$10,000$10,000 so, so, if the survey results are favorable, if the survey results are favorable, build the large plant
Complex Example
Complex Example
2.
2. Given negative survey resultsGiven negative survey results
EMV(4) = (.27)($190,000) + (.73)(
EMV(4) = (.27)($190,000) + (.73)(--$190,000) = $190,000) = --$87,400$87,400 EMV(5) = (.27)($90,000) + (.73)(
EMV(5) = (.27)($90,000) + (.73)(--$30,000) = $2,400$30,000) = $2,400
The EMV for no plant
The EMV for no plant = = --$10,000$10,000 so, so, if the survey results are negative, if the survey results are negative, build the small plant
Complex Example
Complex Example
3.
3. Compute the expected value of the Compute the expected value of the market survey
market survey
EMV(1) = (.45)($106,400) + (.55)($2,400) = $49,200 EMV(1) = (.45)($106,400) + (.55)($2,400) = $49,200
The EMV for no plant
The EMV for no plant = $0= $0 so, given so, given no survey, build the small plant
no survey, build the small plant
4.
4. If the market survey is not conductedIf the market survey is not conducted
EMV(6) = (.5)($200,000) + (.5)(
EMV(6) = (.5)($200,000) + (.5)(--$180,000) = $10,000$180,000) = $10,000 EMV(7) = (.5)($100,000) + (.5)(
The end
The end
Pokok Bahasan Selanjutnya: Pokok Bahasan Selanjutnya:
Teknik Peramalan Teknik Peramalan (F O R E C A S T I N G) (F O R E C A S T I N G)