Minggu 5:

Decision Making and

Information

TI4002-Manajemen Rekayasa Industri

Teknik Industri, FTI

ITB
 FUNGSI MANAJEMEN

Tahapan paling penting

PERENCANAAN

PENGAMBILAN KEPUTUSAN

PENGORGANISASIAN

KEPEMIMPINAN

PENGENDALIAN (Morse & Babcock, 2007)

Minggu 4 – Perencanaan & Peramalan

TI4002 Manajemen Rekayasa Industri 2
 Decision Making
• Key first steps of setting objectives on planning
stage
• Process of making a conscious choice between
two or ore rational alternatives in order to select
the one that will produce more desirable
consequences (benefits) relative to unwanted
consequences (costs)

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

3
 Decision Making
• Type of decisions : routine and non-routine
– Routine decisions can be delegated to lower levels
(within policy limits)
– Non-routine → unstructured situation, incomplete
knowledge, high uncertainty, subjective judgments or
even intuition → goes to higher management
• Objectives and bounded rationality
• Level of certainty

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

4
 Management Science
Characteristics
• A system view of the problem
• The team approach
• An emphasis on the use of formal mathematical
model

System Engineering is an interdisciplinary approach and means

to enable the realization of successful systems. It focuses of
defining customer needs and required functionality early in the
development cycle, documenting requirements, then proceeding
with design synthesis and system validation while considering
the complete problem
TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen
5
 Model and Their Analysis
• Model is abstraction or simplification
• Most models of management science are
mathematical models:
Net income = revenue – expenses - taxes
Real World Simulated (Model) World

1. Formulate the problem

2. Construct a mathematical model.

3. Test the model

4. Derive a solution form model

5. Apply the model solution

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

6
 Tools for Decision Making (DM)
• Three categories :
– under certainty
– under risk
– under uncertainty
• Payoff Table

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

7
 Payoff Table
State of Nature / Probability
N1 N2 ... Nj ... Nn
Alternative P1 P2 ... Pj ... Pn
A1 O11 O12 .. O1j … Oin
A2 O21 O22 .. O2j … O2n
.. .. .. .. .. .. ..
Ai Oi1 Oi2 .. Oij … Oin
…
Am Om1 Om2 .. Omj … Oim

N = nature O=Outcomes P=Probability A: Alternative

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

8
 Decision under Certainty
• Probability P1=1 and all others = 0
• Linear Programming
– State the problem
– What is decision variables?
– Objective function?
– Constraints

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

9
 Example
• Poligon consider to produce two types of bike : BMX dan
MountBike. They estimate profit margin BMX $10/unit,
and MountBike $14/b.
• How many product mix consists of BMX and MountBike
should be produces in order to maximize total profits?
• Now they have 3 production lines dan 2 assembly lines
operated 40 hours/week
• 1 unit BMX requires 3 hours on production line and 1
hour on assembly line.
• 1 unit Montbike requires 2 hours on production line and
2 hour on assembly line.

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

10
 Model Formulation
• State the problem : how many unit BMX and
MontBike to produce to maximize profit?
• Decision variables:
– Let x = number of product BMC to produce/week
– Let y= number of product MontBike to produce/week
• Objective function:
Maximize P = 10x + 14y
• Constraints
3x + 2y ≤ 120 (hours of machining time)
x + 2 y ≤ 80 (hours of assembly time)
x ≥ 0 and y ≥ 0 (number of production)

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

11
 IsoProfit lines

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

12
 Graph Solution

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

13
 Decision Making Under Risk
• Nature of risk :
– there exist a number of possible future states of
nature (Nj).
– Each Nj has a known probability (pj)
– Each alternative (Ai) has a known outcomes for each
future states (Oij)
• Expected Value
𝑛

𝐸𝑖 = ෍ 𝑝𝑗 𝑂𝑖𝑗
𝑗=1

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

14
 Well Drilling Example

State of Nature (Probability)

N1 N2 N3 Expected
Alternative
Dry Hole Small Well Big Well Value
p1=0.6 p1=0.3 p1=0.1
A1: Don’t drill $0 $0 $0 $0
A2: Drill alone -500,000 300,000 9,300,000 720,000
A3: Farm out 0 125,000 1,2500,000 162,500

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

15
 Decision Tree
• Another technique for finding expected value
• Begin with a single decision node (by square or
rectangle), from which a number of decision
alternatives radiates
• Each alternative ends in chance node (by circle)

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

16
 Example : Asuransi
• Manajer keamanan sedang mempertimbangkan
keputusan untuk mengambil atau tidak asuransi
kebakaran dari asset senilai US$ 100.000.
Berikut data-data yang tersedia
– Biaya Asuransi kebakaran US$ 200. Peursahaan
asuran akan memberikan biaya pengganti sebesar
US$ 99.800
– Diperkirakan kemungkinan terjadi kebakaran hanya
0,1%
• Keputusan mana yang harus diambil?

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

17
 Solusi- decision tree
– don’t insure

Decision Chance (Outcome) (Probability) (Expencted Value)

Node Ai Node Ai Oij x =
Pj Ei

No fire (-200) x (0.999) = -199.8

Insure + = $ -200
Fire (-200) x (0.001) = -0.2

No fire (0) x (0.999) = 0

Don’t + = $ -100
Insure
Fire (-100,000) x (0.001) = -100

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

18
 Risk as Variance
• Risk is variability of outcomes, measured by the
variance (vx) or (more often) its square root, the
standard deviation (x)
𝑛
2
𝑉𝑥 = ෍ 𝑝𝑗 𝑂𝑗 − 𝐸(𝑋)
𝑗=1

𝜎𝑥 = 𝑉𝑥

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

19
 Example : Project Risk
Project X Project Y
Probability Profit ($) Probability Profit ($)
0,1 3.000 0,1 2.000
0,2 3.500 0,25 3.000
0,4 4.000 0,3 4.000
0,2 4.500 0,25 5.000
0,1 5.000 0,1 6.000
E(X)=0,1(3.000)+0,2(3.500)+0,4(4.000)+0,2(4.500)+0,1(5.000) = $4.000

E(Y)=0,1(2.000)+0,25(3.000)+0,3(4.000)+0,25(5.000)+0,1(6.000) = $4.000

VX=0,1(3000-4000)2+0,2(3500-4000)2+0,4(4.000-4000)2+0,2(4500-4000)2
+0,1(5.000-4000)2 = $300.000
VY=0,1(2000-4000)2+0,25(3000-4000)2+0,3(4.000-4000)2+0,25(5000-4000)2
+0,1(6.000-4000)2 = $1.300.000

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

20
 Proyek Y lebih berisiko 𝜎𝑥 = $548
dari pada proyek X

𝜎𝑦 = $1,140

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

21
 Decision under uncertainty
Different approach :
• The optimistic decision maker (Risk Taker)→ highest
possible outcome (the maximax solution)
• The pessimist decision maker (Risk averse)→
highest of worse outcome (the maximin solution)
• Hurwicz → somewhere between optimist and
pessimist
• Equally likely
• Smallest difference between best and worst
outcomes (minimax regret)

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

22
Example : DC under uncertainty
Outcome

Maximum Minimum Hurwicz Equally

Alternative
(N3) (N2) (N1) (=0.2) Likely
p3=0,1 p2=0,3 p1=0,1
A2 $9,300,000 $300,000 $ - 500,000 $1,460,000 $3,033,333
A3 1,250,000 125,000 0 250,000 458,333

Hurwicz :  (best outcome) + (1-)(Worst outcome) Alternative A2

σ𝑁
𝑗=1 𝑂𝑢𝑡𝑐𝑜𝑚𝑒 − 𝑗
𝐸𝑞𝑢𝑎𝑙𝑙𝑦 𝐿𝑖𝑘𝑒𝑙𝑦 = Alternative A2
𝑁
Optimist Alternative A2 Pessimist Alternative A3

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

23
 Example :
Minimum Maximum Regret (A2)
State of Nature (Probability)
Alternatives N1 : Dry Hole N2 : Small Well N3 : Big Well
p1=0.6 p1=0.3 p1=0.1
A1: Don’t drill $0 $0 $0
A2: Drill alone -500,000 300,000 9,300,000
A3: Farm out 0 125,000 1,250,000

Regret Maximum
Alternatives
Dry Hole Small Well Big well Regret
A1: Don’t drill $0 $300,000 $9,300,000 $9,300,000
A2 : Drill alone 500,000 0 0 500,000
A3 : Farm out 0 175,000 8,050,000 8,050,000

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

24
 Computer-based Information
Systems
• Integrated Database
– Laser scan
– Hospital
• Management Information System (MIS)
– Structured
• Decision Support System (DSS)
– Semi and Unstructured

TI4002 Manajemen Rekayasa Industri Minggu 1 - Engineering dan Manajemen

25