Lecture 2 –

Understanding Dynamics of Systems

IE2141 Systems Thinking and Dynamics

Dr. LI Haobin, Senior Lecturer

Department of Industrial Systems Engineering and Management (ISEM)
College of Design and Engineering
National University of Singapore

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore
 2.1 – Introduction to
Behavior-Over-Time Graphs (BOTGs)

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 2
 What is BOTG?
▪ A Behaviour over Time Graph (BOTG) is a simple line graph that shows a
pattern of change over time – it shows how something increases and
decreases as time passes.
▪ This foundational Systems Tool is instrumental in modeling and understanding many
systems.
▪ BOTGs are designed to represent our thinking. All BOTG graphs allow our 'mental
models' to take a visual form so we can share them, or analyze them ourselves.

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 3
 -allows to
visually observe
how beharron of system changes

What is BOTG? over time in an intrite manner

▪ BOTGs are plots of one or more variables (y axis) over time (x axis).

Value

Variable
of

->
IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore Time 4
 Why BOTG? Identity & understand
d patterns trends
▪ Systems thinkers often create BOTG as an initial step to understand
a complex system.
▪ BOTGs are typically constructed early in the planning phases of a
research study, project, or program to Discover periodic variations in besystem
change in speed of hends
-

▪ formulate a research question, -Potential key time

▪ understand the problem of interest, points
Bedat Lutelbekender
▪ draw out participants’ mental models, and
▪ generate and compare hypotheses about key determinants of the problem
and strategies for action.

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 5
 Ignasmaltine Changing numeics over time
infoutput of
* keytrends Umajartrends" -
input&
output
->
dude
dkwhich-into them

Drawing BOTGs
dem und rate?
ver of
system certain variables that varis
represent emeral elementskey subsystems demand-> unmet
need entthe for prob
& demand
retranceof I cer Especif definitions Fret
key system vor slight variation
metfor shaft
picture of systemissal May
be
lost
E fer each demand ?
org-> LEE , system
Structure mightchang of
I
indefinition key system ver.
& Time - 1. Select state variables completely
I

Seal
2. Choose timeframe diff boths
assfrends& generated

patterson 3. Identify general trends over time, and

difffihl
4. Identify the key factors they
front
hypothesize are most responsible for
driving those trends over time
Remarks: In IE2141 assignments / project,
▪ All state variables selected for BOTGs shall be well defined; Source: CLExchange
▪ All BOTGs shall have clearly specified time frame (scale on x-axis);
▪ Numerical scales on y-axis are NOT required as BOTGs shows only the general trend;
▪ All general trends shown shall be explained by hypothesis, whereas fluctuations other than the general
trends shall be neglected from BOTGs.
(Example to be shown in Tutorial 2)
IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 6
 Scat, 14 major trends

2.2 – General Trends for

Behaviour-Over-Time

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 7
 Steady State Behaviour
▪ A system or a process is in a steady state if
the variables (called state variables) which

Performance
define the behavior of the system or the
process are unchanging in time.
▪ In continuous time, this means that for those
properties of the system, the partial derivative
with respect to time is zero and remains so.

Time
Behaviour I Constant Still
but
undergoing changes ,

majerbends" remain steadily

state,
IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 8
 Not commonly
I obsened
Steady State Behaviour usually result of
active intervention
stakeholders
by system's key
▪ Example:
values
num
reflected Grad

Performance
H
target
A
sen
values

Time
weed elsehigter
(competen Abacost prod
Unemployment Rate in Singapore (Seasonally Adjusted) ame
↳can be result ofsuccessful gat, interventions < cannottl ↳diff
IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore
pougul foresidents) 9
 Constant Behaviour (Increasing)
▪ The state variable increases at a
constant rate.
▪ In continuous time, this means that for

Performance
those properties of the system, the
partial derivative with respect to time is a
positive constant and remains so.

Time

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 10
 Kose association Steady state betarian
Constant Behaviour (Increasing)
-Establishment of vace Petres in
early 201
.

I
▪ Example:

Performance
Time

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 11
 Constant Behaviour (Decreasing)
▪ The state variable decreases at a
constant rate.
▪ In continuous time, this means that for

Performance
those properties of the system, the
partial derivative with respect to time is a
negative constant and remains so.

Time

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 12
 Constant Behaviour (Decreasing) Org .
ofpetroleum
exported counties
intervened by OPEC &
-
highly regulated , gots
ad
Bod reduction targets established , actual prod
-

▪ Example:
rate until red target
actively charges& constant
,

↑ is achieved , regardless of other factors

Saudi Arabia Oil Production -

Performance
Time

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 13
 Exponential Behaviour (Growth)
Difffam
Polynomial Quad Diff underlying
system dynamic
▪ Exponential growth occurs when the mechanisms for growth

instantaneous rate of change of an increasing

quantity with respect to time is proportional to

Performance
the quantity itself. (J-Shaped Curve)
▪ The quantity is an exponential function of time,
that is, the variable representing time is the
exponent (in contrast to other types of growth,
such as quadratic growth).

𝑑𝑁
= 𝜆𝑁 𝑁 𝑡 = 𝑁0 𝑒 𝜆𝑡 Time
𝑑𝑡

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 14
 Initial) Endy stage rate ofnew entrations
-
- dir proportional to nemble
infections

Exponential Behaviour (Growth)

of
existing
ten
Butone got health measures
reacted,
certain level of head immunity
a. ofnfected indiichals inst rateof ver infections
no
larger proportina
,
Rob of infection
to no of
easting infections
infected also mess
.

▪ Example: among thosenot Growth derite fam exponential

.
-

Performance
Time

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 15
 reverse of expo growth .

Exponential Behaviour (Decay)

▪ A quantity is subject to exponential decay if it
decreases at a rate proportional to its current
value.

Performance
▪ Symbolically, this process can be expressed
by the following differential equation, where
𝑁 is the quantity and 𝜆 is a positive rate called
the exponential decay constant:

𝑑𝑁
= −𝜆𝑁 𝑁 𝑡 = 𝑁0 𝑒 −𝜆𝑡 Time
𝑑𝑡

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 16
 Exponential Behaviour (Decay)
-
No , of students admitted is according fodits trends
-
enestran ofdisciplite , ederational reputation , socialultr logeti factors
Graditions
▪ Example:
,

World-senfinnaus change Otheremerging disapplies

.

whe engineering etc capture other plattention trad. disciples

,

,
whe
theology fall proportional decline in attention & interest
Number of yerafter gar

Performance
Seminarians in Seattle
-

study to bleame a

priest

Time

1965 2015

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 17
 reverself exp growth along - asis

Exponential Behaviour (Collapse)

▪ Exponential collapse is the negative
process of exponential growth, where
the instantaneous rate of decreasing

Performance
is proportional to the decrement.
▪ Exponential Collapse is not
Exponential Decay

𝑑𝐷
= 𝜆𝐷 𝐷 𝑡 = 𝐷0 𝑒 𝜆𝑡 𝑁 𝑡 = 𝑁0 − 𝐷0 𝑒 𝜆𝑡 Time
𝑑𝑡

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 18
 -Rapidl deasetrad
- Need suftresponse

Exponential Behaviour (Collapse) -

↳o
Garcial to find solutions
launter salads b4 facing
Declining recep -> melted antrecup forms greater danger
blue Mowed seawater , absorbing
miEsdarkat , generate worth
▪ Example: Lutter accelerating melting of
,

escap

Performance
Time

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 19
 Goal Seeking Behaviour (Increasing)
▪ The quantity of the state variable starts
below a goal level and over time

Performance
moves toward the goal.
▪ In increase slower while the quantity is
approaching the goal.
trend where difference between currentstate He target

exhibits
"expordecary" Pattern engine system goo exists Time

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 20
 Goal Seeking Behaviour (Increasing)
-

Achieving maintaining anecessary

- Once
pop ,
approaches target value ,
of St
Rep .

population ntal for autering sustained

tighten immigration policies fo show
rateof pop
▪ Example: eson. development
grante to entere
aut socio-econ
:

.
refaces

cannot follow expograntie

-

other bigger countries

Performance
Time

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 21
 Goal Seeking Behaviour (Decreasing)
▪ The quantity of the state variable starts
above a goal level and over time

Performance
moves toward the goal.
▪ It decreases slower while the quantity
is approaching the goal.

Time

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 22
 Goal Seeking Behaviour (Decreasing)
Apollutione Significant issue plaguing Maltriaised nations
atte

i
Appropriat target unz
▪ Example: Air pollution in UK holonger hard Signate
cont impact
on human health
measures d
Ongoing

Performance
emironment
improve ,
zemissionat
particles gradually
decease
harmful
based on level of improvement
already actiered
target perhaps techold
fersafehuman expa
E to RM2 Sessions
. Time

https://www.gov.uk/government/statistics/emissions-of-air-
pollutants/emissions-of-air-pollutants-in-the-uk-particulate-
matter-pm10-and-pm25

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 23
 ·ht in information & controlR citation
-
-

Oscillation Behaviour (Constant)

cause of behaviour is not alternating change intarget value ,

but rather He presence of a certain some delay obtaining

in

deviation information
▪ The quantity of the state variable target feedback of I within a
to adjust
feedback control system as well efforts made
as

fluctuates around a certain level, effects He dev or

generate
.

and the amplitude is a constant targetvalue

Performance
I
E
with respect to time.
limitof amplitude
When value of inceasing Hand approach upper ,

that
seems to be a force within the system
te adjust
value back to lower limit of amplitude
Time

·Oueoscillation temraiable
perform
ene

a behanbur achrered effetched

Herdeslas
IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore Lanierebal 24
 Center value of Oscilla
many coexist ~another major
Oscillation Behaviour (Constant) trend
&
Elen cycle Bon mean target value
stays constant,
but amphaude range ofstill. is
constant
▪ Example: => 1

US Economy

Performance
Time
Gre exhibit constant os all be hander,
E -

superimposed anbehander of
IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore
exponential growth 25
 Oscillation Behaviour (Damped)
▪ The quantity of the state variable
fluctuates around a certain level,
and the amplitude is decreasing

Performance
value
with respect to time. atarget ,

mentee
a

henhavedelagmastersadjustmentbehendent e reduce
↳thintte system are rapid righ , system can gradually
tese deviations -> decreasing amp .
Time

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 26
 Oscillation Behaviour (Damped)
Leadingmechanismal owed e thusfarung each ane
e
feedback unthin tesystem
▪ Example: amp of zad
info I vert

-
accumulate
fomware/ Significantly
↑ narrower

Performance
Breakout of
COVID-19 in US

I
2nd Wave

I Time

impart ot

pandeme & He Year 2020 new equi

neuequr .
port fr He exchange rats uncertain point after
IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 3 mths 27
 Oscillation Behaviour (Expanding)
▪ The quantity of the state variable
fluctuates around a certain level,
and the amplitude is increasing

Performance
with respect to time.
feedback &
Additional delungs in internal information
Control operations enzsystem
Greed more time to correct Hederations
variable derates Lutter framtarget) Time

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 28
 Lack of trading experience &
Oscillation Behaviour (Expanding) sentering historical knowledge , teacher
made the market
signficant pure Huc twin
Attract me made Oftention req
various e mortime frsystem to
.
, -
into channels reportant feedbaction
digest prode information
▪ Example: price deviation

Performance
Time

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 29
 S-Shaped Growth seeking
endpert" increasing goo keer

▪ A pattern of growth in which, the quantity ~

reacch until
increases slowly initially, in a positive instabilises , growtht O
acceleration phase; then increases rapidly,
approaching an exponential growth rate; but ↑

Performance
o
then declines in a negative acceleration phase
until at zero growth rate the quantity stabilizes.
▪ This decline reflects increasing environmental
resistance which becomes proportionately
more important at higher quantity. becomes ligte
eni Constrants become more
.

significantas gty ↓ Time

18tpost
expogromth
IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 30
 (1985-2003) How of goods -> cate con garth dining dem and
for goods movement
S-Shaped Growth 1009-201519Even difficulties presented batters to Ster

movement
goods
growth in throughput continue
⑧

2008 financial crisis ,

▪ Example: notably but

after
rate stow
down
approaching target
I

growth
↑
to
grow ,

per limit

Performance
Time

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 31
 S-Shaped Growth with Overshoot
system Pitrete ,

▪ S-shaped growth combined with andphase delay i

Ist port system growth doesn
oscillation whose amplitude expo ↓ gradually
slow

down orstabilit

⑳0
gradually declines over time.

Performance
↳
reaching
targetvalue

Time

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 32
 I Schepegrowth is -> growth
inst phase
dealy exced target threshold
↑Dis .
bet His
.

if subseamentosell Noticeable

->
.

S-Shaped Growth with Overshoot

Japan export value &GDP
▪ Example:
Japan Economy

Performance
Time euibate
lower Harold
-
oscillationt , grad approach .

expogrant
IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore
↳ 2008 francial visis 33
 Overshoot and Collapse
aukly surpasses
▪ Overshoot and collapse is a Behaviour max
freshold
R
system
of delay
can reach control
aut
in which the growth of a quantity relies wee
infeedback
upon a depletable resource.
seeten

Performance
▪ When the resource is abundant, the

08
quantity grows quickly, which depletes
the resource. ultimately
wirchfatter
reading orgoing
to overuse -> accelerates port expocollapse
resources decline
system
of
↑ Terefathe
rapoddechte
intarget threshold
consequence of rapidly
exceeding threshold
ot He internal ↳
a part
for
responsible
,
gronthedibted bydepletable
diesoncertain
Resystem

resources within the E

OS andplatene
patience to
Time
↓
stil bes
resources
maintaing target equr value
to

has been
severely depleted &
system ~8 for damped fr equilibrium
values backto threshold range
IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 34
INRRECOVERABLE system
 morepublicity , - mistakenly believe smoking
enigorating & metaltberefts
ofsmoker grow
no ,
expo,
Overshoot and Collapse ↑
Commercial
I
operationsI Applesiensmoking

Y
of tabuccomenket
as fashbrabte
I

▪ Example: consumption behande

"
e

Performance
Milangane Consequently ste
decinetobac
ratenearly prop .
Consumption after
1970S
fore of,

smokers ,
- ↑
ehisception
inversible
althou time ->

un
lagifford ↳
- lungcancer Association bet Time
25
smoking fundamental resources sustaining growth
of
erected autof offabul consumption , te confidence
I
He mertt fortobales
drastic mercase
in ling
IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore
mortality , lead
cancer depletion of 35
 General Trends for BOTGs

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 36
 2.3 – Classification of Systems

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 37
 Classification of Systems
▪ Classification by system characteristics
▪ Static vs. Dynamic Systems
▪ Causal vs. Non-Causal Systems
▪ Time-Variant vs. Time-Invariant Systems
▪ Linear vs. Non-Linear Systems
▪ Invertible vs. Non-Invertible Systems
▪ Stable vs. Unstable Systems

Embrofmu fage
Ditt *
Defen of system net
include multiple inputs & outputs
inte diff categories
IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 38
 value of system's outputa
time only determined
any given
at that some
by iput value
Static vs. Dynamic Systems ?! time point without
-
any intere

-warg
frompost/future inputs
-unchers that stationary remain
↑
▪ Static System – output of system depends only on present values of input
▪ Memoryless system
▪ Dynamic System – output of system depends on past or future values of
X input at any instant of time
-

contributeto present
not Stati
(system develop memory
system ▪ System with memory ↳ form de history)
Example of static systems:
𝑌 𝑡 = 𝑋 𝑡 + 3, 𝑌 𝑡 = 2𝑋 𝑡
𝑋 𝑡−𝛿
Example of dynamic systems:
𝑋 𝑡 𝑌 𝑡
SYSTEM 𝑌 𝑡 = 𝑋 𝑡 − 1 , 𝑌 𝑡 = 3𝑋 𝑡 + 2 ,
𝑋 𝑡+𝛿 𝑌 𝑡 = 𝑋 𝑡 + 2𝑋 𝑡 − 1

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 39
 noterery deposit &same the
Stat

Static vs. Dynamicy

Statch
Systems &
Currentprincipal input
↑
intral depantlimput)
Multipletstara total principal +Coutput]
aput) ants (outputs

E
current interest

dynamic deposited interest trend

,
S
expected& the tie

-↑
▪ Examples Total principal
when intral deposit made
contputt
U

interest&any giventime ↑
▪ Static systems – furniture, dishes, buildings, bridges, fix deposit, one- -

time investment
(simplification, approximation or abstraction of real-world dynamic systems)
▪ Dynamic systems – human body, computer, machinery, car, property,
trading strategy

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 40
 Bu dynamicsystem can be causal non-cans al
or

Causal vs. Non-Causal Systems

uzrecessarily involve solely determinedby past present inputs (future N
dependence on future inputs -output
↑
▪ Causal System – output of system is independent of future values of input
▪ All real-life system, all practical or physically realizable systems are causal systems
▪ Non-Causal System – output of system depends on future values of input at any
alw4 instant of time Untachirabtel
dynamic ▪ Anti-causal system – output of system only depends on future values of the input implementable
auzof Cinelependentof past historical inputs in reality
belong t
category Imitations imposed
Static system of caus al
cuz
systems Example of cause systems: &
beyourtland spall
no
reliance
on future
Mputt 𝑌 𝑡 = 𝑋 𝑡 + 3, 𝑌 𝑡 = 2𝑋 𝑡 + 𝑋 𝑡 − 1
𝑋 𝑡−𝛿
Example of non-causal systems:
𝑋 𝑡 𝑌 𝑡
SYSTEM 𝑌 𝑡 = 𝑋 𝑡 − 1 + 2𝑋 𝑡 + 3𝑋 𝑡 + 2
𝑋 𝑡+𝛿
Example of anti-causal systems:
𝑌 𝑡 = 2𝑋 𝑡 + 1
IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 41
 Causal vs. Non-Causal Systems
▪ Examples
▪ Causal systems – furniture, dishes, buildings, bridges, fix deposit,
one-time investment, human body, computer, machinery, car, property,
trading strategy
▪ Non-causal systems – an ideal predictive maintenance system, a
perfect just-in-time system
(not practical or implementable in real-life)
production
current plans detemmed
based on futuresales
deta as inputs
A
Just-In-Time
&reliability
IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore
Predictive Maintenance 42
conditions
Baseden inputs regarding future Lauren
 yes
dates
espects ,
miths

Time-Variant vs. Time-Invariant Systems

▪ Time-Variant (TV) System – a system whose output response depends on
moment of observation as well as moment of input signal application.
▪ In other words, a time delay or time advance of input not only shifts the output signal
in time but also changes other parameters and behavior.
▪ Time variant systems respond differently to the same input at different times. for both scenarios
,

Ifoutputsame
▪ Time-Invariant (TIV) System – a system where the opposite is true for. true invariant
-

outputdiff,
die variant system

u
′
𝑋 𝑡 𝑌 𝑡 𝑌 𝑡 = 𝑌 𝑡 − 𝑡0
SYSTEM Delay by 𝑡0
𝑌 ′′ 𝑡 ≠ 𝑌 ′ 𝑡
𝑋 ′ 𝑡 = 𝑋 𝑡 − 𝑡0
Delay by 𝑡0 SYSTEM
𝑌 ′′ 𝑡 = 𝑌 ′ 𝑡
IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore
- 43
 might not ceviston t not aptuld -
Some
say the
variant

many offer inputs Reality diff fo catch everything

Time-Variant vs. Time-Invariant Systems

regardlesshenapplied a
input I The
wery in which syste system responds to same

C
respond to diff inputs changes as time progresses
frame input doesnt changeover fire L

▪ Examples -> D respondence

system
& which iput
▪ Time-variant (TV) systems – investment in stocks -provided
-Buy & difftines
-
▪ Time-invariant (TIV) systems – investment in CPF Accounts might red diff
(TIV systems are relative in real-life, e.g., reaching age of 55) results
↓ as long as 8tr
wo matter
-endeposit samedatee
interest consistent

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 44
 Linear vs. Non-Linear Systems
▪ Linear System – a system which follows the principle of superposition
▪ Law of Additivity + Law of Homogeneity
▪ Non-Linear System – a system for which the principle of superposition is
violated.

𝑋1 𝑡 𝑌1 𝑡
SYSTEM
𝑋1 𝑡 + 𝑋2 𝑡 𝑌 ′′ 𝑡 = 𝑌 ′ 𝑡 𝑌 ′ = 𝑌1 𝑡 + 𝑌2 𝑡
∑ SYSTEM ∑
𝑌 ′′ 𝑡 ≠ 𝑌 ′ 𝑡
they obey

↑
SYSTEM Law of Additivity
𝑋2 𝑡 𝑌2 𝑡
Bowde 2diff inputs to the system
-
-

Robserve 2diff sets of output & observed

add thes 2 sets of inputtgt provide 45
output ,
IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore If
out put is sumot
previously
-

Hem to be system , resulting

&
 Linear vs. Non-Linear Systems
expect effectiveness of system's outputs can
increase/decrease to the inputs proportionally
Al
▪ Linear System – a system which follows the principle of superposition
satisfy both-
▪ Law of Additivity + Law of Homogeneity
▪ Non-Linear System – a system for which the principle of superposition is
↓ violated. Select I set of input obtain
set outputs
corresponding
-

of
Either of
-Choose coeff K . & scate inputs to system up/downby
a factor of k
the flaws also scated by the factor of k compared
It
resulting output
satisfied
-

𝑋 𝑡 ′
𝑌 𝑡 𝑌 = 𝑘𝑌 𝑡

obes
not ↓
Meorigina out
t , system
SYSTEM 𝑘

𝑘𝑋 𝑡 𝑌 ′′ 𝑡 = 𝑌 ′ 𝑡
𝑘 SYSTEM
𝑌 ′′ 𝑡 ≠ 𝑌 ′ 𝑡
Law of Homogeneity

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 46
 total bingthe
Realty -> most complex systems nothier
ofpleasing
System ->
to
no ,

-
-

Linear vs. Non-Linear Systemsfexed

limputty
stowthe
2 .

9 ·
Aty of oranges bought mea seats
I
-

presold
prepard(output) total worthing timeforthose ->) lin.
related to no .
of Ppk

▪ Examples who alrbung be fareusing

↑
wring
↑
▪ Linear systems – pricing for groceries at FairPrice, total time spent by a in

crowd watching a movie dining party limited whingfine franstomers

stone serve I backofline keeps increasing
,
- -
-
↑

▪ Non-linear systems – pricing for stocks at SGX, total time spent by a

crowd queueing for a restaurant Fer buying mar
, char pul will t - donline
(real-life systems are difficult to control as many of them are non-linear)

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 47
 dednes what inputs system and
Seeoutput , abteto
Invertible vs. Non-Invertible Systems

S
↳ multipkinputs many
lead to same output
▪ For an invertible system, there should be one to one mapping
between input and output at each and every instant of time
Invertible
of districtinputs One to one mapping Many to one mapping
radto district
outputs ,or of 1 𝑎 2
𝑎
imeesystem 2 𝑏 4
ext

3 𝑐 6 𝑏

𝑋 𝑡 𝑌 𝑡 𝑋 𝑡 𝑌 𝑡
SYSTEM SYSTEM

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 48
 Invertible vs. Non-Invertible Systems
▪ For an invertible system, there should be one to one mapping
between input and output at each and every instant of time indula
understand only need observe respondence bet system's import puts - learn
ul
out
invet Dunnid deeply .
t

essesstee
t

SeniputtSandHeoptwasof a idea
↳ ma
ExperimentadjustdiffSambhatta
ate the
wonist
-
the inputs given
desired output

𝑋 𝑡 Invertible 𝑌 𝑡 Inverse 𝑌′ 𝑡 = 𝑋 𝑡 ± 𝛿

System System 𝑌′ 𝑡 ≠ 𝑋 𝑡 ± 𝛿

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 49
 Invertible vs. Non-Invertible Systems
▪ Examples
▪ Invertible systems – identifying a person by his/her IC, knowing a
person by his/her spouse, sending emails by an internet user
▪ Non-invertible systems – paying bills, scoring in an exam, sending
Samedr
emails by a hacker ↳outputotbellingamt cannot
Grading
WI cannot reflect
, ,

drectly traceimputof items purchased detauted learning achievements ofeffort

M a or
student

(real-life systems are complex as many of them are non-invertible) Rut they

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 50
 Stable vs. Unstable Systems
▪ For a stable system, output should be bounded for bounded
input (BIBO) at each and every instant of time
Centre imputvalues within certain
range , observe system output
2 Whethercan control system imputrange
for system output
Strandstarte spectrange to ensure output remains within a

di 𝑋 𝑡 𝑌 𝑡 manageablerange
SYSTEM
Stablesystem
𝑋 𝑡 ∈ 𝐿𝑋 , 𝑈 𝑋 - 𝑌 𝑡 ± 𝛿 ∈ 𝐿𝑌 , 𝑈 𝑌

𝑌 𝑡 ± 𝛿 ∈ −∞, ∞
does not
solly intert systems
where output tends to approach either
Dor-

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 51
 Sustainable development
I Any instability
withoutre Gowth - Artohaness
infinite
opportunities
I
Controllable conditions of- Bices war
tr returns
growth
of mature system expect continuouslyrise
supply & demand
, system becomes aconcern system - limited system iputs
it tobe stable
↓

Stable vs. Unstable Systemsdemand &

& production capacity
But itexpand boundaries
a.

g buyer
.
&seperactions
socio-econ falters,
though pulls finetrafe, wer Rotter
.

even customer
salessystems L

speerfe range whr

demand & supply
remains within certain range, pres can be controlled
within a
a
aty anno larger be

↑
controlled as inputs ,

▪ Examples may become unstable

system

▪ Stable systems – market of iPhone 13, restaurant, public transit

notalways
negativel ▪ Unstable systems – climate change, financial crisis
undesirable

E
(unstable systems are relative, as the output can be always bounded under
a larger system, however it is beyond our control or not in favor)
of imports
accurately determine set
to control
cannot
dimate within manageable range
I

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 52
 fr unlimited
*

Recognise attributes of system & diffstages i Identity probssystem ,

Startup _ wi
growth opportunities
Ropose appropriate solutions Si returns
infutuel k
hand
Systems to study in this module?
on

Intuit finte resomes now

cool
stagnant
,

significant Stab but

inputs outputs focus
most
indicate
Boundaries ,
an

*
-
/ perf would
.

↳ Emultipl systemprob
▪ Classification by system characteristics
▪ Static vs. Dynamic Systems
▪ Causal vs. Non-Causal Systems
▪ Time-Variant vs. Time-Invariant Systems
▪ Linear vs. Non-Linear Systems
▪ Invertible vs. Non-Invertible Systems
▪ Stable vs. Unstable Systems

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 53
 Let's protect the environment
Lecture notes are subject to change before
the class for further improvement

THANK YOU
The preparation of this module has been made possible by
the support from IE2141 teaching team, and the past
teaching materials developed by A/Prof. Aaron Chia.

IE2141 Systems Thinking and Dynamics – ISEM Department, National University of Singapore 54