Post on 18-Aug-2018
Lecture # 5: Diagrams for Conceptualization, Explanation, and Presentation
Four types:
Causal-loop diagrams (influence diagrams)
Citizeninvolvement in
planning
Perception thatcitizens are powerlessand disenfranchised
Willingness of citizensto become involved
-+
-
Visitors peryear
Crowding Naturalarea
Contact areaDamage
per visitor
Incidenceof
damageEnvironmental
damage
Damagedensity Regeneration
time
Disappearanceof damage
R
PAD 724 lecture notes Page 28
Averagecommuting time
Averagecommutingdistance
Averagecommuting speed
Tolerablecommuting time
Pressure for more andwider high speed roads
Lane-miles of highspeed roads
Attractivenessof outlying areas
Population inoutlying areas
Number ofcommuters
Net migration tooutlying areas
Highwayconstruction
Far better with loops named in some descriptive, explanatory, or evocative way:
Averagecommuting time
Average commutingdistance
Averagecommuting speed
Tolerablecommuting time
Pressure for moreand wider highspeed roads
Lane-miles of highspeed roads
Attractiveness ofoutlying areas
Population inoutlying areas
Number ofcommuters
Net migrationto outlying
areasHighwayconstruction (B1) Highway
expansion
(B3) Commutingvolume
(B2) Commutingdistance
PAD 724 lecture notes Page 29
Causal loop diagrams with stocks explicit, but flows implicit in the arrows:
Workforce
Net hiring
Progress rate
Cumulative realprogress
Tasks perceivedremaining
Effortperceivedremaining
Indicated worforce
Gross productivity
Initial project definition
Time remainingTime perceived required
Indicated completion date
Scheduledcompletion date
Net additions toschedule
<wcwf>
Undiscoveredrework
Cumulativepreceivedprogress
Fraction satisfactory
Cumulative effort
Perceived productivity
Workforce sought
Willingness tochange
workforceFraction
perceivedcomplete
<Workforce sought>
Detection ofundiscovered
rework
Time to detect rework
<tdrw>
The picture is complete and quantifiable, and may be easier for some audiences to take in because it leaves out the visual complexity of explicit flows as pipes. On the other hand, the fact that the flows are captured as arrows means that the diagram contains less visual information than the same diagram with flows draawn as pipes.
Below is a more complex example of the same idea. Here, pipes would undoubtedly have been too much. The diagram could make sense only as “unfolded,” piece by piece.
PAD 724 lecture notes Page 30
Fund
ing
for
toba
cco
cont
rol p
rogr
ams
Gov.
inco
me
Toba
cco
cont
rol
prog
rams
Smok
ers
Public
awar
enes
s of
toba
cco
health
risk
Pres
sure
on
toba
cco
compa
nies
to
redu
cemar
keting
act
ivities
Toba
cco
mar
keting
activities
Tax
reve
nues
from
smok
ers
+
+
+
-
+
Peop
le q
uitt
ing
smok
ing
-
Smok
ing
as a
social n
orm
Peop
le s
tart
ing
smok
ing
Toba
cco
reve
nues
+
+
+
++
Hea
lth
care
cos
ts
+
Fund
ing
to s
uppo
rtto
bacc
o qu
itting
effo
rts
+
+
Rese
arch
ers
awar
ness
of
toba
cco
health
risk
Fund
ing
for
toba
cco
health
rese
arch
+
+
Govt
awa
rene
ssof
tob
acco
health
risk
+
+
+
+
-
Pro-
toba
cco
cont
itue
ncies
Ant
i-to
bacc
oco
nstitu
encies
++
+
+
Toba
cco
prod
ucts
availabi
lity
+
Toba
cco
grow
ers
+
+
+
++
+
Govt
willingn
ess
toac
t ag
ains
t to
bacc
o
-+
+
+
+
Toba
cco
taxe
s+
+ +
-
Govt
fun
ding
of
toba
cco
cont
rol
+
+
Tren
d in t
obac
coco
mpa
ny r
even
ues
+
-
Ant
i-to
bacc
olegislat
ion
-
+
+
+
+
Ant
i-to
bacc
olit
igat
ion
<Ant
i-to
bacc
olit
igat
ion>
PAD 724 lecture notes Page 31
Causal-loop diagrams with explicit stocks and flows (“stock-and-flow/word-and-arrow” diagrams).
This sort of diagram was almost unheard of when Richardson and Pugh proposed it as a “hybrid” diagram (part causal loop diagram, and part “stock-and-flow/feedback” diagram) and suggested it was probably the most appropriate diagram to use for most purposes. Now, of course, it is the standard we experience in Vensim.
New personnel Experiencedpersonnel
AssimilationrateHiring rate
Experiencedleave rate
Time tobecome
experienced
Frac expleaving p year
Total personnel
Target growthrate
Fraction experienced
Mentoring time neededMentoring time available
Mentoring timeneeded per New
Mentoring timeavailable per Exp
Adequacy ofmentoring
Perceived timeto become Exp
Pressure for mentoring
PAD 724 lecture notes Page 32
All of these loop diagrams are improved by naming the loops: see Sterman, p. 748 for another example.
The structure below is a “concept model” used in a group model building effort focusing on problems in the polypropylene/polyethylene industry (essentially a commodity cycle model).
InventoryProducing Shipping
Demand
Productioncapacity
Inventorycoverage
Fraction of orders
shipped
Planned growth incapacity
Capacity inconstruction
Buildingcapacity
Capacityordering
Capacity
acquisition delay
Deliverydelay
Delivery delayperceived by the
market
Phantomordering
Availability loop
Price and capacityadjustment loop
Phantomordering loops
+-
Ordering
Price
The diagram on the next page is a more complex example, with prominent loops named. It comes from a Center for Technology group modeling effort focusing on CTG’s work to facilitate teamwork and information integration in interagency information technology projects.
PAD 724 lecture notes Page 33
Inte
grat
edin
form
atio
nU
nint
egra
ted
info
rmat
ion
inte
grat
ing
info
Reso
urce
sal
loca
ted
toin
tegr
atio
npr
ojec
tad
ding
to
reso
urce
sto
inte
grat
ion
Prob
lem
sge
nera
ted
inte
grat
ing
info
gene
rati
ngpr
oble
ms
redu
cing
prob
lem
s
Effo
rt t
oin
tegr
atin
g in
fo
Ease
of
inte
grat
ing
info
Effo
rt t
ore
duci
ng p
robl
ems
Perc
eive
d va
lue
of in
tegr
ated
info
rmat
ion
Prob
lem
s ge
nera
ted
per
info
uni
t in
tegr
ated
Will
ingn
ess
to a
lloca
tere
sour
ces
to in
tegr
atio
n
(R) P
robl
ems
com
poun
d
(R)
Succ
ess
enha
nces
reso
urce
s
(B) L
owha
ngin
g fr
uit
(B) P
robl
ems
impe
de p
rogr
ess
(B) A
tten
ding
to
prob
lem
s ea
ses
inte
grat
ing
but
slow
s pr
ogre
ss
Uni
nteg
rate
din
fo w
ithi
n sc
ope
Scop
e of
inte
grat
ion
effo
rt
(B) P
robl
ems
thre
aten
sco
pe
(R) P
erce
ived
val
ueen
hanc
es s
cope
(B) P
robl
ems
rob
reso
urce
s
Reso
urce
spl
anne
d to
info
inte
grat
ion
frac
tion
of
effo
rtto
pro
blem
s
Tole
rabl
epr
oble
ms
Prob
lem
s re
duce
dpe
r ho
ur o
f ef
fortPr
oble
mse
veri
ty
frac
tion
of
effo
rtto
inte
grat
ing
PAD 724 lecture notes Page 34
Sector overview diagramsThese are maps to models, a kind of navigational aid. Be sure to name the links between sectors. Avoid showing too much. See also Sterman, p. 606, for another example.
PAD 724 lecture notes Page 35
Policy structure diagramsSee also Sterman, p. 710, for another example.
PAD 724 lecture notes Page 36