DataPDF Available
-"1"-"
Model&documentation&for:&
&
Modelling&the&dynamics&of&Technological&Innovation&Systems&
&
&
Bob"Walrave*,"Rob"Raven+"
"
*"Eindhoven"University"of"Technology,"School"of"Industrial"Engineering,"Eindhoven,"The"
Netherlands"
+"Utrecht"University,"Copernicus"Institute,"Utrecht,"The"Netherlands&
&
"
"
&
&
&
&
&
&
" "
-"2"-"
The"model"was"developed"in"Ventana"System’s"VENSIM"software."Figure"1"gives"an"
overview"of"the"full"model."The"model,"grounded"in"the"literature,"was"subjected"to"
robustness"analyses"and"served"to"run"the"different"scenarios"in"the"context"of"technological"
innovation"systems.""
Following"modelling"best"practices"(Holtz,"2011),"we"focus"only"on"one"part"the"
overarching"transition"dynamic."That"is,"rather"than"modelling"a"full"transition,"we"aim"to"
replicate"growth"(or"decline)"of"a"TIS"in"the"context"of"a"dominant"socio-technical"regime."As"
such,"the"‘goal’"of"any"given"simulation"is"to"arrive"at"the"state"where"the"‘Market"motor"
loop’"(R.5,"Figure"1)"grows"strong"enough"(i.e.,"generates"enough"resources)"to"drive"the"
whole"system,"independent"of"any"form"of"external"resources,"in"the"context"of"a"dominant"
socio-technical"regime."
The"six"loops"(Figure"1)"are"grounded"in"the"concept"of"motors"of"innovation"(Suurs,"
2009),"which"is"combined"with"the"notion"of"‘transition"pathways’"(Geels"and"Schot,"2007)."
These"six"feedback"loops,"which"are"discussed"in"the"manuscript,"are"presented"in"different"
colours"with"the"variable"names"written"in"black"(in"Figure"1)."The"green"variables"concern"
resources"that"are"being"fed"into"the"system—following"from"‘Fixed"external"resources’"
[Euro]"and/or"the"‘Niche"market’"[dmnl]"itself—needed"to"drive"the"system."The"brown"
variable"denotes"strong"exogenous"influences"(‘Landscape"pressure’"[dmnl]"and"‘Fixed"
external"resources’"[Euro])"on"the"system."The"orange"variables"indicate"adjustment"times"
(delays)"or"ratios."The"pink"constants"represent"resource"distribution"factors."The"purple"
variables"cause"for"random"variations"to"selected"flows."The"unit"of"time"in"the"model"is"
months"and"the"total"simulation"time"is"600"months."The"simulation"algorithm"is"Euler’s"
method"with"a"step"size"(dt)"of"0.0625"months."As"explained"in"the"manuscript,"the"model"
assumes"a"‘single"TIS"–"single"regime’"situation"and"is"not"attuned"to"a"technology-specific"
innovation"system."
The"main"ingredients"of"system"dynamics"models"are"stocks,"flows,"and"variables."In"
the"diagramming"notation,"stocks"are"represented"by"rectangles"and"denote"a"particular"
level"(e.g.,"the"amount"of"technological"knowledge"developed)."Flows"are"depicted"as"pipes"
with"valves"and"are"responsible"for"changes"(e.g.,"decrease"of"increase"of"the"technological"
knowledge)"in"the"stocks."Variables"are"those"constructs"that"are"dependent"on"stocks"and"
flows"(e.g.,"the"‘Resources"to"technology"development’)."Finally,"the"clouds"represent"
infinite"sources"or"outcomes"of"particular"flows"that"are"beyond"the"scope"of"the"model."
-"3"-"
"
"
Figure'1.'Overview'of'the'endogenous'model'structure.'
Regime
resistance loop
-"4"-"
MODEL&DESCRIPTION&
As"explained"in"the"manuscript,"the"model"is"composed"of"two"main"parts:"technology"
development"and"market"development."
Technology&development&
The&knowledge&development&motor&loop&–&Loop&B.1&
The"knowledge"development"motor"serves"as"prerequisite"for"innovation"processes"to"
occur."Activities"that"underlie"this"loop"are,"for"instance,"academic"studies"and"laboratory"
trials."Such"research"and"development"(R&D)"related"activities"constitute"a"source"of"
variation"to"the"system"(Suurs,"2009)"–"and"it"is"assumed"that"this"results"in"the"
development"of"a"certain"path-breaking"technology."We"capture"the"current"state-of-the-
art,"with"respect"to"technological"knowledge"developed,"by"the"stock"‘Technological"
knowledge"developed’"[dmnl]."(This"stock"thus"captures"all"the"knowledge,"distributed"over"
all"different"agents,"related"to"the"path-breaking"invention.)"This"stock"can"vary"between"0"
(lowest"level"="no"knowledge)"and"1"(highest"level"="a"nearly,"for"that"moment"in"time,"
‘complete’"understanding"of"the"technology)."
It"is"widely"acknowledged"that"the"improvement"of"technological"performance"
becomes"increasingly"more"difficult"once"that"technology"matures"and"starts"to"reach"it"
performance"limit"(De"Liso"&"Filatrella,"2008)."Indeed,"development"of"technological"
performance"typically"follows"a"S-shaped"growth"curve"over"time"(Christensen,"1992a,"
1992b;"Schilling"&"Esmundo,"2009;"Suurs,"2009)."We"adopt"this"conception"in"our"model"by"
making"technological"progress"increasingly"difficult"the"higher"the"state"of"
development/diffusion."More"specifically,"increasingly"more"resources"are"needed"to"
achieve"technological"performance"growth,"the"higher"the"state-of-the-art."In"this"respect,"
growth"in"this"part"of"the"system"is"bounded"by"a"performance"ceiling."This"makes"these"two"
loops"‘goal-seeking’"in"nature."
If,"for"some"reason,"knowledge"development"stops"(or"is"severely"limited),"the"
current"body"of"knowledge"becomes"obsolete"over"time"(due"to"new"technological"
developments,"etc.)."As"such,"the"‘Technological"knowledge"developed’"[dmnl]"stock"is"
subject"to"an"outflow,"called"‘Knowledge"decay"rate’"[dmnl/month]."Here"we"assume"an"
exponential"decay"structure"(Sterman,"2000)"(i.e.,"knowledge"decreases"at"a"rate"
-"5"-"
proportional"to"its"current"value:"The"more"knowledge"there"is,"the"more"knowledge"is"
subject"to"becoming"out-dated)."As"such,"we"adopt"an"adjustment"time"(Sterman,"2000)"
given"by"the"constant"‘Knowledge"decay"time’"[month]."Equation"1"captures"the"behaviour"
of"the"knowledge"decay"rate"(where"the"‘Knowledge"development"rate’"[dmnl/month]"is"
given"in"Equation"2)."
(1)$Knowledge$decay$rate$[dmnl/month]:$
d$(Technological$knowledge$developed)$/$dt$=$–$Technological$$knowledge$developed$/$
Knowledge$decay$time$[+$Knowledge$development$rate]$
"
Vice"versa,"due"to"investments"in,"for"instance,"R&D"(captured"by"the"variable"
‘Resources"to"R&D’"[Euro]),"the"stock"‘Technological"knowledge"developed’"[dmnl]"
increases."This"increase"is"captured"by"the"inflow"‘Knowledge"development"rate’"
[dmnl/month]."Here,"we"assume"an"exponential"growth"structure."That"is,"we"assume"that"
improvement"of"technological"performance"becomes"progressively"more"difficult"once"that"
technology"matures"and"starts"to"reach"it"performance"limit"(De"Liso"and"Filatrella,"2008)."
This"growth"characteristic"is"captured"by"Equation"2,"by"the"introduction"of"an"adjustment"
time"(i.e.,"the"‘AT"TKDe’"[month])"in$combination$with"a"increasingly"limited"potential"
maximum"‘Technological"knowledge"developed’"(i.e.,"1"–"‘Technological"knowledge"
developed)."This"setup"allows"the"stock"‘Technological"knowledge"developed’"to"follow"a"S-
shaped"growth"curve,"typical"for"technological"developments"(Christensen,"1992a,"1992b;"
Schilling"and"Esmundo,"2009;"Suurs,"2009)."Subsequently,"we"calculate"the"inflow"as"given"in"
Equation"2—where"the"constant"‘Euro"to"knowledge"development"rate’"[dmnl/Euro]"
captures"how"much"knowledge"is"(initially)"gained"per"invested"Euro—and"where"the"
‘Knowledge"decay"rate’"[dmnl/month]"is"presented"in"Equation"1."
(2)$Knowledge$development$rate$[dmnl/month]:$
d$(Technological$knowledge$developed)$/$dt$=$+$((1$-$Technological$knowledge$developed)$*$
(Resources$to$R&D$*$Euro$to$knowledge$development$rate))$/$AT$TKDe$[–$Knowledge$decay$
rate]$
"
The"stock"‘Technological"knowledge"developed’"[dmnl]"(partially)"determines"the"
levels"of"the"stocks"‘Guidance"of"the"search’"[dmnl]"and"‘Resource"mobilization’"[Euro]."As"
-"6"-"
can"be"seen"in"Figure"1,"these"stocks"are"also"part"of"‘The"knowledge"diffusion"motor’"(loop"
B.2)"and,"therefore,"are"discussed"in"the"next"section."The"latter"stock"captures"the"total"
amount"of"resources"(in"Euro’s)"put"to"use"for"both"technology"development"and"diffusion."
In"this"respect,"a"certain"percentage"of"the"available"budget"(denoted"by"‘Knowledge"
resource"distribution’"[dmnl%])"is"made"available"to"knowledge"development,"while"the"
rest"(i.e.,"1"–"‘Knowledge"resource"distribution’)"is"directed"toward"knowledge"diffusion."
Subsequently,"the"‘Resources"to"R&D’"[Euro]"can"be"calculated"as"in"Equation"3."
(3)$Resources$to$R&D$[Euro]$=$Knowledge$resource$distribution$*$Resource$mobilization$
"
The"‘Resources"to"R&D’"[Euro]"subsequently"drives"the"already"discussed"‘Knowledge"
development"rate’"[dmnl/month]."As"such,"this"completes"the"description"of"the"balancing"
‘Knowledge"development"motor"loop’"(loop"B.1)."
The&knowledge&diffusion&motor&loop&–&Loop&B.2&
In"order"for"the"technology"and"a"market"to"develop,"the"developed"technological"
knowledge"need"to"diffuse"among"relevant"actors"(through,"e.g.,"conferences,"workshops,"
and"alliances)."We"capture"the"current"state"of"knowledge"diffused"with"the"stock"
‘Technological"knowledge"diffused’"[dmnl]."This"stock"also"varies"between"0"(implying"no"
technological"knowledge"is"diffused)"and"1"(meaning"that"nearly"all"currently$available"
technological"knowledge"is"diffused)."This"stock"is"also"subject"to"exponential"decay"due"to"
knowledge"loss"over"time,"which"is"captured"by"the"‘Knowledge"loss"rate’"[dmnl/month],"
with"adjustment"time"‘Knowledge"decay"time’"[month]."Furthermore,"if"new"knowledge"is"
developed"(i.e.,"a"positive"‘Knowledge"development"rate’"[dmnl/month]),"the"level"of"
‘Technological"knowledge"diffused’"[dmnl]"should,"pro-rata,"decrease."This"then"results"in"
Equation"4—where"the"‘Knowledge"diffusion"rate’"[dmnl/month]"is"explained"by"Equation"5."
(4)$Knowledge$loss$rate$[dmnl/month]:$
d$(Technological$knowledge$diffused)$/$dt$=$–$(Technological$knowledge$diffused$/$
Knowledge$decay$time)$–$(Technological$knowledge$diffused$*$(Knowledge$development$rate$
/$Technological$knowledge$developed))$[+$Knowledge$diffusion$rate]$
"
The"level"of"‘Technological"knowledge"diffused’"[dmnl]"increases"through"
investments"(captured"by"the"variable"‘Resources"to"knowledge"diffusion’"[Euro]"–"see"
-"7"-"
Equation"6)."Similar"to"the"structure"underlying"the"‘Knowledge"development"rate’"
[dmnl/month],"we"assume"that"it"becomes"progressively"more"difficult"to"make"progress"
and,"as"such,"introduce"an"adjustment"time"‘AT"TKDi’"[month]"in"combination"with"a"
increasingly"limited"potential"maximum"‘Knowledge"diffusion"rate’"[dmnl/month]"(i.e.,"1$–$
‘Technological$knowledge$diffused’)."Equation"5"captures"this"–"where"the"constant"‘Euro"to"
knowledge"diffusion"rate’"[dmnl/Euro]"denotes"how"much"knowledge"is"(initially)"gained"per"
invested"Euro—and"where"the"‘Knowledge"loss"rate’"[dmnl/month]"is"given"in"Equation"4."
(5)$Knowledge$diffusion$rate$[dmnl/month]:$
d$(Technological$knowledge$diffused)$/$dt$=$+$((1$–$Technological$knowledge$diffused)$*$
(Resources$to$knowledge$diffusion$*$Euro$to$knowledge$diffusion$rate))$/$AT$TKDi$[–$
Knowledge$loss$rate]$
"
Following"the"logic"underlying"Equation"3,"the"‘Resources"to"knowledge"diffusion’"
[Euro]"can"be"calculated"as"follows"(Equation"6)."
(6)$Resources$to$knowledge$diffusion$[Euro]$=$(1$–$Knowledge$resource$distribution)$*$
Resource$mobilization$
"
‘Technological"knowledge"developed’"and"‘Technological"knowledge"diffused’"shape"
the"‘guidance"of"the"search’."For"example,"the"successful"realisation"of"a"research"project,"
contributing"to"‘Technological"knowledge"developed’"and/or"to"'Technological"knowledge"
diffused’"can"result"in"high"expectations"which"contributes"to"the"function"‘Guidance"of"the"
Search’"(Suurs"and"Hekkert,"2012)."As"such,"we"consider"the"interaction"between"the"
‘Technological"knowledge"diffused’"[dmnl]"and"the"‘Technological"knowledge"developed’"
[dmnl]"to"reflect"the"status"of"such"shared"development"vision,"influencing"how"effectively"
the"‘Resources"to"technology"development’"can"be"utilized."Yet,"we"consider"only"a"certain$
percentage"(i.e.,"captured"by"the"constant"‘Effectiveness"and"efficiency"factor"funding’"
[dmnl—%])"of"all"available"resources"for"technology"development"(i.e.,"‘Resources"to"
technology"development’"[Euro])"to"be"subject"such"the"interaction—as"in"Equation"7,"
where"‘AT"GoS’"[month]"is"the"adjustment"time:"
"
"
-"8"-"
(7)$Change$in$Guidance$of$the$Search$[Euro/month]:$
d$(Guidance$of$the$Search)$/$dt$=$((Technological$knowledge$developed$*$Technological$
knowledge$diffused$*$Effectiveness$and$efficiency$factor$*$Resources$to$technology$
development$+$(1$–$Effectiveness$and$efficiency$factor)$*$Resources$to$technology$
development)$–$Guidance$of$the$Search)$/$AT$GoS$
"
Furthermore,"as"there"is"a"time"delay"between"the"accreditation"of"resources"and"the"
actual"resource"division"and"utilization"(i.e.,"‘Resource"mobilization’"[Euro]),"we"utilize"a"
simple"first-order"adaptive"structure"(Sterman,"2000)"for"calculating"the"change"to"‘Resource"
mobilization’"[Euro]."As"such,"we"can"describe"Equation"8"–"where"‘AT"RM’"[month]"denotes"
the"adjustment"time:"
(8)$Change$in$Resource$mobilization$[Euro/month]:$
d$(Resource$mobilization)$/$dt$=$(Guidance$of$the$search$–$Resource$mobilization)$/$AT$RM$
"
The"‘Resources"to"knowledge"diffusion’"[Euro]"drives"the"already"discussed"
‘Knowledge"diffusion"rate’"[dmnl/month]."This"then"completes"the"description"of"the"
balancing"‘Knowledge"diffusion"motor"loop’"(loop"B.2)."
Market&development&
The&entrepreneurial&motor&loop&–&Loop&R.3&
The"manuscript"outlines"that"the"‘Perceived"legitimacy"of"the"TIS’"is"dependent"on"both"
technological"legitimacy"and"market"legitimacy."We"consider"the"interaction"between"
‘Technological"knowledge"developed’"[dmnl]"and"‘Technological"knowledge"diffused’"[dmnl]"
to"represent"such"for"former"type"of"legitimacy."This"latter"type"of"legitimacy"follows"from"
(a)"the"level"of"‘TIS"structures’"[dmnl];"which"positively"influences"the"legitimacy—see"
Equation"16);"and"(b)"the"level"of"‘Regime"resistance"toward"TIS’"[dmnl],"negatively"
influencing"the"legitimacy—see"Equation"24)."The"perceived"technological"legitimacy"and"
perceived"market"legitimacy"combined"determine"the"stock"‘Perceived"legitimacy"of"the"TIS’"
[dmnl]."This"stock"can"vary"between"0"(no"legitimacy)"to"1"(full"legitimacy)."
In"general,"perceptions"adjust"to"new"circumstances"with"a"certain"delay,"which"can"
be"modelled"in"terms"of"the"behaviour"of"a"first-order"adaptive"system"(Sterman,"2000)."As"
-"9"-"
such,"we"model"change"to"the"stock"‘Perceived"legitimacy"of"the"TIS’"[dmnl]"as"a"first-order"
adaptive"system,"in"Equation"9,"with"adjustment"time"‘AT"PLT’"[month]."
(9)$Change$in$Perceived$legitimacy$of$the$TIS$[dmnl/month]:$
d$(Perceived$legitimacy$of$the$TIS)$/$dt$=$((Technological$knowledge$diffused$*$Technological$
knowledge$developed$*$TIS$structures$*$(1$–$Regime$resistance$toward$TIS))-Perceived$
legitimacy$of$the$TIS$)$/$AT$PLT$
"
The"stock"‘Entrepreneurial"activity’"[dmnl]"represents"the"number"of"firms"and"
entrepreneurs"becoming"active"in"the"innovation"system."This"stock,"like"the"formerly"
described"stocks,"also"varies"between"0"(implying"no"entrepreneurial"activity)"and"1"
(meaning"full"‘saturation’"with"respect"to"entrepreneurial"activity)."
The"stock"‘Entrepreneurial"activity’"[dmnl]"is"expected"to"change,"with"a"certain"
delay,"as"the"result"of"change"within"the"‘Perceived"legitimacy"of"the"TIS’"[dmnl]."As"
explained"in"the"manuscript,"we"assume"a"non-linear"relationship,"which"is"modelled"by"
means"of"a"so-called"lookup"function"(denoted"by"the"lookup"variable"‘Effect"of"PLT"on"EA’)."
This"relationship"is"depicted"in"Figure"2."This"relation"thus"follows"a"‘band-wagon"effect’"
(Geels,"2005),"where"the"rate$of$change"in"entrepreneurial"activity"increases"the"more"
entrepreneurial"activity"is"installed.""
"
"
0""""""""""""""""""""0.25"""""""""""""""""""0.5"""""""""""""""""""0.75"""""""""""""""""""1"
Perceived"legitimacy"of"the"TIS"
Figure&3.&Non-linear&relation&between&‘Perceived&legitimacy&of&the&TIS’&and&
‘Entrepreneurial&activity’.&
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0.55
0.6
0.65
0.7
0.75
0.8
0.85
0.9
0.95
1
Entrepreneurial activity
Perceived legitimacy of the TIS
.75"
.5"
.25"
0"
Entrepreneurial"activity"
-"10"-"
Entrepreneurial"activity"can,"furthermore,"be"spurred"by"external"actors,"for"
instance,"by"external"agents"approaching"firms"and"research"institutes"to"participate"in"
development"initiatives"aimed"at"the"realization"of"pilots"and"demonstrations."In"this"
respect,"it"is"very"likely"that"for"instance"governments,"aiming"to"expedite"the"development"
and"introduction"of"the"socially"desirable"new"technology"and"belonging"solutions"(e.g.,"
more"sustainable),"are"backing"entrepreneurs"with"project"specific"subsidies"(Suurs,"2009)."
This"reduces"perceived"entrepreneurial"risks—effectively"resulting"in"more"‘Entrepreneurial"
activity’"[dmnl]"respective"to"the"current"level"of"‘Perceived"legitimacy"of"the"TIS’"[dmnl]."We"
modelled"this"by"introducing"the"variable"‘Effect"funding"on"PLT’"[dmnl]"(see"Equation"11)"
within"the"lookup"section"of"Equation"10."Equation"10"captures"the"described"dynamics—
with"adjustment"time"‘AT"EI’."(Note"that"growth"in"‘Entrepreneurial"activity’"[dmnl]"is"
bounded"by"the"‘1"–"Entrepreneurial"activity’"term.)"
(10)$Entrepreneurial$interest$rate$[dmnl/month]:$
d$(Entrepreneurial$activity)$/$dt$=$+$((1$–$Entrepreneurial$activity)$*$(Effect$of$PLT$on$EA$
(Perceived$legitimacy$of$the$TIS$+$Effect$funding$on$PLT))$/$AT$EI$[–$Entrepreneurial$
disinterest$rate]$
"
In"this"respect,"following"equation"11,"the"variable"‘Effect"funding"on"PLT’"[dmnl]"
effectively"converts"the"amount"of"resources"available"to"stimulate"‘Entrepreneurial"activity’"
[dmnl]"(i.e.,"‘Funding"EA’"[Euro],"see"Equation"23)"to"an"increase"in"the"level"of"‘Perceived"
legitimacy"of"the"TIS’"[dmnl],"through"introduction"of"the"constant"‘Euro"to"change"in"PLT’"
[dmnl/Euro]."For"model"robustness,"we"limit"the"maximum"outcome"of"Equation"11"to"1"by"
means"of"a"so-called"‘MIN’"statement.""
(11)$Effect$funding$on$PLT$[dmnl]$=$MIN$(Euro$to$change$in$PLT$(Funding$EA),$1)$
"
Yet,"if"the"‘Perceived"legitimacy"of"the"TIS’"[dmnl]"drops,"so"does"entrepreneurial"
interest"in"the"TIS—albeit"with"a"certain"delay."We"adopt"an"exponential"decay"structure"to"
model"this;"with"adjustment"time"‘AT"EDI’"[month],"as"presented"by"Equation"12."
(12)$Entrepreneurial$disinterest$rate$[dmnl/month]:$
d$(Entrepreneurial$activity)$/$dt$=$–$Entrepreneurial$activity$/$AT$EDI$[+$Entrepreneurial$
interest$rate]$
-"11"-"
As"‘Entrepreneurial"activity’"[dmnl]"increases,"so"do"the"resources"provided"to"
further"system"development,"as"initial"entrepreneurial"profits"feedback"into"the"system."We"
model"this"through"introducing"the"constant"‘Variable"input"percentage’"[dmnl—%],"which"
makes"a"part"of"the"total"amount"of"external"resources"(i.e.,"Fixed"external"resources’"
[Euro],"see"the"section"‘Exogenous"inputs’)"dependent"on"the"level"of"‘Entrepreneurial"
activity’"[dmnl]."Equation"13"details"the"mechanism."
(13)$External$resources$[Euro]$=$(1$–$Variable$input$percentage)$*$Fixed$external$resources$+$
Variable$input$percentage$*$Fixed$external$resources$*$Entrepreneurial$activity$
"
The"resources"are"subsequently"distributed"over"technology"development—through"
‘Funding"technology"development"(External"resources)’"[Euro]—and"market"development
‘Funding"market"development"(External"resources)’"[Euro],"through"a"given"percentage"
(denoted"by"‘External"resources"distribution’"[dmnl—%])."From"here,"we"can"describe"
Equation"14"and"15."
(14)$Funding$technology$development$(External$resources)$[Euro]$=$External$resources$*$
External$resources$distribution$
"
(15)$Funding$market$development$(External$resources)$[Euro]$=$External$resources$*$(1$–$
External$resources$distribution)$
"
The"‘Funding"technology"development"(External"resources)’"[Euro]"(Equation"14)"
subsequently"serves,"partly,"as"input"for"the"variable"‘Resources"to"technology"development’"
[Euro]"(see"Equation"22),"which"in"its"turn"feeds"into"‘Resource"mobilization’"[Euro]"
(Equation"8)."The"resources"for"market"development"(i.e.,"Equation"15)"are"further"split"into"
resources"directed"to"stimulate"entrepreneurial"activity"(i.e.,"‘Funding"EA’"[Euro],"Equation"
23);"and"funds"directed"to"stimulate"growth"in"TIS"structures"(i.e.,"‘Funding"TISS’"[Euro],"
Equation"24)—see"section"on"the"‘Market"motor"loop’"(loop"R.5)."As"such,"this"completes"
the"description"of"the"reinforcing"‘Entrepreneurial"motor"loop’"(loop"R.3)."
"
-"12"-"
The&system&building&motor&loop&–&Loop&R.4&
As"argued"in"the"manuscript,"as"‘Entrepreneurial"activity’"[dmnl]"increases,"‘TIS"structures’"
[dmnl]"arise."The"‘TIS"structures’"[dmnl]"is"modelled"as"a"stock,"and"can"vary"between"0"(no"
TIS"structures)"and"1"(a"‘complete’"set"of"TIS"structures)."We"assume"that"the"influence"of"
entrepreneurs"on"the"formation"processes"of"‘TIS"structures’"[dmnl]"becomes"increasingly"
large"as"‘Entrepreneurial"activity’"[dmnl]"increases—reflecting"the"need"for"a"certain"critical$
mass"before"substantial"influence"can"be"exercised."We"model"this"relation"between"
‘Entrepreneurial"activity’"[dmnl]"and"‘TIS"structures’"[dmnl]"through"a"lookup"function;"
denoted"by"the"variable"‘Effect"of"EA"on"TISS’"[dmnl]"and"depicted"in"Figure"3."
"
"
0""""""""""""""""""""0.25"""""""""""""""""""0.5"""""""""""""""""""0.75"""""""""""""""""""1"
Entrepreneurial"activity"
Figure&4.&Non-linear&relation&between&‘Entrepreneurial&activity’&and&‘TIS&structures’.&
"
Other"actors"can"also"stimulate"the"development"of"‘TIS"structures’"[dmnl]."For"
instance,"governments"can"support"the"building"of"institutions"by"introducing"new"
environmental"standards"fit"to"the"emerging"TIS"(Suurs,"2009)."We"model"this"influence"by"
introducing"the"variable"‘Effect"funding"on"TISS’"[dmnl/Euro]"in"Equation"16,"which"serves"to"
translate"the"input"that"originates"from"the"variable"‘Funding"TISS’"[Euro]"(see"Equation"24)"
into"an"increase"in"the"‘TIS"structures’"[dmnl]."Note"that,"once"more"for"model"robustness,"
we"limit"the"potential"‘Effect"funding"on"TISS’"[dmnl]"to"1"by"means"of"a"‘MIN’"statement."
From"here,"we"can"define"Equations"16"and"17."
"
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0.55
0.6
0.65
0.7
0.75
0.8
0.85
0.9
0.95
1
Formal institutions and systems
Entrepreneurial activity
1"
"
.7
5"
"
.5"
"
.2
5"
TIS"structures"
-"13"-"
(16)$Change$in$TIS$structures$[dmnl/month]:$
d$(TIS$structures)$/$dt$=$(1$–$TIS$structures)$*$((Effect$of$EA$on$TISS$(Entrepreneurial$activity)$+$
Effect$funding$on$TISS)$–$TIS$structures)$/$AT$TISS$
"
$(17)$Effect$funding$on$TISS$[dmnl/Euro]$=$MIN$(Euro$to$change$in$TISS$*$Funding$TISS,$1)$
"
The"existence"of"‘TIS"structures’"[dmnl]"influences"the"further"development"of"the"TIS"
in"two"manners:"First,"the"existence"of"FIS"provides"face"validity"for"the"emerging"
technology,"directly"influencing"the"‘Perceived"legitimacy"of"the"TIS’"[dmnl]"(see"Equation"9)."
Furthermore,"the"existence"of"‘TIS"structures’"[dmnl]"negatively"influences"the"‘Regime"
resistance"toward"TIS’"[dmnl]"as"broadening"networks"and"the"establishment"of"
intermediaries"enforce"growth"of"the"emerging"niche"market,"thereby"counteracting"regime"
resistance."Also"see"Equation"25."This"concludes"the"discussion"on"the"reinforcing"‘System"
building"motor"loop’"(loop"R.4)."
The&market&motor&loop&–&Loop&R.5&
As"explained"in"the"manuscript,"growth"in"‘Entrepreneurial"activity’"[dmnl]"also"results,"
albeit"with"a"delay"(‘AT"NM’"[month]),"in"‘Niche"market’"[dmnl]"growth"(once"more,"we"
adopt"a"first-order"adaptive"system)."Yet,"the"‘Niche"market’"[dmnl]"can"only"truly"develop"
when"innovation"system"actors"successfully"navigate"the"creation"of"‘TIS"structures’"[dmnl]."
As"such,"we"consider"the"interaction"between"‘Entrepreneurial"activity’"[dmnl]"and"‘TIS"
structures’"[dmnl]"to"indicate"change"in"the"‘Niche"market’"[dmnl]—See"Equation"18."
(18)$Change$in$Niche$market$[dmnl/month]:$
d$(Niche$market)$/$dt$=$((Entrepreneurial$activity$*$TIS$structures)$–$Niche$market)$/$AT$NM$
"
The"reinforcing"‘Market"motor"loop’"(loop"R.5)"can,"once"dominant,"sustain"all"other"
motors"(albeit"that"this"requires"a"rather"large"‘Niche"market’),"by"generating"‘Niche"input’"
[Euro]"to"the"funds"going"into"technology"development."In"such"situation,"there"is"no"need"
for"‘External"resources’"[Euro]"anymore,"as"a"self-sustaining"TIS"has"been"created."The"size"of"
the"variable"‘Niche"input’"[Euro]"is"determined"by"multiplying"the"size"of"the"‘Niche"market’"
[dmnl]"with"the"constant"‘Niche"to"Euro"factor’"[Euro/dmnl]—as"in"Equation"19."
"
-"14"-"
(19)$Niche$input$[Euro]$=$Niche$to$Euro$factor$*$Niche$market$
"
Subsequently,"the"available"resources,"following"from"the"‘Niche"market’"[dmnl]"are"
distributed"over"technology"development"and"market"development."The"constant"‘Niche"
input"distribution’"[Dmnl—%]"determines"this"division;"similar"to"the"constant"‘External"
resources"distribution’—see"Equations"14"and"15."From"here,"we"formulate"Equations"20"
and"21."
(20)$Funding$technology$development$(Niche)$[Euro]$=$Niche$input$*$(Niche$input$
distribution)$
"
(21)$Funding$market$development$(Niche)$[Euro]$=$Niche$input$*$(1$–$Niche$input$
distribution)$
"
The"‘Funding"technology"development"(Niche)’"[Euro]"(Equation"20)"and"the"‘Funding"
technology"development"(External"resources)’"[Euro]"(Function"14)"determine"the"variable"
‘Resources"to"technology"development’"[Euro]"(which"serves"as"input"for"Equation"8),"as"
described"by"Equation"22."
(22)$Resources$to$technology$development$[Euro]$=$Funding$technology$development$
(External$resources)$+$Funding$technology$development$(Niche)$
"
The"resources"following"from"‘Funding"market"development"(External"resources)’"
[Euro]"and"from"‘Funding"market"development"(Niche)’"[Euro]"are"utilized"for"the"
development"of"‘Entrepreneurial"activity’"[dmnl]"and"‘TIS"structures’"[dmnl]."We"do"so"by"
introducing"two"constants:"(1)"‘Market"resource"distribution"(External"resources)’"[dmnl
%],"and"(2)"‘Market"resource"distribution"(Niche)’"[dmnl%]."This"results"in"Equations"23"
and"24—which"feed"into"Equations"11"and"17."
(23)$Funding$EA$[Euro]$=$Funding$market$development$(External$resources)$*$Market$
resource$distribution$(External$resources)$+$Funding$market$development$(Niche)$*$Market$
resource$distribution$(Niche)$$
"
-"15"-"
(24)$Funding$TISS$[Euro]$=$Funding$market$development$(External$resources)$*$(1$–$Market$
resource$distribution$(External$resources))$+$Funding$market$development$(Niche)$*$(1$–$
Market$resource$distribution$(Niche))$
$
This"concludes"the"description"of"the"reinforcing"‘Market"motor"loop’"(loop"R.5)."
The&regime&resistance&loop&–&Loop&B.6&
The"final"loop"in"the"model"is"the"‘Sailing"ship"loop’"(B.6)."This"loop"is"added"to"the"existing"
TIS"framework"to"account"for"the"potential"feedback"loops"between"a"TIS"and"its"context."
First,"the"growth"of"a"TIS"‘Niche"market’"[dmnl]"triggers"the"‘Sailing"ship"effect’"[dmnl],"which"
implies"an"increase"in"the"‘Regime"resistance"toward"TIS’"(ranging"from"0-1:"0"implying"no"
selection"pressure"at"all,"1"means"a"severe"selection"environment;"i.e.,"no"‘Niche"market’"
[dmnl]"can"develop—see"Equation"9)."The"stock,"model"as"a"first-order"adaptive"system,"is"
subject"to"adjustment"time"‘AT"RRT’"[month]."
As"explained,"we"assume"that"the"development"of"‘TIS"structures’"[dmnl]"lowers"the"
‘Regime"resistance"toward"TIS’"[dmnl]."In"order"to"control"for"the"effect"that"‘TIS"structures’"
[dmnl]"has"on"‘Regime"resistance"toward"TIS’"[dmnl],"we"introduce"the"constant"‘Effect"FIS"
on"RRT’"[dmnl%]."More"specifically,"this"latter"constant"causes"that"a"fully"developed"‘TIS"
structures’"[dmnl]"diminishes"the"‘Regime"resistance"toward"TIS’"[dmnl]"to"a"certain"
degree—rather"then"completely"(this"is"in"line"with"our"aim"to"replicate"growth/decline"of"a"
TIS"in"the"context"of"a"dominant"regime,"rather"than"modelling"a"full"transition)."Yet,"
simultaneously,"due"to"‘Niche"market’"[dmnl]"growth"the"‘Regime"resistance"toward"TIS’"
[dmnl]"increases"as"a"result"of"the"‘Sailing"ship"effect’"[dmnl]"(De"Liso"and"Filatrella,"2008)."
Note"that,"in"Equation"25,"we"bounded"growth"to"the"stock"‘Regime"resistance"
toward"TIS’"[dmnl]"to"1—by"means"of"a"‘MIN’"statement."The"assumed"persistence"of"the"
‘Sailing"ship"effect’"[dmnl],"in"combination"with"a"decline"in"the"‘TIS"structures’"[dmnl]"(as"
the"result"of"a"failing"TIS)"can,"without"such"‘MIN’"statement,"result"in"the"situation"where"
the"‘Regime"resistance"toward"TIS’"grows"larger"than"the"theoretically"possible"value"of"1"
(and"compromise"model"robustness).""
Finally,"the"‘Regime"resistance"toward"TIS’"[dmnl]"is"subject"to"‘Landscape"pressure’"
[dmnl]—see"section"on"‘Exogenous"inputs’."Such"influences"(e.g.,"oil"price"changes,"climate"
change,"etc.)"put"significant"pressure"on"the"currently"dominant"regime"(Suurs,"2009),"
-"16"-"
causing"a"drop"in"the"‘Regime"resistance"on"TIS’"[dmnl]."As"such,"we"can"model"change"in"the"
‘Regime"resistance"toward"TIS’"[dmnl]"as"follows"(Equation"25)."
(25)$Change$in$the$Regime$resistance$toward$TIS$[dmnl/month]:$
d$(Regime$resistance$toward$TIS)$/$dt$=$(MIN$(1$–$TIS$structures$*$Effect$FIS$on$RRT$+$Sailing$
ship$effect$–$Landscape$pressure,$1)$–$Regime$resistance$toward$TIS)$/$AT$RRT$
"
The"modelled"‘Sailing"ship"effect’"can"be"described"in"the"following"manner:"
phase"A) Initially,"the"‘Sailing"ship"effect’"will"build"up"as"a"result"of"‘Niche"market’"
[dmnl]"growth."
phase"B) Subsequently,"once"a"certain"‘Niche"market’"[dmnl]"size"has"been"crossed,"
the"‘Sailing"ship"effect’"will"retain"its"maximum"effect"size"(and"not"grow"
any"stronger)."
phase"C) Then,"if"the"‘Niche"market’"[dmnl]"has"been"able"to"keep"a"certain"size,"for"
a"given"amount"of"time,"the"‘Sailing"ship"effect’"declines"again."
As"such,"in"order"to"determine"the"‘Sailing"ship"effect’"[dmnl],"two"elements"of"the"
‘Niche"market’"[dmnl]"need"to"be"captured:"First,"its"historically"largest"size."And"second,"the"
duration"the"‘Niche"market’"[dmnl]"exceeded"a"certain"assumed"threshold"size."That"is,"the"
sailing"ship"effect"is"not"likely"to"last"indefinitely"and,"as"such,"a"niche"market,"exceeding"a"
certain"size"(captured"by"the"constant"‘Size"threshold"SSE’"[dmnl])"for"a"given"amount"of"
time"(captured"by"the"constant"‘Duration"threshold"SSE’"[month]),"is"likely"to"overcome"the"
‘Sailing"ship"effect’"[dmnl]."
For"the"former"element,"we"introduce"the"stock"‘Historically"largest"niche"market’"
[dmnl],"which"records"the"historically"largest"‘Niche"market’"as"a"one-way"adaptive"system"
(implying"the"stock"can"never"decline)"with"no"delay"(adjustment"time"equals"‘Time"Step’)."
As"in"Equation"26."
(26)$Change$in$Historically$largest$Niche$market$[dmnl]:$
d$(Historically$largest$Niche$market)$/$dt$=$IF$THEN$ELSE$(Historically$largest$niche$market$<$
Niche$market,$(Niche$market$–$Historically$largest$niche$market)$/$TIME$STEP,$0)$
"
The"latter"element"is"captured"by"the"stock"‘Duration"above"threshold’"[Month]"
(once"more,"an"one-way"adaptive"system"with"no"delay),"which"keeps"track"of"the"amount"
-"17"-"
of"months"the"‘Niche"market’"exceeded"the"given"‘Size"threshold"SSE’"[Dmnl]—as"described"
by"Function"27."
(27)$Change$in$Duration$above$threshold$[month]:$
d$(Duration$above$threshold)$/$dt$=$IF$THEN$ELSE$(Niche$market$>$Size$threshold$SSE,$1,$0)$/$
TIME$STEP$
"
Equation"26"and"27"combined"determine"the"‘Sailing"ship"effect’"[dmnl]"(following"
the"outlined"reasoning)."In"order"to"do"so,"we"first"determine"the"‘SSE"status’"[dmnl]"(a"
stock,"ranging"from"0"to"1)."This"stock"serves"to"translate"the"values"of"‘Historically"largest"
niche"market’"[dmnl]"and"‘Duration"above"threshold’"[dmnl]"into"a"specific"value"that"can"be"
used"to"calculate"the"‘Sailing"ship"effect’"[dmnl]."We"do"so"as"follows:"If"the"‘Historically"
largest"niche"market’"[dmnl]"did"not"exceed"the"‘Size"threshold"SSE’"[dmnl],"the"‘Sailing"ship"
effect’"[dmnl]"is"calculated"with"the"‘Historically"largest"niche"market’"[dmnl]"(phase"A)."If"
the"size"threshold"has"been"passed,"we"need"to"determine"if"the"duration$threshold"has"
been"passed."Note"that"we"assumed"the"‘Sailing"ship"effect’"[dmnl]"to"be"highly"persistent,"
and"its"effect"is"not"likely"to"decline"before"the"duration"threshold"has"been"reached."As"
such,"if"the"duration"threshold"has"not"been"passed,"then"the"‘Sailing"ship"effect’"[dmnl]"is,"
calculated"by"means"of"the"‘Size"threshold"SSE’"[dmnl]."This"implies"the"highest"possible"
effect"size"(phase"B)."If"the"‘Duration"threshold"SSE’"[month]"has"been"exceeded,"then"the"
‘Sailing"ship"effect’"[dmnl]"is,"once"more,"determined"by"the"‘Historically"largest"niche"
market’"[dmnl];"following"Figure"5,"this"implies"a"potential"decline"of"the"‘Sailing"ship"effect’"
[dmnl]"(phase"C)."Figure"4"outlines"the"reasoning"behind"Equation"28."
"
Figure&4.&Reasoning&behind&Equation&28."
!"#$%#&'()*+,&-.+/#$$0)
$#.1',&)2+/3')4#.5'&)6)7+8')
&3.',3-$9)77!
:!7);<%+$9=%>)>3#,')-?)&3')
7#+$+21),3+>)'??'/&=>3#,')@AB)
Historically+largest+niche+
market
CD);7+8')&3.',3-$9)3#,)E''2)
.'#/3'9AB)!"#$%#&'()
Duration+above+threshold+<+
Duration+threshold+SSE
:!7);7#+$+21),3+>)'??'/&)
.'4#+2,)#&)4#F+4%4)$'"'$)
%2&+$)9%.#&+-2)&3.',3-$9)3#,)
E''2)'F/''9'9=>3#,')<A()
Size+threshold+SSE
CD);G%.#&+-2)&3.',3-$9)
.'#/3'9H)9'/$+2')+2)7#+$+21)
,3+>)'??'/&)>-,,+E$'=>3#,')
IA()Historically+largest+niche+
market
-"18"-"
Given"that"this"effect"follows"from"the"response"of"incumbents"on"TIS"growth,"we"
model"the"delay"‘AT"SSEs’"[month]."This"results"in"Equation"28."
(28)$Change$in$SSE$status$[dmnl/month]:$
d$(SSE$status)$/$dt$=$IF$THEN$ELSE$($Historically$largest$niche$market$<$Size$threshold$SSE,$
Historically$largest$niche$market,$IF$THEN$ELSE$($Duration$above$threshold$<$Duration$
threshold$SSE,$Size$threshold$SSE,$Historically$largest$niche$market)$/$AT$SSEs$
"
Subsequently,"the"‘SSE"status’"[dmnl]"determines"the"variable"‘Sailing"ship"effect’"
[dmnl]."Given"the"non-linear"nature"of"this"relationship"(De"Liso"and"Filatrella,"2008),"we"
adopted"a"lookup"variable"‘Effect"SSEs"on"SSE’"(see"Figure"5)."
"
"
Figure&5.&Non-linear&relation&between&‘SSE&status’&and&the&‘Sailing&ship&effect’."
"
As"can"be"seen"in"Figure"5,"the"potential"maximum"‘Sailing"ship"effect’"[dmnl]"equals"
1."Following"Equation"25,"such"effect"size"would"result"in"a"‘Regime"resistance"toward"TIS’"of"
1"and"a"‘Perceived"legitimacy"of"the"TIS’"of"0;"and"thus"guaranteed"TIS"failure."As"such,"we"
scale"the"‘Sailing"ship"effect’"(i.e.,"the"outcome"of"the"lookup"function"given"by"‘Effect"SSEs"
on"SSE’)"with"the"constant"‘Maximum"effect"size"SSE’"[Dmnl]—as"in"Equation"29."
(29)$Sailing$ship$effect$[dmnl]$=$Effect$SSEs$on$SSE$(SSE$status)$*$Maximum$effect$size$SSE$
"
The"Sailing"ship"effect"then"feeds"into"the"‘Regime"resistance"toward"TIS’"[dmnl]"
(Equation"25),"which"subsequently"determines"part"of"the"‘Perceived"legitimacy"of"the"TIS’"
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0.000
0.202
0.207
0.212
0.217
0.222
0.227
0.232
0.237
0.241
0.246
0.251
0.256
0.261
0.266
0.271
0.276
0.280
0.285
0.290
0.295
Sailing ship effect
SSE status
-"19"-"
[dmnl]"(Equation"9)."As"such,"this"concludes"the"description"of"the"balancing"‘Sailing"ship"
effect"loop’"(loop"B.6)."
EXOGENOUS&INPUTS&
The"model,"as"described"so"far,"is"fully"endogenous"in"nature."Yet,"as"noted"throughout"this"
text,"there"are"two"important"exogenous"inputs"that"influence"TIS"development."First"of"all,"
there"is"‘Landscape"pressure’"[dmnl],"which"lowers"the"‘Regime"resistance"toward"TIS’"
[dmnl]."Secondly,"the"model’s"dynamics"are"started"by"means"of"‘Fixed"external"resources’"
[Euro]."As"explained,"these"funds"can"feed"into"technology"development"as"well"as"the"
market"development"(through"‘External"resources’"[Euro],"Equation"13).""
In"an"attempt"to"keep"the"model"parsimonious,"we"describe"both"inputs"in"terms"of"
(a)"size"[dmnl"or"Euro],"(b)"timing"[month],"and"(c)"duration"[month]."Equation"30"details"the"
‘Landscape"pressure’"[dmnl];"and"Equation"31"details"the"‘Fixed"external"resources’"[Euro].""
(30)$Landscape$pressure$[dmnl]$=$IF$THEN$ELSE$(Time$>=$Timing$LP$:AND:$Time$<$Timing$LP$+$
Duration$LP,$Size$LP,$0)$
"
$(31)$Fixed$external$resources$[Euro]$=$IF$THEN$ELSE$(Time$>=$Timing$FER$:AND:$Time$<$
Timing$FER$+$Duration$FER,$Size$FER,$0)$
"
RANDOM&VARIATIONS&
As"outlined"in"the"manuscript,"we"included"sources"of"random"variation"to"the"model."More"
specifically"to"the"growth"and"decline"in"‘Technological"knowledge"developed’,"
‘Technological"knowledge"diffused’,"‘Entrepreneurial"activity’,"‘TIS"structures’,"and"‘Niche"
market’."We"modelled"this"by"means"of"so-called"‘Pink"noise’"[dmnl%]"(Sterman,"2000)."
More"specifically,"the"following"stock"and"flow"structure"was"used—with"adjustment"time"
‘AT"noise’"[month]."
(32)$Change$in$Pink$noise$[dmnl—%]$
d$(Pink$noise)$/$dt$=$(1$+$Normal$distribution$–$Pink$noise)$/$AT$noise$
"
(33)$Normal$distribution$[dmnl]$=$RANDOM$NORMAL$(-5,$5,$0,$5,$Seed$x)$
"
-"20"-"
Five"different"‘seeds’"(see"Equation"33)"were"used"to"create"five"different"‘versions’"of"noise,"
each"version"influencing"one"or"more"flows."Table"1"detail"which"flow"was"subject"to"what"
‘version’"of"noise."""
"
Table&1.&Pink&noise&–&Flow&relationships.&
Pink&noise&‘version’&
Introduced&to&flow&
Pink"noise"1"
Knowledge"development"rate"(Equation"2);"Knowledge"loss"rate"
(Equation"4)"
Pink"noise"2"
Knowledge"decay"rate"(Equation"1);"Knowledge"diffusion"rate"
(Equation"5)"
Pink"noise"3"
Entrepreneurial"interest"rate"(Equation"10)"
Pink"noise"4"
Change"in"niche"market"(Equation"18)"
Pink"noise"5"
Entrepreneurial"disinterest"rate"(Equation"12);"Change"in"FIS"
(Equation"18)"
"
We"multiplied"the"‘Pink"noise’"[dmnl—%]"with"the"numerator"of"the"targeted"flow."The"
following"adjustments"were"made"to"Equations"1,"2,"4,"5,"10,"12,"16,"and"18—resulting"in"
Equations"1a,"2a,"4a,"5a,"10a,"12a,"16a,"and"18a."We"present"the"new"functions"below."For"
the"reader’s"convenience,"first"the"original"function"is"given,"after"which"the"adjusted"
equation"is"presented."
(1)$Knowledge$decay$rate$[dmnl/month]:$
d$(Technological$knowledge$developed)$/$dt$=$–$Technological$$knowledge$developed$/$
Knowledge$decay$time$[+$Knowledge$development$rate]$
"(1a)$Knowledge$decay$rate$[dmnl/month]:$
d$(Technological$knowledge$developed)$/$dt$=$–$(Technological$knowledge$developed$*$Pink$
noise$2)$/$Knowledge$decay$time$[+$Knowledge$development$rate]$
"
(2)$Knowledge$development$rate$[dmnl/month]:$
d$(Technological$knowledge$developed)$/$dt$=$+$((1$-$Technological$knowledge$developed)$*$
(Resources$to$R&D$*$Euro$to$knowledge$development$rate))$/$AT$TKDe$[–$Knowledge$decay$
rate]$
-"21"-"
(2a)$Knowledge$development$rate$[dmnl/month]:$
d$(Technological$knowledge$developed)$/$dt$=$+$((1-Technological$knowledge$developed)$*$
Pink$noise$1$*$(Resources$to$R&D$*$Euro$to$knowledge$development$rate))$/$AT$TKDe$[–$
Knowledge$decay$rate]$
"
(4)$Knowledge$loss$rate$[dmnl/month]:$
d$(Technological$knowledge$diffused)$/$dt$=$–$(Technological$knowledge$diffused$/$
Knowledge$decay$time)$–$(Technological$knowledge$diffused$*$(Knowledge$development$rate$
/$Technological$knowledge$developed))$[+$Knowledge$diffusion$rate]$
(4a)$Knowledge$loss$rate$[dmnl/month]:$
d$(Technological$knowledge$diffused)$/$dt$=$–$(Technological$knowledge$diffused$/$
Knowledge$decay$time)$–$(Technological$knowledge$diffused$*$(Knowledge$development$rate$
*$Pink$noise$1$/$Technological$knowledge$developed$))$[+$Knowledge$diffusion$rate]$
"
(5)$Knowledge$diffusion$rate$[dmnl/month]:$
d$(Technological$knowledge$diffused)$/$dt$=$+$((1$–$Technological$knowledge$diffused)$*$
(Resources$to$knowledge$diffusion$*$Euro$to$knowledge$diffusion$rate))$/$AT$TKDi$[–$
Knowledge$loss$rate]$
(5a)$Knowledge$diffusion$rate$[dmnl/month]:$
d$(Technological$knowledge$diffused)$/$dt$=$+$((1$–$Technological$knowledge$diffused)$*$(Pink$
noise$2$*$Resources$to$knowledge$diffusion$*$Euro$to$knowledge$diffusion$rate))$/$AT$TKDi$[–$
Knowledge$loss$rate]$
"
(10)$Entrepreneurial$interest$rate$[dmnl/month]:$
d$(Entrepreneurial$activity)$/$dt$=$+$((1$–$Entrepreneurial$activity)$*$(Effect$of$PLT$on$
EA(Perceived$legitimacy$of$the$TIS$+$Effect$funding$on$PLT))$/$AT$EI$[–$Entrepreneurial$
disinterest$rate]$
(10a)$Entrepreneurial$interest$rate$[dmnl/month]:$
d$(Entrepreneurial$activity)$/$dt$=$+$((1$-$Entrepreneurial$activity)$*$Pink$noise$3$*$Effect$of$PLT$
on$EA(Perceived$legitimacy$of$the$TIS$+$Effect$funding$on$PLT))$/$AT$EI$[–$Entrepreneurial$
disinterest$rate]"
-"22"-"
(12)$Entrepreneurial$disinterest$rate$[dmnl/month]:$
d$(Entrepreneurial$activity)$/$dt$=$–$Entrepreneurial$activity$/$AT$EDI$[+$Entrepreneurial$
disinterest$rate]$
(12a)$Entrepreneurial$disinterest$rate$[dmnl/month]:$
d$(Entrepreneurial$activity)$/$dt$=$–$Entrepreneurial$activity$*$Pink$noise$5$/$AT$EDI$[+$
Entrepreneurial$disinterest$rate]$
"
(16)$Change$in$TIS$structures$[dmnl/month]:$
d$(TIS$structures)$/$dt$=$(1$–$TIS$structures)$*$((Effect$of$EA$on$TISS(Entrepreneurial$activity)$+$
Effect$funding$on$TISS)$–$TIS$structures)$/$AT$TISS$
(16a)$Change$in$TIS$structures$[dmnl/month]:$
d$(TIS$structures)$/$dt$=$(1$–$TIS$structures)$*$Pink$noise$5$*$((Effect$of$EA$on$
TISS(Entrepreneurial$activity)$+$Effect$funding$on$TISS)$–$TIS$structures$)$/$AT$TISS$
"
(18)$Change$in$Niche$market$[dmnl/month]:$
d$(Niche$market)$/$dt$=$((Entrepreneurial$activity$*$TIS$structures)$–$Niche$market)$/$AT$NM$
(18a)$Change$in$Niche$market$[dmnl/month]:$
d$(Niche$market)$/$dt$=$(MIN$(Entrepreneurial$activity$*$TIS$structures$*$Pink$noise$4,$1)$–$
Niche$market)$/$AT$NM$
"
Due"to"the"‘Pink"noise’"[dmnl]"introduced"in"Equation"18a,"the"stock"‘Niche"market’"[dmnl]"
can"grow"larger"than"1"(which"should,"theoretically,"not"be"possible)."As"such,"a"‘MIN’"
statement"is"introduced"to"prevent"the"stock"from"growing"larger"than"1."""
"
EXPERIMENTAL&SETUP&
In"order"to"conduct"the"experiments"as"outlined"in"the"manuscript,"the"following"constants"
were"adjusted"(Note"that"all"the"constants"[incl."units"and"values]"are"listed"in"the"section"
‘Constants"and"initial"values’.):"
(a) ‘Landscape"pressure’"[dmnl]"was"the"same"(in"terms"of"size,"timing,"duration)"for"
every"experiment."
-"23"-"
(b) The"‘Fixed"external"resources’"[Euro]"varied"among"the"different"experiments"in"
terms"of:"
1. Timing:"a"proactive"resource"condition"(i.e.,"‘Fixed"external"resources’"[Euro]"
introduced"before"‘Landscape"pressure’"[dmnl])"versus"a"reactive"resource"
condition"(i.e.,"‘Fixed"external"resources’"[Euro]"introduced"at"the"same"time"
as"‘Landscape"pressure’"[dmnl)."
2. Duration:"Vary"between"10"and"15"years"of"‘External"resources’"(while"
keeping"the"‘size’"the"same).""
(c) We"varied"the"‘Regime-TIS"relation’"by"manipulating:"
1. The"‘Duration"threshold"SSE’"[month]"(shorter"for"a"‘Symbiotic’"relationship"
versus"longer"for"a"‘Competitive’"relationship)."
2. The"‘Maximum"effect"size"SSE’"[dmnl]"(lower"for"a"‘Symbiotic’"relationship"
versus"higher"for"a"‘Competitive’"relationship)."
(d) Finally,"we"modelled"three"resource"conditions"through"adjustment"of"the"constant"
‘External"resources"distribution’"[dmnl—%]:"
1. Technology-oriented"resource"conditions."
2. Market-oriented"resource"conditions."
3. Hybrid"conditions."
Table"2,"as"also"presented"in"the"manuscript,"provides"an"overview"of"all"conducted"
experiments.&
&
Table&2.&Experimental&setup."
&
De-alignment&and&re-
alignment&
Reconfiguration&
Transformation&
Technological&
substitution&
Regime-TIS&relation"
Symbiotic"
Symbiotic"
Competitive"
Competitive"
TIS&developed&
before&landscape&
event?"
No"
Yes"
No"
Yes"
Technology-oriented&
resource&conditions&
Experiment"1a"
Experiment"4a"
Experiment"7a"
Experiment"10a"
&
Experiment"1b"
Experiment"4b"
Experiment"7b"
Experiment"10b"
Market-oriented&
resource&conditions&
Experiment"2a"
Experiment"5a"
Experiment"8a"
Experiment"11a"
&
Experiment"2b"
Experiment"5b""
Experiment"8b"
Experiment"11b"
Hybrid&conditions&
Experiment"3a"
Experiment"6a"
Experiment"9a"
Experiment"12a"
&
Experiment"3b"
Experiment"6b"
Experiment"9b"
Experiment"12b"
-"24"-"
CONSTANTS&AND&INITIAL&VALUES&
Table"3"lists"all"the"constants"(incl."values"and"units)."Table"4"gives"all"the"initial"values"(incl."
values"and"units)."
"
Table&3.&List&of&all&the&constants&used."
Name&of&constant&
Value&
Unit&
Remark&
AT"EI"
12"
Month"
"
AT"TISS"
60"
Month"
"
AT"GoS"
3"
Month"
"
AT"NM"
24"
Month"
"
AT"noise"
6"
Month"
"
AT"PLT"
12"
Month"
"
AT"RM"
6"
Month"
"
AT"RRT"
12"
Month"
"
AT"SSEs"
12"
Month"
"
AT"TKDe"
3"
Month""
"
AT"TKDi"
3"
Month""
"
Duration"FER"
180"OR"240"
Month"
Experiments"a"versus"experiments"b"(see"
Table"2)"
Duration"LP"
60"
Month"
"
Duration"threshold"SSE"
12"OR"180"
Month"
Symbiotic"OR"Competitive"
Effect"size"FIS"on"RRT"
0.4"
Dmnl—%"
"
Effectiveness"and"efficiency"
factor"
0.25"
Dmnl—%"
"
Euro"to"change"in"TISS"
1/50.000"
Euro"/"
dmnl"
"
Euro"to"change"in"PLT"
1/15.000"
Euro"/"
Dmnl"
"
Euro"to"knowledge"
development"rate"
1/10.000"
Dmnl"/"
Euro"
"
Euro"to"knowledge"diffusion"
rate"
1/1.000"
Dmnl"/"
Euro"
"
External"resources"distribution"
0.75"OR"0.25"OR"
‘hybrid’1"
Dmnl—%"
Technology"push"OR"Market"pull"OR"Hybrid."
See"footnote.""
Knowledge"decay"time"
60"
Month""
"
Knowledge"resource"
distribution"
0.8"
Dmnl—%"
"
Market"resource"distribution"
(External"resources)"
0.5"
Dmnl—%"
"
Market"resource"distribution"
(Niche)"
0.1"
Dmnl—%"
"
""""""""""""""""""""""""""""""""""""""""""""""""""""""""
"
1"The"hybrid"development,"as"outlined"in"the"manuscript,"goes"from"0.9"to"0.1"over"a"period"
of"36"months"(after"36"month,"it"remains"0.1"until"the"‘Duration"FER’"has"been"reached—as"
determined"by"Equation"31)."This"was"modelled"as"follows:"
$
IF$THEN$ELSE$(Time$>$Timing$FER$:AND:$Time$<$Timing$FER$+$36,$0.9$–$(0.8$/$36)$*$
(Time$–$Timing$FER),$IF$THEN$ELSE$(Time$<=$Timing$FER,$0.9,$0.1))$
-"25"-"
Name&of&constant&
Value&
Unit&
Remark&
Maximum"effect"size"SSE"
0.05"OR"0.25"
Dmnl"
Symbiotic"OR"Competitive"
Niche"input"distribution"
0.05"
Dmnl—%"
"
Niche"to"Euro"factor"
175.000"
Euro"/"
dmnl"
"
Seed"1"
3425"
Dmnl"
"
Seed"2"
11"
Dmnl""
"
Seed"3"
4312"
Dmnl""
"
Seed"4"
12355"
Dmnl""
"
Seed"5"
424"
Dmnl"
"
Size"FER"
3.600.000"
Euro"
This"number"can"be"scaled"if"also"the"four"
‘Euro"to"…’"constants"are"adjusted."
Size"LP"
0.15"
Dmnl"
"
Size"threshold"SSE"
0.05"OR"0.25"
Dmnl"
Symbiotic"OR"Competitive"
Timing"FER"
60"
Month"
"
Timing"LP"
60"OR"120"
"
Reactive"OR"Proactive"strategy"
Variable"input"percentage"
0.25"
Dmnl—%"
"
&
Table&4.&List&of&all&the&stock&and&initial&values&used."
Name&of&stock&
Initial&value&
Unit&
Technological"knowledge"developed"
0.01"
Dmnl"
Technological"knowledge"diffused"
0.01"
Dmnl"
Guidance"of"the"Search"
0"
Euro"
Perceived"legitimacy"of"the"TIS"
0"
Dmnl"
Entrepreneurial"activity"
0"
Dmnl"
Niche"market"
0"
Dmnl"
Regime"resistance"toward"TIS"
1"
Dmnl"
SSE"status"
0"
Dmnl"
Historically"largest"niche"market"
0"
Dmnl"
Duration"above"threshold"
0"
Dmnl"
Resource"mobilization"
0"
Euro"
Pink"noise"1"
1"
Dmnl—%"
Pink"noise"2"
1"
Dmnl—%"
Pink"noise"3"
1"
Dmnl—%"
Pink"noise"4"
1"
Dmnl—%"
Pink"noise"5"
1"
Dmnl—%"
"
CALIBRATION&AND&ROBUSTNESS&TESTS&
As"explained"in"the"manuscript,"we"devised"a"context"where"knowledge"diffusion"requires"
the"least"amount"of"resources;"followed"by"knowledge"development;"followed"by"the"
development"of"entrepreneurial"activity"and,"finally,"the"development"of"TIS"structures."
Following"Table"3,"this"implies"1/1.000"[dmnl/Euro];"1/10.000"[dmnl/Euro];"1/15.000"
[dmnl/Euro];"and"1/50.000"[dmnl/Euro]"respectively."In"this"respect,"we"modelled"a"certain"
(fixed)"resource"requirement"per"loop,"or"‘relative"resistance’—relative"to"its"state"(i.e.,"the"
higher"the"value"of"the"stock"in"question,"the"more"resources"are"required"to"sustain"that"
-"26"-"
state)."For"calibration"purposes,"the"relative"resistance"of"the"different"loops"(motors)"needs"
to"be"aligned"and"matched"to"the"context"under"investigation."In"the"manuscript,"the"
modelled"context"is"loosely"based"on"the"empirical"observations"that,"in"particular"for"
radical"innovations,"costs"increase"substantially"when"moving"from"invention"to"innovation"
and"deployment,"because"new"infrastructures"and"institutions"need"to"be"developed"(and,"
hence,"technology"development"is"significantly"easier"to"achieve"than"market"development)."
Furthermore,"loop"resistance"needs"to"be"aligned"with"resource"input"from"(a)"‘Fixed"
external"resources’"[Euro]"and"(b)"‘Niche"input’"[Euro]"(the"latter"being"determined"by"the"
‘Niche"to"Euro"factor’"[Euro/dmnl])."The"‘Fixed"external"resources’"should"be"set"in"such"a"
manner"that"TIS"development"starts."The"‘Niche"to"Euro"factor’"should"be"modelled"in"such"a"
manner"that"the"‘Market"motor"loop’,"once"dominant,"can"drive"the"system"(i.e.,"sufficient"
resources"should"originate"from"‘Niche"input’—in"relation"to"the"‘Euro"to"…’"constants)."The"
correct"value,"for"both"constants"(‘Niche"to"Euro"factor’"and"‘Niche"input’),"is"best"
determined"by"means"of"experimentation—while"carefully"considering"the"set"‘relative"
resistance’."(Furthermore,"the"resource"distribution"factors"might"also"need"adjustment"to"
suit"the"modelled"context/policies—see"the"pink"variables"in"Figure"1.)"
The"chosen"setup"(e.g.,"the"idea"of"relative"loop"resistance)"implies"that"many"
different"contexts"can"(rather"easily)"be"simulated"by"means"of"the"presented"model,"which"
we"hope"is"going"to"proof"beneficial"for"future"studies."
Extreme&conditions&tests&
The"model"should"behave"realistically,"even"under"extreme"conditions"(Sterman,"2000)."We"
subjected"the"model"to"extreme"conditions"by"setting"extreme"exogenous"inputs."More"
specifically,"the"two"main"exogenous"inputs"were"varied"between"extreme"low"and"extreme"
high"values."We"did"so"in"the"context"of"experiment"3b"(randomly"chosen)—while"keeping"
the"random"noise"constant"(‘Pink"noise’"stocks)."More"specifically,"the"model"was"run"with"a"
‘Fixed"external"resources’"[Euro]"of"0"and"40.000"(base"scenario"is"20.000"Euro);"and"with"a"
‘Landscape"pressure’"[dmnl]"of"0"and"1"(base"scenario"is"0.15"percent"points"dmnl)."Figure"6"
and"7"illustrate"the"result"of"these"experiments"(of"course,"more"variables,"stocks,"and"flows"
were"assessed)—both"tests"indicate"model"robustness"as"all"results"remain"within"the"
theoretically"set"boundaries"(0"and"1).""
-"27"-"
"
Figure&6.&Extreme&conditions&test:&Perceived&legitimacy&of&the&TIS.&(FER"="‘Fixed"External"
Resources’."LP"="‘Landscape"pressure’.)"
&
Figure&7.&Extreme&conditions&test:&Niche&market."(FER"="‘Fixed"External"Resources’."LP"="
‘Landscape"pressure’.)&
Perceived legitimacy of the TIS
0.3
0.225
0.15
0.075
0
0 60 120 180 240 300 360 420 480 540 600
Time (Month)
Dmnl
Perceived legitimacy of the TIS : Extreme - FER 40000
Perceived legitimacy of the TIS : Extreme - FER 0
Perceived legitimacy of the TIS : Extreme - LP 0
Perceived legitimacy of the TIS : Extreme - LP 1
Niche market
0.9
0.675
0.45
0.225
0
0 60 120 180 240 300 360 420 480 540 600
Time (Month)
Dmnl
Niche market : Extreme - FER 40000
Niche market : Extreme - FER 0
Niche market : Extreme - LP 0
Niche market : Extreme - LP 1
-"28"-"
Sensitivity&analyses&
The"sensitivity"analyses"are"best"directed"toward"the"most"important"elements"of"the"
model."For"the"discussed"model,"this"implies"testing"of"the"‘Euro"to"…’"constants"(i.e.,"‘Euro"
to"knowledge"diffusion"rate’,"‘Euro"to"knowledge"development"rate’,"‘Euro"to"change"in"PLT’,"
‘Euro"to"change"in"TISS’)."As"such,"we"varied"these"constants"randomly"(-5%"/"+5%)"over"250"
experiments"each"(i.e.,"univariate"tests)"(once"more,"in"the"context"of"experiment"3b—while"
keeping"the"random"noise"constant)."Figure"8,"9,"10,"and"11"show"the"results"(for"the"stock"
‘Niche"market’)."
"
"
Figure&8.&Sensitivity&analysis:&Euro&to&knowledge&diffusion&rate—output:&Niche&market.&
"
Sensitivity Knowledge Diffusion
50% 75% 95% 100%
Niche market
0.4
0.3
0.2
0.1
00 150 300 450 600
Time (Month)
-"29"-"
"
Figure&9.&Sensitivity&analysis:&Euro&to&knowledge&development&rate—output:&Niche&
market.&
"
Figure&10.&Sensitivity&analysis:&Euro&to&change&in&PLT—output:&Niche&market.&
"
Sensitivity Knowledge Development
50% 75% 95% 100%
Niche market
0.5
0.375
0.25
0.125
00 150 300 450 600
Time (Month)
Sensitivity PLT
50% 75% 95% 100%
Niche market
0.9
0.675
0.45
0.225
00 150 300 450 600
Time (Month)
-"30"-"
"
Figure&11.&Sensitivity&analysis:&Euro&to&change&in&TISS—output:&Niche&market.&
What"becomes"apparent"from"these"results"is"that"the"model"can"be"considered"robust"as"
the"confidence"bounds"are"rather"narrow"for"all"experiments."More"specifically,"for"the"
technology"development"side"(Figures"8"and"9),"this"is"true"for"the"complete"model"run"(until"
t"="600);"for"the"market"development"side"(Figure"10"and"11),"the"model"can"be"considered"
robust"until"about"t"="400."(The"results"thus"indicate"that"the"higher"‘resistance’"loops"are"
also"more"sensitive.)"With"respect"to"the"latter,"the"rather"long"model"run"(600"months),"in"
combination"with"the"existence"of"a"‘tipping"point’,"makes"that"some"of"the"model"runs"
result"in"a"self-sustaining"niche,"while"other"model"runs"result"in"a"declining"niche."This"then"
also"explains"the"rather"large"confidence"bounds"in"Figure"10"and"11"at"t$="600,"which"is,"as"
such,"not"necessarily"the"result"of"an"overly"‘sensitive’"model."This"latter"can"also"be"seen"in"
Figure"12,"which"shows"the"result"illustrated"in"Figure"11"in"a"different"manner."More"
specifically,"rather"than"displaying"confidence"bounds"(as"in"Figure"11),"individual"traces"are"
shown"in"Figure"12."The"effect"of"the"tipping"point,"from"about"t"="400"is"clearly"visible.""
"
Sensitivity TISS
50% 75% 95% 100%
Niche market
0.8
0.6
0.4
0.2
00 150 300 450 600
Time (Month)
-"31"-"
"
Figure&12.&Sensitivity&analysis:&Euro&to&change&in&TISS—output:&Niche&market&(individual&
traces).&
"
REFERENCES&
Aldrich,"H.E.,"Fiol,"C.M."1994."Fools"rush"in?"The"institutional"context"of"industry"creation."
Acad."Manage."Rev."19,"645–670.""
Christensen,"C.M.,"1992."Exploring"the"Limits"of"the"Technology"S-Curve."Part"I:"Component"
Technologies."Prod."Oper."Manag."1,"334–357."
Christensen,"C.M.,"1992."Exploring"the"limits"of"the"technology"S-curve."Part"II:"Architectural"
technologies."Prod."Oper."Manag."1,"358–366."
De"Liso,"N.,"Filatrella,"G.,"2008."On"Technology"Competition:"A"Formal"Analysis"of"the"
“sailing-Ship"Effect.”"Econ."Innov."New"Technol."17,"593–610.""
Geels,"F.W.,"Schot,"J.,"2007."Typology"of"sociotechnical"transition"pathways."Res."Policy"36,"
399417."
Geels,"F.W.,"2005."Processes"and"patterns"in"transitions"and"system"innovations:"Refining"
the"co-evolutionary"multi-level"perspective."Technol."Forecast."Soc."Change"72,"681–
696."
Gort,"M.,"Klepper,"S.,"1982."Time"paths"in"the"diffusion"of"product"innovations."Econ."J."92,"
630653.""
Sensitivity TISS
Niche market
0.8
0.6
0.4
0.2
00 150 300 450 600
Time (Month)
-"32"-"
Holtz,"G.,"2011."Modelling"transitions:"An"appraisal"of"experiences"and"suggestions"for"
research."Environ."Innov."Soc."Transit."1,"167–186."
Schilling,"M.A.,"Esmundo,"M.,"2009."Technology"S-curves"in"renewable"energy"alternatives:"
Analysis"and"implications"for"industry"and"government."Energ."Policy"37,"17671781.""
Sterman,"J.D.,"2000."Business"Dynamics:"Systems"Thinking"and"Modeling"for"a"Complex"
World,"McGraw"Hill,"New"York."
Suurs,"R.A.A.,"2009."Motors"of"Sustainable"Innovation:"Toward"a"theory"on"the"dynamics"of"
technological"innovation"systems."Utrecht"University,"Utrecht."
"
... To elucidate the feedback story, I rely on the work of Walrave & Raven (2016a, 2016b, who have translated the literature on Technological Innovation Systems (TIS) framework (Bergek et al., 2008;Hekkert et al., 2007;Markard & Truffer, 2008) into a system dynamics model. TIS theory posits that seven key functions undergird the formation and growth of innovation systems, namely, "(1) entrepreneurial activities, (2) knowledge development, (3) knowledge diffusion, (4) guidance of search, (5) market formation, (6) mobilization of resources, and (7) creation of legitimacy" (Hekkert et al., 2007;Walrave & Raven, 2016b, p. 1834Wicki & Hansen, 2017). ...
... Infrastructure (Lee et al., 2018). Entrepreneurs contribute to the "development of formal market rules, establishment of intermediary networks, the building of infrastructure, or the development of formal regulations" (Walrave & Raven, 2016a, p. 1837. Through establishing the Market Infrastructure for market formation, entrepreneurial activity "contribute to the creation of a demand for the emerging technology" (Suurs, 2009, p. 56 The System Legitimacy loop, R17, encompasses the aforementioned smaller loops R15 and R16, and "constitutes the most powerful self-reinforcing loop, potentially able to drive the whole system" (Walrave & Raven, 2016b, p. 1838 Market formation of a new technology is bound to precipitate "resistance from actors with interests in the incumbent" regime (Suurs, 2009, p. 57). ...
... As discussed previously, this subsystem (Figure 3.6) was adapted from a published SD model by Walrave & Raven (2016a, 2016b, which adequately represents the complexity of market formation in technological innovation systemsand thus appropriate for the nascent digital prosthetics market. Here, I discuss the main differences in modelling choices. ...
Thesis
Full-text available
The World Health Organization has predicted a doubling in the global population for persons with amputation by 2050 because of steady population growth, ageing populations, and climbing rates chronic conditions such as peripheral arterial disease and diabetes mellitus. Without proper and timely prosthetic interventions, amputees with major lower-limb loss experience adverse mobility outcomes, including the loss of independence, lowered quality of life, and decreased life expectancy. Yet, a vast majority of amputees still do not have access to prosthetic services given the present capacity constraints, lack of proximity to services, high costs and poor healthcare coverage. Today, the entry of digital technology to the prosthetics services industry (e.g., 3D-printed sockets) is touted to be a plausible solution to this problem. This thesis aims to assess the impact of digital prosthetics on the amputee mobility outcomes – specifically, the proportion of amputees who successfully regain mobility from using a prosthesis and the health-economic consequences of such mobility. Using the system dynamics approach, this study presents a computational simulation model ¬– representing the patient-care continuum and digital prosthetics system – that provides a feedback-rich causal theory of how digital prosthetics impacts amputee mobility outcomes over time. In general, this study has found that with sufficient resources for market formation and capacity expansion for digital prosthetics services, substantial improvements to mobility outcomes for amputees can be expected. In doing so, it serves as proof-of-concept for the viability of scaling digital prosthetics for enabling mobility and bolstering the social impact of providing a prosthesis. Based on the high-leverage policy levers found in the system, this study further discusses the model-based insights that could inform policy design for alleviating the barriers to access and enhancing the health-economic outcomes of prosthetics care.
Article
Full-text available
The World Health Organization estimates that 5 to 15% of amputees in any given population have access to a prosthesis. This figure is likely to worsen as the amputee population is expected to double by 2050, straining the limited capacity of prosthetics services. Without proper and timely prosthetic interventions, amputees with major lower-limb loss experience adverse mobility outcomes, including the loss of independence, lowered quality of life, and decreased life expectancy. Presently, the use of digital technology in prosthetics (e.g., 3D imaging, digital processing, and 3D printed sockets) is contended as a viable solution to this problem. This paper uses system dynamics modeling to assess the impact of digital prosthetics service provision. Our simulation model represents the patient-care continuum and digital prosthetics market system, providing a feedback-rich causal theory of how digital prosthetics impacts amputee mobility and the corollary socio-health-economic outcomes over time. With sufficient resources for market formation and capacity expansion for digital prosthetics services, our work suggests an increased proportion of prosthesis usage and improved associated health-economic outcomes. Accordingly, our findings could provide decision support for health policy to better mitigate the accessibility problem and bolster the social impact of prosthesis usage.
Article
Full-text available
Now organizations are always vulnerable to the liabilities of newness, but such pressures are especially severe when an industry is in its formative years. We focus on one set of constraints facing entrepreneurs in emerging industries-their relative lack of cognitive and sociopolitical legitimacy. We examine the strategies that founders can pursue, suggesting how their successful pursuit of legitimacy may evolve from innovative ventures to broader contexts, collectively reshaping industry and institutional environments.
Article
Formal modelling is potentially a useful method to help understand the many mechanisms underlying transitions and their complex interactions. This article reviews existing models of transitions. It examines the mechanisms they cover, and to what extent the models contribute to understanding core characteristics of transitions, notably multi-domain interactions, inertia to change resulting from path-dependency, and the processes that induce, and contribute to self-reinforcement of, change. It is further assessed how well the models capture real world mechanisms and how this is motivated. It is concluded that models addressing the full extent of transitions suffer from a weak micro-foundation and empirical validation, which is largely due to the scope of transitions and the fragmented theoretical basis of transition research. It is proposed that transitions be broken down into particular problems and components, and that where possible associated areas of interest be defined and addressed in corresponding models.
Article
The technology S-curve is a useful framework describing the substitution of new for old technologies at the industry level. In this paper I use information from the technological history of the disk drive industry to examine the usefulness of the S-curve framework for managers at the firm level in planning for new technology development. Because improvements in over-all disk drive product performance result from the interaction of improved component technologies and new architectural technologies, each of these must be monitored and managed. This paper focuses on component technology S-curves, and a subsequent paper, also published in this issue of the journal, examines architectural technology Scurves. Improvement in individual components followed S-curve patterns, but I show that the flattening of S-curves is a firm-specific, rather than uniform industry phenomenon. Lack of progress in conventional technologies may be the result, rather than the stimulus, of a forecast that the conventional technology is maturing, and some firms demonstrated the ability to wring far greater levels of performance from existing component technologies than other firms. Attacking entrant firms evidenced a distinct disadvantage versus incumbent firms in developing and using new component technologies. Firms pursuing aggressive Scurve switching strategies in component technology development gained no strategic advantage over firms whose strategies focused on extending the life of established component technologies.
Article
This is the second of two papers in which I use information from the technological history of the disk drive industry to examine the usefulness of the S-curve framework for managers at the firm level in planning for new technology development. In this article I show that it is in architectural, rather than component innovation, that entrant firms exhibit an attacker's advantage. A conventionally drawn sequence of intersecting S-curves is a misleading conceptualization of the substitution process for new architectural technologies, because it characterizes architectural innovations strictly in technical terms. Innovations in architectural technologies frequently redefine the functionality of products and address product performance needs in new or remote markets, rather than mainstream ones. Such innovations in architectural technologies entail market innovation as much as technology development, and it is in their ability to aggressively enter emerging or remote markets that entrant firms exhibit an attacker's advantage. I propose a different S-curve framework for processes of architectural technology change that comprehends both its technological and market aspects.
Article
This article investigates transitions at the level of societal functions (e.g., transport, communication, housing). Societal functions are fulfilled by sociotechnical systems, which consist of a cluster of aligned elements, e.g., artifacts, knowledge, markets, regulation, cultural meaning, infrastructure, maintenance networks and supply networks. Transitions are conceptualised as system innovations, i.e., a change from one sociotechnical system to another. The article describes a co-evolutionary multi-level perspective to understand how system innovations come about through the interplay between technology and society. The article makes a new step as it further refines the multi-level perspective by distinguishing characteristic patterns: (a) two transition routes, (b) fit–stretch pattern, and (c) patterns in breakthrough.
Article
Plotting the performance of a technology against the money or effort invested in it most often yields an S-shaped curve: slow initial improvement, then accelerated improvement, then diminishing improvement. These S-curves can be used to gain insight into the relative payoff of investment in competing technologies, as well as providing some insight into when and why some technologies overtake others in the race for dominance. Analyzing renewable energies from such a technology S-curve perspective reveals some surprising and important implications for both government and industry. Using data on government R&D investment and technological improvement (in the form of cost reductions), we show that both wind energy and geothermal energy are poised to become more economical than fossil fuels within a relatively short time frame. The evidence further suggests that R&D for wind and geothermal technologies has been under-funded by national governments relative to funding for solar technologies, and government funding of fossil fuel technologies might be excessive given the diminishing performance of those technologies.