How to involve inhabitants in urban design planning by using digital tools? An overview on a state of the art, key challenges and promising approaches

ScienceDi ec A ailable online a www.sciencedi ec .com P ocedia Compu e Science 112 (2017) 2391–2405 1877-0509 © 2017 The Au ho s. Published by Else ie B.V. Pee - e iew unde esponsibili y o KES In e na ional 10.1016/j.p ocs.2017.08.102 10.1016/j.p ocs.2017.08.102 © 2017 The Au ho s. Published by Else ie B.V. Pee - e iew unde esponsibili y o KES In e na ional 1877-0509 A ailable online a www.sciencedi ec .com ScienceDi ec P ocedia Compu e Science 00 (2017) 000–000 www.else ie .com/loca e/p ocedia 1877-0509 © 2016 The Au ho s. Published by Else ie B.V. Pee - e iew unde esponsibili y o KES In e na ional. 21s In e na ional Con e ence on Knowledge Based and In elligen In o ma ion and Enginee ing Sys ems, KES2017 How o in ol e inhabi an s in u ban design planning by using digi al ools? An o e iew on a s a e o he a , key challenges and p omising app oaches Sande Müns e a*, Ch is ophe Geo gia, Ka ina Heijneb, Ke in Klame a, Jö g Raine Noenniga, Ma hias Pumpa, Benjamin S elzlea, Han an de Mee b aTU D esden, Ge many bTU Del , The Ne he lands Abs ac Di e en cases o public disag eemen in di e en Eu opean coun ies ha e shown ecen ly ha pe using a ho ough planning p ocess is by no means a gua an ee o a b oad public accep ance o an en isioned u ban p ojec . Consequen ly, he employmen o digi al media and ools o enable pa icipa ion o inhabi an s in u ban planning p ocesses on a massi e scale is a p omising, bu cu en ly no comp ehensi ely analyzed app oach. Ou esea ch ac i i ies a e in ended o gain an o e iew on a s a e o he a o esea ch on communica ion channels, me hods and bes p ac ices as well as o iden i y key challenges and p omising s a egies and ools o o e come hese challenges wi h speci ic ega ds o la ge numbe s o use s and digi al suppo ed app oaches. The la e aspec s comp ise he in es iga ion o phenomena like pa icipan selec ion, aming e ec s and gami ied app oaches o digi al-media ed pa icipa o y p ocesses as well as na i e language p ocessing echniques o examine opinions as well as ideas o ele ance om massi e public eedback. To examine, we pe o med li e a u e e iews o se e al hund ed esea ch a icles, in- es iga ed cases in Ge many, F ance and he Ne he lands by in e iews and wo kshops wi h s akeholde s and employed me hods o p o o yping o concep ualize, de elop and assess some p omising app oaches such as sen imen analysis in de ail. © 2016 The Au ho s. Published by Else ie B.V. Pee - e iew unde esponsibili y o KES In e na ional. Keywo ds: u ban planning, su ey, public pa icipa ion, s a e o he a * Co esponding au ho . Tel.: +49-351-46332530, E-mail add ess: sande .muens e @ u-d esden.de A ailable online a www.sciencedi ec .com ScienceDi ec P ocedia Compu e Science 00 (2017) 000–000 www.else ie .com/loca e/p ocedia 1877-0509 © 2016 The Au ho s. Published by Else ie B.V. Pee - e iew unde esponsibili y o KES In e na ional. 21s In e na ional Con e ence on Knowledge Based and In elligen In o ma ion and Enginee ing Sys ems, KES2017 How o in ol e inhabi an s in u ban design planning by using digi al ools? An o e iew on a s a e o he a , key challenges and p omising app oaches Sande Müns e a*, Ch is ophe Geo gia, Ka ina Heijneb, Ke in Klame a, Jö g Raine Noenniga, Ma hias Pumpa, Benjamin S elzlea, Han an de Mee b aTU D esden, Ge many bTU Del , The Ne he lands Abs ac Di e en cases o public disag eemen in di e en Eu opean coun ies ha e shown ecen ly ha pe using a ho ough planning p ocess is by no means a gua an ee o a b oad public accep ance o an en isioned u ban p ojec . Consequen ly, he employmen o digi al media and ools o enable pa icipa ion o inhabi an s in u ban planning p ocesses on a massi e scale is a p omising, bu cu en ly no comp ehensi ely analyzed app oach. Ou esea ch ac i i ies a e in ended o gain an o e iew on a s a e o he a o esea ch on communica ion channels, me hods and bes p ac ices as well as o iden i y key challenges and p omising s a egies and ools o o e come hese challenges wi h speci ic ega ds o la ge numbe s o use s and digi al suppo ed app oaches. The la e aspec s comp ise he in es iga ion o phenomena like pa icipan selec ion, aming e ec s and gami ied app oaches o digi al-media ed pa icipa o y p ocesses as well as na i e language p ocessing echniques o examine opinions as well as ideas o ele ance om massi e public eedback. To examine, we pe o med li e a u e e iews o se e al hund ed esea ch a icles, in- es iga ed cases in Ge many, F ance and he Ne he lands by in e iews and wo kshops wi h s akeholde s and employed me hods o p o o yping o concep ualize, de elop and assess some p omising app oaches such as sen imen analysis in de ail. © 2016 The Au ho s. Published by Else ie B.V. Pee - e iew unde esponsibili y o KES In e na ional. Keywo ds: u ban planning, su ey, public pa icipa ion, s a e o he a * Co esponding au ho . Tel.: +49-351-46332530, E-mail add ess: sande .muens e @ u-d esden.de 2392 Sande Müns e e al. / P ocedia Compu e Science 112 (2017) 2391–2405 Au ho name / P ocedia Compu e Science 00 (2016) 000–000 1. In oduc ion Di e en cases o public disag eemen in di e en Eu opean coun ies ha e shown ecen ly ha pe using a ho - ough planning p ocess is by no means a gua an ee o a b oad public accep ance o an en isioned u ban p ojec . Cu en ly exis ing measu es o in o m he public and collec po en ial disag eemen a e based on classical media and equi e o lookup plans and p ojec desc ip ions du ing he opening hou s o municipal o ices whe e hose ma e ials a e a display. As seen o se e al ecen planning cases o e un by public disag eemen such as S u ga 21 [c . 1], he 3 d unway o he Hea h ow ai po [2] o he wi hd awn Olympic games applica ion o he ci y o Hambu g, his p ocedu e becomes ou moded in he digi al age. Also, a public deba e on u ban planning become inc easingly con- using in e ms o communica ion channels, p inciples and opinions [3]. A p omising app oach o o e come hese p oblems is o ac i ely in ol e inhabi an s in public p ocesses. Du ing he las decades his idea go e y popula in o de o e ilize accep ance and o educe scep icism o ci izens owa ds decisions o publically high awa eness [eg. 4, 5, 6]. Mo eo e , public in ol emen is o en seen – bu no undoub ed - as medium o enhance a quali y o deci- sion making in hese con ex s especially in e ms o he p o ision o lay knowledge and locally adjus ed solu ions [7, 8]. How o in ol e inhabi an s in u ban design planning? Acco ding o B yson [9] a majo p e equisi e o se up a success ul pa icipa i e p ocess is o ca e ully assess and design o an in ended con ex and pu pose. Mos la ge scale-p ojec s ha e o icial p ojec websi es and employ social media channels like Facebook and Twi e , and so do ini ia i es agains planning p ojec s [eg. 3]. Consequen ly, he employmen o digi al media and ools o enable pa icipa ion o inhabi an s in u ban planning p ocesses on a massi e scale is a p omising, bu cu en ly no comp e- hensi ely analyzed app oach. Since ou esea ch will base he de elopmen o a digi ally enhanced pla o m o pa icipa o y planning suppo [10], his a icle is dedica ed o gain an o e iew on (a) a s a e o he a o commu- nica ion channels, me hods and bes p ac ices, (b) key challenges and (c) p omising app oaches o o e come hese challenges wi h speci ic ega ds o digi al suppo ed app oaches. 2. Resea ch Design Wi hin ou esea ch, we pe o med h ee s ages o in es iga ion: A li e a u e e iew ocussed on he examina ion o a cu en s a e o esea ch. To gain p ac ical g ounded knowledge on cu en challenges and s a egies o cope wi h la ge scale complex u ban planning p ojec s we in es iga ed cases in Ge many, F ance and he Ne he lands by in- e iews and wo kshops wi h s akeholde s. Finally, we employed me hods o p o o yping o concep ualize, de elop and assess some p omising app oaches such as sen imen analysis. S age 1: Li e a u e Re iew Since he e is much esea ch on public pa icipa ion in u ban and a chi ec u al planning, ou me hodical app oach is o e iew and classi y exis ing li e a u e o examine basic p inciples and key insigh s and gain implica ions o he design o he a ge ed pla o m. F om a me hodical poin o iew a li e a u e e iew is “a sys ema ic sea ch o pub- lished wo k o ind ou wha is al eady known abou he in ended esea ch opic” [11, p. 58], o p o ide „an in- o med e alua ion o ha li e a u e“ [12]. A li e a u e e iew is a ela i ely low s anda dized me hod and elies on s ages o da a sea ch and c i ical e alua ion [13]. Wi hin ou s udy he pu pose o he e iew is o “ amilia ise he esea che wi h he la es de elopmen s in he a ea o esea ch” and „s udy he de ini ions used in p e ious wo ks as well as he cha ac e is ics o he popula ions in es iga ed, wi h he aim o adop ing hem o he new esea ch” [14, p. 20]. S age 2: Case esea ch Case s udy esea ch is a quali a i e esea ch pa adigm o in es iga e complex phenomena [15]. To gain g ounded insigh in challenges and s a egies un eiled du ing o me pa icipa o y p ocesses an in-dep h s udy was pe o med on speci ic cases in Ge many, F ance and he Ne he lands. These cases we e selec ed o hei cha ac e is ics, which make hem clea a ge p ojec examples o he designa ed pla o m: la ge-scale, complex, dynamic, poli ically sensi i e and in ol ing huge numbe s o s akeholde s. Fo each case a whi e a ie y o quali a i e da a was e- ie ed, e.g. in e iews wi h p ojec leade s and o he s akeholde s; analysis o he w i en epo s on he cases; anal- ysis o he communica ion channels; and obse a ions du ing case wo kshops o ganized o he public. Finally, he Sande Müns e e al. / P ocedia Compu e Science 112 (2017) 2391–2405 2393 Au ho name / P ocedia Compu e Science 00 (2016) 000–000 3 cases we e compa ed in o de o iden i y simila i ies and di e ences, which led o a lis o challenges ha mos o such la ge-scale p ojec s ace ega ding pa icipa o y planning, co-c ea ion and communica ion. S age 3: P o o ypes P o o yping is an es ablished me hod especially in enginee ing o ( e-)sea ch solu ions “by doing”, o build common g ound o communica ion as well as enable apid es ing [16]. P o o yping includes al e na ing s ages o concep ion, de elopmen and es ing. Wi hin ou esea ch we use p o o yping me hods o o i y and assess p omis- ing app oaches o e ilize public pa icipa ion. While a so wa e p o o ype o sen imen analysis is ye o be es ed, he concep ual design o u he p o o ypes a e e e ed o in pa ag aph 4.4 and in [17]. 3. S a e o he a on a me hodology o pa icipa ion in u ban planning A i s s age is dedica ed o examine a s a e o he a on pa icipa i e u ban planning. This comp ises in pa icu- la a e iew o communica ion channels, exis ing me hods o pa icipa ion and digi al suppo ed app oaches. 3.1. A pa icipa i e u ban planning p ocess How does a p ocess o u ban planning ypically ake place? And how o enhance ha p ocess by employing mechanisms o public pa icipa ion? To in es iga e hese issues we esea ched 61 a icles on u ban planning and pa icipa ion [c . 10]. T adi ionally, u ban design and de elopmen p ocesses employ s ages and phases as app op i- a e ins umen s o ensu e success ul design ou comes. P ocesses used o pass om p e-design analysis and b ie ings h ough design and planning phases o inal execu ion (c . Fig. 1). Figu e 1. U ban Design P ocess scheme In he ea ly phases, p o essional p e-design me hods such as P og amming [18] we e used o collec and o ganize p ojec ele an da a, some imes h ough s uc u ed o ma s o mul iple s akeholde s b ie ings and wo kshops. I has been a gued his simple linea model is no a all he igh way o sol e eal p oblems o he eal wo ld. In mo e ecen s udies, he p ocess ge s di ided in mul iple p ocesses ha un pa allel and o e lap, e.g. p ojec managemen , in o ma ion inding, con en inding, accep ance inding [19]. Agains ha backg ound, he key idea, howe e , is o in ol e end-use s as ea ly as possible in he o e all p ocess, and o main ain di ec pa icipa ion as a as possible [20]. Since use in e ac i e and pa icipa o y planning app oaches ha e been epo ed om many coun ies, includ- ing he Uni ed S a es [21], Uni ed Kingdom [22], Belgium [23], Scandina ia [24, 25] and he Ne he lands [26], he me hodology, as i was es ed and elabo a ed in a chi ec u al and u ban design p ac ice, ecommends o s a wi h he in ol emen o a la ge numbe o end-use s and o e ine he gene a ed in o ma ion in smalle wo king g oups wi h ac i is s [27]. A la e s ages, when expe s engage in o communica ion wi h end-use s o ge eedback, he pa icipa ion p ocess will be again opened o a la ge numbe o esponden s. A se ious me hodological p oblem is aised in ega ds o le el o in o ma ion, complexi y, and secu i y [28]. The lowe le els a e he public igh o know, o be in o med implying no in ol emen in o he design p ocess. Medium le els include pa icipa ion in “de- ining in e es s and he agenda” and in assessing he isks and ecommending solu ions”. The highes le el o pa ic- ipa ion is eached when pa icipa ion akes he o m o “pa ne ship in he inal decision”. 3.2. A s a e o he a on communica ion channels Pa icipa o y planning ac i i ies ( ools and me hods) can be deli e ed h ough physical o i ual communica ion channels, o a combina ion o bo h. Figu e 2 and Table 1 p o ide an o e iew o possibili ies, which can be di ided in o 1-way and 2-way communica ion. 1-way communica ion does no allow o pa icipa o y planning ac i i ies, P e-designanalysis B ie ing P o essionaldesign Execu ion 2394 Sande Müns e e al. / P ocedia Compu e Science 112 (2017) 2391–2405 Au ho name / P ocedia Compu e Science 00 (2016) 000–000 since ac ual pa icipa ion can only ake place in in e ac ion, hus 2-way. Howe e , as he 1-way communica ion channels may be needed o each he a ge audience in he i s place and o in o m hem, hey a e included as well. Figu e 2. O e iew o communica ion channels: physical, i ual, 1-way and 2-way Table 1. O e iew o pa ame e s o conside when selec ing he igh communica ion channel(s) o i he pu pose o he speci ic pa icipa o y planning ac i i y. Pa ame e : Sho explana ion: Conside : Re : Le el o pa ici- pa ion A ns ein’s pa icipa ion ladde desc ibes i e le els o pa icipa ion and se s he ole o pa icipan s (ac i e o eac i e). - In o ming, consul ing, in ol ing, collabo a ing o empowe ing [29- 31] Le el o pa ici- pan in e ac ion Pa icipan s can in e ac wi h each o he in di e en compo- si ions. - Indi idually, in pai s, in g oups, o a p og ess h ough hese ypes (me-we-us me hod) - Synch onous s: asynch onous [32- 34] Pu pose o he p oc ess s age Di e en s ages in he u ban planning p ocess equi e di e en mindse s and ules. E.g. he p ocess o ga he ing opinions o ob aining ideas di e s a lo om making deci- sions. - Accep ance inding, con en inding, in o - ma ion inding - P oblem inding, idea inding, solu ion inding - Di e ging, clus e ing, con e ging [9, 35] Reach I is impo an o a ge audiences whe e he ba ie s o pa icipa e a e minimized. The ease o pa icipa ion also plays a la ge ole. - Home, public and ansi spaces, p ojec si e, e en cen es/ own halls, and online space - The access need o mobile de ices - Time and e o o pa icipa e [36- 38] Scalabili y In addi ion o he each, he ques ion is how many pa ici- pan s you wan o each. E.g. he acili a o /pa icipan a io in a ace- o- ace wo kshop is 1/8. - 1-8 pa icipan s, 8-100 pa icipan s, 100-1000 pa icipan s, 1000+ pa icipan s [35, 39, 40] Pa icipan selec ion Closely linked o he p e ious wo poin s is whe he he eached pa icipan s ep esen he ull ange o opinions. See also pa ag aph 4.1 - Open o e e yone (sel -selec ion), s akeholde ep esen a i e, demog aphically ep esen a i e, o speci ic indi iduals [9, 37, 40] Pa icipan skills Conside whe he he pa icipan s would need any aining in o de o pa icipa e h ough he channel. - Ini ial/no pa icipan aining equi ed [41] Cul u al ap- plicabili y Cul u al backg ound should be conside ed, since i ela es o pa icipan s’ social beha iou in ce ain se ings e.g. ela ed o powe dis ance. - Online eadiness - People beha iou in open s. anonymous se ing [37, 42- 44] Cos s Some channels a e mo e expensi e o ope a e han o he s. - Cos s pe pa icipan - To al cos s o pa icipa o y planning ac i i y [45, 46] In e ac ion quali y and dep h No all me hods ha e he same in e ac ion quali y and dep h. - Deli e ing high s. low pe sonal con ac and impac [9, 47, 48] Requi ed labou and expe ise Linked o he p e ious wo poin s: cos s and in e ac ion quali y and dep h. Some channels equi e li le high-skilled labou , while o he s need e.g. a closely in ol ed expe acili a o . - Amoun o labou and expe ise necessa y o con igu a ion, engagemen , in ol emen , u iliza- ion, analyzing, e c. [46, 48] Sande Müns e e al. / P ocedia Compu e Science 112 (2017) 2391–2405 2395 Au ho name / P ocedia Compu e Science 00 (2016) 000–000 5 Wi hin each channel, he e is a wide a ie y o ools and me hods possible. Fo ins ance, “wo kshops” can ange om se ious gaming, o li ing labs, o gene a i e sessions, e c. Each channel has i s own oppo uni ies and limi a- ions. The e o e, i is impo an o delibe a ely choose he igh channel(s) o he igh pu pose o he pa icipa o y p ocess ac i i y. To unde s and which pa ame e s should be aken in o conside a ion, an in es iga ion on pu poses o pa icipa o y planning was done which led o he se o pa ame e s lis ed in able 1. Fo compiling his o e iew he gene al snowballing guidelines [49] we e used. The o e iew is la gely based on h ee me a-analyses om In- ol e [50], B yson e al. [9] and Leighninge [48] and supplemen ed by addi ional public pa icipa ion li e a u e as well as design esea ch li e a u e. This o e iew may no ye be comple e, bu should be in e p e ed as a s a ing poin o a highe pu pose, which is de eloping a amewo k o suppo selec ing he igh pa icipa o y planning o co-c ea ion ool(s) o a speci ic s age o si ua ion in an u ban planning p ojec . Howe e , his o e iew al eady shows ha a lo o di e en combina ions a e possible in o de o deli e he igh pa icipa o y p ocess and achie - ing he desi ed esul . Since each si ua ion equi es speci ic ypes o communica ion channels, i makes sense ha he di e en ools he planned pla o m should consis o will be using (a combina ion o ) di e en communica ion channels. In addi ion, depending on he pa ame e s desc ibed abo e, one can choose i ual o physical communica- ion channels o a combina ion o bo h. I is likely ha a combina ion o bo h will be used o an op imal expe ience and deli e y o he pa icipa o y p ocess. The s eng hs o physical ( ace- o- ace) channels a e he pe sonal con ac , he high in e ac ion quali y and dep h and he possibili y o u ilizing non- e bal communica ion (e.g. body lan- guage). These componen s a e e y sui able o accep ance and consensus inding. These bene i s come a he p ice o lowe scalabili y, highe cos s pe pa icipan and he need o mo e expe labou . The ad an ages o i ual me hods a e he high each and scalabili y o in ol e and engage a la ge c owd, which could be e y powe ul in c owd sou cing ools and me hods. 3.3. S a e o he a o me hods Ci izens ha e a b oad ange o possibili ies o communica e in u ban planning like public hea ings, neighbou hood mee ings, communi y ou eaches o ci izen o ums [51]. Since po als as Pa icipedia.ne [52] p o ide a gene al o e iew on me hods o pa icipa ion, ou speci ic in e es was o name me hods speci ically o u ban planning. Fo he esea ch o he ools, in e alia he used me hods o 30 ci ies we e ex ac ed om mul iple publica ions [eg. 53, 54, 55] which we e w i en by he municipali ies o p oclaim hei pa icipa ion concep . Figu e 3. Layou o he pa icipa ion me hod da abase To o ganize hese in o ma ion wi hin a da abase (c . Fig. 3) a ca ego iza ion in o he le el o pa icipa ion as de- 2396 Sande Müns e e al. / P ocedia Compu e Science 112 (2017) 2391–2405 Au ho name / P ocedia Compu e Science 00 (2016) 000–000 sc ibed abo e ook place, whe eby he p io i y we e consul a ion and collabo a ion me hods. In addi ion, se e al o he ca ego ies such as numbe o pa icipan s o leng h o he p ocess we e added. Cu en ly he da abase includes o e 70 me hods. Only me hods o in o mal pa icipa ion a e co e ed, because o mal pa icipa ion in u ban de el- opmen is p ecisely desc ibed in legal amewo ks. Being associa ed o he EU H2020 p ojec “U_CODE U ban Collec i e Design En i onmen ”, he da abase will be launched online wi hin 2017 on he p ojec websi e www.u- code.eu. The esea ch shows ha mos ci ies only ha e a small numbe o me hods, which hey a e used o and which a e no oo complex o unde ake. Common me hods a e di e en low-le el wo kshop o ma s, ci y walks o he handou o in o ma ion ma e ial. Mo e elabo a e me hods such as c owdsou cing pla o ms a e seldom bu g ow- ing. Tha means o u u e pla o ms ha hey should suppo he municipali ies wi h knowhow and conc e e ins uc- ions o assis he use o mo e cus omized me hods. Mos o he esea ched me hods wo k o line (only 14 pu e online me hods), bu many o he o line ools can be assis ed by digi al ools. A pe ec example a e passi e dis- plays (in con as o in e ac i e displays) which can p o ide way mo e in o ma ion and add ess ano he a ge g oup ia a displayed i ual eali y. Fo u u e pla o ms, ha means ha hey should suppo he use o sophis ica ed high-le el echnologies. I mo e complex me hods esul in a be e ou come is ye o be examined. 3.4. U ilizing digi al echnologies o pa icipa ion in u ban planning As men ioned in he desc ibed me hods, only ew o ma s a e capable o managing la ge pa icipan numbe s. A key concep o add ess he p oblem o limi ed each is he combina ion o In o ma ion Technology, Human Com- pu e In e ac ion and u ban planning. In ecen yea s, an inc easing numbe o web pla o ms we e de eloped o sha e in o ma ion abou spa ial p ojec s wi h local people and ge o know hei needs and knowledge [56]. Fo cu en planning cases like he Tempelho Field in Be lin [57], he Reallabo o sus ainable mobili y cul u e in S u ga [58] o o di e en ansi p ojec s in he USA [59], ci y au ho i ies ha e published online pa icipa ion pla o ms. Online communi ies and discussion o ums p o ide excellen pa icipa o y ins umen s o in eg a e expe knowledge and ci izen commi men wi hin decision making: “[In social ne wo ks and in e ne pla o ms] ci izens can be easily in ol ed in impo an decision making p ocesses. […] And ci izens hemsel es become ac i e. In Blogs hey discuss ea ly on abou p ojec s. E en hough ac s may no be ue, hey s ill sp ead excessi ely as . Emo ions o en escala e. P ojec de elope s and in es o s a e well ad ised o ecognise hese channels and con ib- u e o an objec i e discussion […] ea ly on pa icipa ion […] may coun e ac ea s and p ejudices o ci izens.” [47, p. 8]. Fu he oppo uni ies o u ban planning based on communi y engagemen come om c owdsou cing, co- c ea i e digi al media, and neogeog aphy ools [60]. So-called Expanded Pa icipa o y Design [61] “suppo s an expansion o he pa icipa o y e-planning locus owa ds collabo a i e wo k be ween expe s and non-expe s, an- cho ed in mu ual digi al-media p oduc ion and sha ing” [61, p.25]. Saad-Sulonen highligh s he di e ence be ween he collabo a i e app oach and u ban planning whe e planne s “ acili a e and o ches a e he delibe a i e ac i i ies”. Also, pa icipa ion ools in he o ma o “c owdsou cing pla o ms” ha e been es ablished and come in o use o e he pas yea s. Al hough he e is s ill li le esea ch abou hei e icacy and me hodology, social media-based con- ibu ions o open design spaces appea o be a mos p omising a enue also o u ban planning, as online pa icipa- o y design pla o ms like OWELA and RECORD sugges [62, 63]. 3.5. Implica ions o digi al pa icipa ion echnologies in u ban planning Digi al pa icipa ion echnologies ha e he po en ial o ho oughly eshape he p o essional p ac ice in u ban design and de elopmen . I is necessa y and mos meaning ul o enable cons uc i e pa icipa ion and co-design ac i i ies o he public in he ea ly p e-design phases, when changes a e s ill possible on a easible basis [64]. T anspa en ways o decision gene a ion [eg. 50], p ocedu al ai ness and objec i i y [eg. 65] and “win–win” solu ions [eg. 66] a e essen ial. Thei impac can be summa ized by ou new quali ies: • C owdsou cing Knowledge: Digi al pa icipa ion ools enable he u ilisa ion o a wide knowledge basis as in con en ional design p ac ices be o e. I becomes possible o ap on aluable expe iences and c ea i i y o non-p o essionals oo, especially local ci izen expe s. • Design E idence: New echnologies like design sen imen analysis p o ide new e idence o designe s and planne s al eady in he design p ocess. I is possible o de elop p ojec s in acknowledgemen o public e- sponse, and o es public a i ude and accep ance al eady in ea ly p ojec s ages. Sande Müns e e al. / P ocedia Compu e Science 112 (2017) 2391–2405 2397 Au ho name / P ocedia Compu e Science 00 (2016) 000–000 7 • In e ac ion: Besides echnical ad ances, digi al pa icipa ion echnologies p o ide o an in e ac i e and communica ion-o ien ed planning p ocess. Di ec exchange be ween all s akeholde s, especially planne s and use s (ci izens, esiden s) will become a key ac i i y in u ban design. • Agile Design: Pa icipa i e ools and echnologies will ans o m planning wo k in o an i e a i e, agile wo k p ocess, in con as o sequen ial and linea wo k lows ("wa e alls") ha ha e shaped u ban design p ac ice in he pas . 4. Key challenges and p omising app oaches Wha a e majo hu dles o pa icipa i e planning p ocesses and how o o e come? Fo me in es iga ions on public pa icipa ion [6, 9] iden i ied a ious key challenges: 1. Few use s: Especially publically ini ialized pa icipa i e ac i i ies o en lack a su icien numbe o use s. This may be caused by lacking in o ma ion on he p ocess [eg. 67, 68], ba ie s in cul u e, unde s anding o accessibili y [eg. 69] o e en weak mo i a ion o pa icipa e [eg. 59]. 2. W ong use s: As mos pa icipa o y p ocesses emb ace e e ybody, hey a e aced wi h sel -selec ion biases [70, p. 81]. As igu ed ou in many s udies, people willing o pa icipa e in u ban planning p ocesses a ely ep esen a majo i y o inhabi an s o in ol e (po en ial) opinion leade s [eg. 50, 71]. 3. Communica ion issues: Pa icipa o s in public p ocesses a e in luenced by many ac o s such as p io hough s, eelings and belie s [eg. 72], aming e ec s such as p esen a ion o ma s and echniques o me- dia channels [eg. 73], and hei le el o knowledge o he p esen ed p oblem and he objec i es o achie e. Ano he co e p oblem is he communica ion o complex design issues o he public in he ea ly p ojec li e cycle when concep s a e s ill li le shaped and ha d o con ey o non-specialis s, bu easy o in luence and e-di ec . 4. P ocess de ici s: I a public pa icipa o y p ocess lacks anspa ency, inclusion and ai ness, o leads o pub- lically el poo decisions, i could cause mo e se e e disag eemen han i i ne e had aken place. S udying hese challenges and employing insigh s om he in es iga ed cases hese ques ions a ise: • How o each and engage a ep esen a i e g oup o pa icipan s? Mos la ge-scale u ban planning p ojec s ha e a huge amoun o s akeholde s (ci izens, in e es g oups, au ho i ies, poli i ians, e c.). Howe e , he s akeholde s wi h he highes in e es , powe o will o pa icipa e, is mos likely no a ep esen a i e g oup o people who a e a ec ed by he p ojec . Fo success ul public pa icipa ion i is key o p o ide an accu- a e ep esen a ion o all possible opinions in he communi y, meaning a wide a ie y o people om di - e en ages, cul u es and social backg ounds. Especially when using digi al ools, he digi al eadiness, ac- cessibili y and communica ion channels ha e o be conside ed and ho oughly analyzed. A p omising ap- p oach o o e come hese challenges - u he discussed in pa ag aph 4.1 - is o employ mul imodal ap- p oaches o pa icipan selec ion, e.g. by combining online and o line ools. • How o engage and ac i a e cus ome s o pa icipa e in an ea ly s age o he p ojec ? Fi s ly, ci izens do no ca e abou p ojec s un il hey s a a ec ing hem (e.g. 'no in my backya d'). Howe e , wi h e e y p o- jec s age cos s, commi men s, and he le el o elabo a ion inc ease, while isks, solu ion space, and hus he impac o ac i i ies on he p ojec ’s ou come dec ease. Secondly, he conclusions and decisions based on ea ly pa icipa ion may no be ele an anymo e a p ojec comple ion. Especially in ou a ge ed (la ge scale) p ojec s, u ban planning p ojec s can easily ake 15+ yea s om he (p oblem) iden i ica ion s age o ealiza ion. Meanwhile, ci izens’ needs and desi es may ha e shi ed. The use o gami ica ion o e ilize use mo i a ion is p omising app oach - u he discussed in pa ag aph 4.2. • How o communica e a design p ojec o make i unde s andable o he public and allow o collabo a ion? To a oid misunde s andings, non-expe ci izens should be able o unde s and he plans, e en hough hey ind i di icul o unde s and p o essional language, ead plans o echnical d awings, and a e unawa e o ( echnical o p ac ical) cons ain s. Eme ging digi al ools like Vi ual Reali y, Augmen ed Reali y and Ge- oLoca ion may be e y help ul in b idging he communica ion gap be ween p o essionals and ci izens - u - he discussed in pa ag aph 4.3. • How o acili a e pa icipa ion and co-design wi h 10.000+ ci izens? Mee ing and hea ing 100 people du - ing a ace- o- ace wo kshop was a challenge in one o he example cases. As he ac ual each o digi al 2398 Sande Müns e e al. / P ocedia Compu e Science 112 (2017) 2391–2405 Au ho name / P ocedia Compu e Science 00 (2016) 000–000 pla o ms can easily be 10.000+ people, how o make su e hose people s ill eel hea d and can make a sig- ni ican con ibu ion? App oaches om compu e linguis ics such as a sen imen analysis discussed in pa - ag aph 4.4 a e dedica ed o make la ge scale use eedback easible and ensu e anspa ency be ween a pub- lic o e and sugges ions o u he planning. 4.1. Pa icipan selec ion One o he p ominen heo e ical and p ac ical challenges in pa icipa o y u ban planning is he non- ep esen a i e pa icipan selec ion [30, 74]. Cu en ways o o ganize public pa icipa ion a e: 1. In i ing pa icipan s ep esen ing s akeholde o ganiza ions [9] 2. In i ing all membe s o he ele an public by mail, pape s, billboa ds, adio and ele ision [9, 75] 3. Using online su eys o each he comple e ele an public [76] As has been a gued [9] he i s way will lead o he phenomenon called “usual suspec s” wi h ob ious challenges. The las wo ha e ano he challenge o cope wi h called “non- esponse bias” o “selec ion bias”. The h ead o “non- esponse bias” in adi ional sample su eys is long known [see .i. 77]. The highe he h eshold o pa icipa- ion, he bigge he isk o “non- esponse”. So mail su eys show an e en lowe “non- esponse” a e han pa icipa- ion mee ings a ci y own halls [78]. “Non- esponse” leads o an inc ease in a iance as a esul o a educ ion in he ac ual size o he sample and he ecou se o impu a ion. This p oduces a bias i he non- esponden s ha e cha ac e - is ics o in e es ha a e di e en om hose o he esponden s. Fu he mo e, he e is a isk o signi ican ly unde - es ima ing he sampling e o , i impu ed da a a e ea ed as hough hey we e obse ed da a [79, p.59]. Jüni and Ma hias [80] desc ibe a o m o selec ion bias in hei esea ch called a i ion (loss o pa icipan s). This o m o selec ion bias in ol es d opou , non- esponse, wi hd awal and p o ocol de ia o s. In hei esea ch, he example o a die ing es is men ioned. In his example, he esea che simply ejec s e e yone who d ops ou o he ial, when mos o he ones who d op ou a e hose o whom he ea men isn’ wo king. We see he same e ec s happen in pa icipa ion in u ban planning. Conce ning he hi d way, he phenomenon o sel -selec ion bias leads o di e ences o up o 300% in esul s e ie ed om a o al popula ion in con as o sel - ec ui ing pa cipan s in online su eys [81]. In li e a u e h ee mode n ways o o e come “selec ion bias” can be ound: 1. Finding pa icipan s in ele an s akeholde o ganiza ions, bu h ough opinion leade s [see .i. 82] 2. Sampling by ep esen a ion ia andom lo e ies [76] 3. Sampling ia indi ec me hods [83] All h ee mode n ways a e p omising a enues o p e en ing selec ion bias in public pa icipa ion in u ban planning. In ou i s ound o p o o yping [83], some i s esul s we e ound in an online ool o public pa icipa ion p o id- ed his online ool was used in combina ion wi h o line ools. I will need u he esea ch and p o o yping o ind a good mix o ools o public pa icipa ion ha will e lec he public opinion co ec ly wi hou selec ion bias. 4.2. Gami ied app oaches o pa icipa ion Gami ica ion as he “use o game design elemen s in non-game con ex s” [84] p o ides mechanisms o enhance use mo i a ion o pa icipa e and con ibu e o planning p ocesses. In o de o e ie e in o ma ion on his subjec , we pe o med a li e a u e e iew including he da abases EBSCOHos , ACM Digi al Lib a y and Scopus. Table 2. Ca ego ies o ci izen pa icipa ion in u ban planning [85] Ca ego y In o ma ion Consul a ion Collabo a ion Desc ip ion Ci izens in o m o a e being in o med on ecen o u u e plans, decisions and ac ions Ci izens a e asked o gi e inpu and eedback (e.g. opinions, solu ions) Ci izens and o he s ake- holde s ac i ely wo k oge he in decision- making Rela ion One-way Limi ed wo-way Ad anced wo-way Besides, by using he snowball p inciple, addi ional wo ks we e ound ha we e no co e ed by he da abase-d i en keywo d sea ch. As discussed in de ail in [86], 188 publica ions we e e ie ed. Fo a clus e ing o indings we Sande Müns e e al. / P ocedia Compu e Science 112 (2017) 2391–2405 2399 Au ho name / P ocedia Compu e Science 00 (2016) 000–000 9 employed me hods o deduc i e quali a i e con en analysis [eg. 87], based on bo h heu is ic amewo ks [c . 88, 89] and induc i e s ages o quali a i e con en analysis. Acco ding o [85], h ee ways o ci izen pa icipa ion in u ban planning can be dis inguished (c . Tab. 2). When i comes o in o ming he public, isualiza ion echniques such as augmen ed eali y (AR) o e a a ie y o inno a i e solu ions o e ec i ely suppo in o ma ional pu poses, including mobile de ices o display planning in-si u [90, 91]. Tha app oach can easily be gami ied, e.g. by imple- men ing asks and high-sco es. In e ms o consul ing he public in ma e s o u ban planning, o ma s ocus on ci izens as local expe s and s i e o collec neighbou hood knowledge by using mobile da a collec ion ools [92] such as “S e eopublic” and “Widenoise” o cap u e noise le els and display hem on a map [93] o o o e on u ban planning ideas using high-sco e lis s o aising use mo i a ion [93]. Ini ia ing an in ense p ocess o collabo a ion be ween he public and p o essionals is he in en ion o a gami ied mobile app called “Communi y Ci cles”. I e- wa ds use in e ac ions and con ibu ions ha e e o local u ban planning issues wi h digi al poin s [45]. Finally, he web-based online pla o m “Communi y PlanI " is designed o assis ing u ban planning mee ings, ans o ming an u ban ins ance in o a “mission” ha con ains game elemen s such as challenges, leade boa ds and in-game e- wa ds. By comple ing hose missions, he ci izen con ibu es o he planning p ocess, ea ning i ual coins which can be spen o suppo u ban concep s ha ame he opic [93]. Wha a e implica ions o a pla o m design? Fi s o all, mobile de ices ob iously lowe ba ie s o pa icipa ion and aise he chance o add essing a wide ange o pa icipan s. Mo eo e , hey allow displaying u ban design p oposals igh a he spo , which may os e in e es in public pa icipa ion. Fu he mo e, gami ica ion s a egies show po en ial no only o a ouse cu iosi y o pa icipa- ion o ma s bu also o imp o e long- e m use mo i a ion o pa icipa e. Commen ing and a ing design p oposals, sha ing own ideas o playing goal- ela ed p ojec missions can be ewa ded wi h poin s o badges, while o ma s o disco e y mo i a e people o explo e hei dis ic in o de o ind u ban issues o public in e es . A la e s ages, isualiza ion echniques such as 3D en i onmen s and augmen ed eali y ske ches o u u e buildings show po en ial o make u ban p ojec s mo e exci ing and angible. 4.3. Communica ing & decision making One key challenge igge ed by he applica ion o digi al co-design ools in u ban planning is he o e all decision making p ocess. In he con e gen (" a ional") phases o a design p ojec , esponsible decisions mus be made in o de o selec an app op ia e solu ion om a mul i ude o al e na i es c ea ed in ea lie di e gen ("c ea i e") phas- es. Besides he in o ma ion ha is gi en a public e en s o ci izens, online pla o ms p o ide a g ea e possibili y o sha e and ga he da a belonging o a p ojec . This includes p oposals om he ci y, budge , ime schedules, di e en media o ma s like pho os, ideos o 3d-models, commen s, people in ol ed, esul s o o ings and/o ankings, and o example da a collec ions h ough websi e analysing ools. Bu e en echnical da a (e.g. legal amewo k, dimen- sions o cons uc ions) should be a ailable o he c owd [94]. The amoun o da a depends on he le el o pa icipa- ion [95]. A a lowe le el he con en comes om policy-make s, s akeholde o p o essionals ( op-down). The mo e he ci izens a e allowed o pa icipa e, he mo e he e will be a c owdsou cing p ocess (bo om-up) [96]. Impli- ca ions o a well done pla o m design a e di e se and a ec he in o ma ion i sel placed on he si e as well as he way o communica e. The p oblem should be announced clea ly wi h all necessa y da a ha is a ailable, wi h an in e es ing agenda, he goals ha ha e o be eached and why he public is needed [97]. Only hen he c owd will be ac i a ed and be c ea i e o sol e issues[98]. The design o he use in e ace has o be usable o all po en ial use s and allow accessibili y o all de ices (e.g. sma phone, PC). Mo eo e , he in o ma ion abou a p ojec has o be anspa en , well unde s andable and unbiased o a oid aming e ec s [99]. I is e y impo an ha all use s ha e a common unde s anding o he same subjec and do no eac in di e en ways. In he Bos on Sou hwes Co ido p ojec , C ewe [100] ound ou ha “ he mo e designe s alue he inpu o ci izens, he mo e app op ia e hei de- signs will be o he use s conce ned“. Fu he mo e, all in ol ed pe sons (s akeholde s, planne s, designe s, ci y au ho i ies and he ci izens) ha e o ake pa wi hin he communica ion p ocess because all o hem ha e needs and in e es s. I all in ol ed pa ies ake pa in he decision-making p ocess, hen you will show ha you ake he ci izen se iously and o a oid he isk o esis ance [101]. Wi h he help o icons, pic u es, labels and u he desc ip ions, he di e en oles can be made isible o e e y end use . Co esponding o he p o ile, use s ha e di e en igh s, asks and possibili ies o a ec he pa icipa ion p ocess. As desc ibed in pa ag aph 3.1 he pa icipa ion p ocess may consis s o se e al ounds wi h di e en asks and issues. Fo design pu poses digi al pla o ms ha e o com-