GIS-data related route optimization, hierarchical clustering, location optimization, and kernel density methods are useful for promoting distributed bioenergy plant planning in rural areas

This is a sel -a chi ed e sion o an o iginal a icle. This e sion may di e om he o iginal in pagina ion and ypog aphic de ails. Au ho (s): Ti le: Yea : Ve sion: Copy igh : Righ s: Righ s u l: Please ci e he o iginal e sion: CC BY-NC-ND 4.0 h ps://c ea i ecommons.o g/licenses/by-nc-nd/4.0/ GIS-da a ela ed ou e op imiza ion, hie a chical clus e ing, loca ion op imiza ion, and ke nel densi y me hods a e use ul o p omo ing dis ibu ed bioene gy plan planning in u al a eas © 2019 Else ie L d. All igh s ese ed. Accep ed e sion (Final d a ) Laasasenaho, K.; Lensu, Anssi; Lauhanen, R.; Rin ala, J. Laasasenaho, K., Lensu, A., Lauhanen, R., & Rin ala, J. (2019). GIS-da a ela ed ou e op imiza ion, hie a chical clus e ing, loca ion op imiza ion, and ke nel densi y me hods a e use ul o p omo ing dis ibu ed bioene gy plan planning in u al a eas. Sus ainable Ene gy Technologies and Assessmen s, 32, 47-57. h ps://doi.o g/10.1016/j.se a.2019.01.006 2019 GIS-da a ela ed ou e op imiza ion, hie a chical clus e ing, loca ion1 op imiza ion, and ke nel densi y me hods a e use ul o p omo ing2 dis ibu ed bioene gy plan planning in u al a eas3 K. Laasasenahoa*, A. Lensub*, R. Lauhanenc and J. Rin alaa 4 aFacul y o Enginee ing and Na u al Sciences, Tampe e Uni e si y, FI-33014 Tampe e5 Uni e si y , Finland; b Depa men o Biological and En i onmen al Science, Uni e si y6 o Jy äskylä,, Jy äskylä, Finland; cSchool o Food and Ag icul u e, Seinäjoki7 Uni e si y o Applied Sciences, Seinäjoki, Finland8 ka i.laasasenaho@ uni. i (K. Laasasenaho); anssi.lensu@jyu. i (A. Lensu)9 *co esponding au ho s10 11 GIS-da a ela ed ou e op imiza ion, hie a chical clus e ing, loca ion12 op imiza ion, and ke nel densi y me hods a e use ul o p omo ing13 dis ibu ed bioene gy plan planning in u al a eas14 Cu en ly, geog aphic in o ma ion sys em (GIS) models a e popula o s udying15 loca ion-alloca ion- ela ed ques ions conce ning bioene gy plan s. The aim o his16 s udy was o de elop a model o in es iga e op imal loca ions o wo di e en 17 ypes o bioene gy plan s, o a m and cen alized biogas plan s, and o wood18 e minals in u al a eas based on minimizing anspo a ion dis ances. The19 op imal loca ions o biogas plan s we e de e mined using loca ion op imiza ion20 ools in R so wa e, and he op imal loca ions o wood e minals we e de e mined21 using ke nel densi y ools in A cGIS.22 The p esen case s udy showed ha he u ilized GIS ools a e use ul o 23 bioene gy- ela ed decision-making o iden i y po en ial bioene gy a eas and o24 op imise biomass anspo a ion, and help o plan powe plan sizing when25 candida e bioene gy plan loca ions ha e no been de ined in ad ance.26 In he s udy a ea, i was possible o ind logis ically iable loca ions o 13 a m27 biogas plan s (>100 kW) and o 8 cen alized biogas plan s (>300 kW) using a28 10-km h eshold o eeds ock supply. In he case o wood e minals, he esul s29 iden i ied he mos in ensi e wood ese es nea he highes oad classes, and wo30 po en ial loca ions we e de e mined.31 Keywo ds: biogas; ci cula economy; loca ion-alloca ion; ne wo k analysis;32 wood e minal33 1 In oduc ion34 Cu en ly, biomass is he mos used enewable ene gy sou ce in he wo ld [1]. Biomass35 om plan s, o ganic was e, and animal exc e a is equen ly u ilized in bioene gy36 p oduc ion. In u al a eas, se e al ypes o biomass a e a ailable o bioene gy37 p oduc ion depending on local ac o s, such as p esence o ag icul u al esidues (e.g.,38 s aw and manu e) and a ailabili y o o es biomass. Bioene gy and bio uels can be39 c ea ed om biomass h ough se e al echniques, including mechanical, chemical, o 40 biological ea men s such as pelle izing, gasi ica ion, py olysis, o biological p ocesses41 [2]. In ac , he use o biomass o bioene gy p oduc ion appea s o be inc easing, and42 he di e en applica ions o biomass a e expanding because o he shi ing end owa d43 bio and ci cula economies ha eplace adi ional ossil esou ces [2-4]. In di e en 44 u al a eas o Eu ope, in es men in biogas plan s using manu e as uel a e inc easingly45 conside ed while he use o wood biomass as such o as pelle s in bioene gy plan s is46 p omo ed as well. In his con ex , he a ailabili y o biomass o bioene gy p oduc ion47 mus also be gua an eed in he u u e.48 1.1 Planning o bioene gy p oduc ion49 S akeholde s play an impo an ole h oughou he a ious phases o bioene gy50 de elopmen p ojec s om he bioene gy plan planning o p ojec implemen a ion. By51 in eg a ing he di e en s akeholde s, i is possible o iden i y condi ions ha a e52 applicable o bioene gy [5]. Planning loca ions o bioene gy plan s is usually a53 demanding ask because p ecise knowledge abou biomass a ailabili y, yield, and54 chemical cha ac e is ics a e equi ed. Besides he loca ion o he ac ual bioene gy plan ,55 he need o in oduce wood e minals has become especially u gen in No he n56 coun ies o balance he loca ion o wood supplies and o con en ional combined hea 57 and powe (CHP) bioene gy plan s. This is as adi ional win e ime ha es ing o wood58 is becoming di icul due o wa ming win e s, leading o a lack o ha dening os on59 oads wi h low bea ing capaci y [6].60 The ounding o a new bioene gy plan is always a geospa ial ques ion. Biomass61 esou ces a e usually spa sely dis ibu ed, making e e y case unique [7]. Di e en 62 biomasses ha e di e en yields, yea ly schedules, and cha ac e is ics and, acco dingly,63 ha e dis inc economic alues, which in luence, o example, he economic easibili y64 o he equi ed anspo a ion dis ances. The eby, one c ucial s ep o es ablishing65 bioene gy plan s is inding iable loca ions o hem. Me hods based on geog aphic66 in o ma ion sys ems (GISs) ha e been used in many disciplines as decision-making67 ools because hey can sol e loca ion-alloca ion- ela ed p oblems h ough, o example,68 minimizing anspo a ion dis ances [8-9].69 1.2 Feasibili y o GIS ools o alloca ing biomass esou ces o bioene gy70 Globally, se e al s udies ha e mapped biomass esou ces o bioene gy p oduc ion. In71 gene al, s udies can be di ided in o wo GIS-based app oaches: sui abili y analyses and72 op imali y analyses. In sui abili y analyses, which a e some imes called mul i-c i e ia73 e alua ions (MCEs), bu e and spa ial o e lay analyses a e usually used o assess he74 loca ion o po en ial biomass p oduc ion plan s, whe eas op imali y analyses a e used75 o loca ion-alloca ion issues o ma ch biomass supply and he ene gy demands o 76 socie y [9]. Sui abili y analyses ha e been p e iously based on he in eg a ion o 77 di e en models o analy ical echniques in o a GIS en i onmen , including Ma ko 78 chains [10], mul i-c i e ia models [11˗12], analy ic hie a chy p ocess and map algeb a79 [13], and ke nel densi y analysis [14]. Meanwhile, op imali y analyses o bioene gy80 plan s ha e been based on Dijks a’s ou e op imiza ion algo i hm [15], emo e sensing81 da a and GIS-based mixed in ege linea modeling [16], and he modi ied p-median82 p oblem [9]. Many o he s udies using GIS ha e di ec ly examined o assessed gene al83 biomass po en ial o bioene gy p oduc ion [e.g., 17˗22]. Also, some s udies a e84 handling analy ical me hodologies and de elopmen o heu is ics in bioene gy supply85 chain [23]. GIS me hods a e especially use ul in assessing land a ailabili y o ene gy86 c ops [11˗13, 17]. In addi ion, sus ainabili y o bioene gy p ojec s could be imp o ed by87 combining Li e Cycle Assessmen s and GIS ools [24].88 Feasible biomass anspo a ion dis ance is eeds ock dependen and is a ec ed89 by se e al ac o s. The economics o biomass anspo a ion dis ances a e dependen ,90 o example, on biomass composi ion, ene gy alue (e.g., biogas po en ial), mois u e,91 speci ic weigh [25], and aile capaci y. In addi ion, local egula ions a ec was e-92 based managemen p ocedu es and anspo a ion p ac ices, and he e o e ha e a no able93 ole in bioene gy planning [26]. GI Sys ems p o ide se e al ools o sol ing op imal94 logis ic solu ions and minimizing biomass anspo a ion cos s, bu mos o he ools95 equi e ha he use speci ies bo h sou ce and des ina ion loca ions o he anspo s.96 When planning a loca ion o a new acili y, his would equi e p o iding se e al97 possible plan loca ions (des ina ion candida es), and hen we could choose he bes 98 candida e. I such candida es do no exis o i we do no wan o limi he sea ch o 99 bes loca ion o such se o candida es, he ou e op imiza ion me hodology needs o be100 al e ed o op imize ou es om he sou ce poin s o all o he loca ions in he oad101 ne wo k, as we ha e done in his s udy. Taking bo h anspo a ion dis ances and102 biomass supply in o accoun , he op imal size and loca ion o plan s can be de e mined.103 The aim o he p esen s udy was o de elop and assess he easibili y o a GIS-104 based solu ion o selec ing he op imal loca ion o biogas plan s and wood e minals in105 a u al a ea based on minimizing he anspo a ion needs o di e en biomasses. The106 op imal loca ions o biogas plan s and wood e minals we e he e o e de e mined in he107 s udy a ea conside ing spa sely dis ibu ed biomasses. The aim was o c ea e a model108 ha can help local s akeholde s o op imize bioene gy plan loca ions and o de elop109 bioene gy and bio- e ining-based business ac i i ies in u al a eas.110 2 Ma e ials and me hods111 2.1 S udy a ea112 The s udy a ea co esponded wi h he u al Kuudes aan egion in Sou h Os obo hnia,113 Finland. The o al a ea o he egion is 3,121 km2, and he egion con ains 23,646114 inhabi an s [27˗28] ha mos ly li e in wo majo owns (Äh ä i and Ala us wi h 5,968115 and 11,746 inhabi an s, espec i ely). One hund ed and hi y- i e la ge a ms116 (desc ibed in mo e de ail la e ) a e p esen in he egion, and he economy o he egion117 has been adi ionally based on o es y ac i i ies. Cu en ly, he po en ial eeds ocks o 118 bioene gy p oduc ion a e wood, ag icul u al esidues (e.g., s aw and manu e), and119 municipal o ganic was es. Wood is commonly used as uel o hea p oduc ion in120 dis ic hea ing plan s (12 plan s) and in p i a e houses, including a mhouses. The e a e121 h ee majo wood e minals (1–2 ha in size, Me sä G oup) whe e wood is empo a ily122 s o ed and hen anspo ed o a pulp mill and bio e ine y loca ed in Äänekoski, Cen al123 Finland (a e age dis ance o 100 o 150 km). Howe e , so a , no biogas plan s a e124 p esen in he s udy a ea.125 126 Figu e 1. Loca ion o he s udied Kuudes aan egion in Finland. Popula ion cen es127 (adminis a i e bo de s) a e indica ed in da k g ey. Municipali y names a e indica ed by128 capi al le e s, and some majo illages by lowe case le e s.129 2.2 Scena ios and da a o biomass esou ces130 In his s udy, wo biomass use scena ios we e s udied. The i s one aimed o ind131 loca ions o biogas plan s wi h capaci ies om 100 o o e 300 kW. The capaci ies132 we e based on economically easible a m biogas plan and cen alized biogas plan 133 sizing in Finland acco ding o Na u al Resou ce Ins i u e Finland [29]. The o he 134 scena io aimed o loca e wood e minals in he s udy a ea (Fig. 4).135 In he biogas scena io, eeds ocks included di e en manu es om a ms,136 sepa a ed biowas es om municipali ies, oca ional schools, g oce y s o es, and ou is 137 cen es (which is biowas e om ca e ing se ices, bu also includes biowas e and animal138 manu e om Äh ä i Zoo); and sludge om was ewa e ea men plan s (Table 1).139 Fu he mo e, he use o eed cana y g ass (RCG; Phala is a undinacea), which can be140 po en ially g own on cu away pea lands, was conside ed [30]. In ensi e pea ex ac ion141 egions a e p esen in he s udy a ea, and hund eds o hec a es o hese si es will en e 142 in o he a e -use phase in he nea u u e and hus ep esen po en ial g owing si es o 143 ene gy c ops.144 Manu es ( o al 264,273 ) om la ge a ms wi h mo e han 50 heads o ca le,145 500 pigs, 30 ho ses, o 500 heads o poul y in 2016 we e included in he s udy. Thei 146 loca ions (add esses) we e ob ained om he da abases o he Finnish Food Sa e y147 Au ho i y (E i a), he Agency o Ru al A ai s (Ma i), and he Na ional Land Su ey148 o Finland (NLS). The amoun o manu e p oduced pe a m was calcula ed based on149 animal age and species, and he mean amoun s o manu e p oduced pe animal [31].150 The amoun ( esh ma e ; FM) o biowas es was ob ained om he municipali ies and151 ope a o s o municipal was e collec ing se ices. The amoun s ( o al solids; TS) o 152 sewage sludge (municipali ies o Ala us, Äh ä i, and Soini) we e ob ained om he153 En i onmen al P o ec ion da abase [32]. The me hane po en ial o di e en biomasses154 a e p esen ed in Table 2.155 Table 1. Annual amoun s o manu e and biowas e (Mg FM) and sewage sludge (Mg TS)156 gene a ed in he s udy a ea.157 O ganic was e Amoun Ag icul u al manu e 264,273 Biowas e - municipal 127 - shops 103 - ou is cen es 306 - oca ional schools 4 Sewage sludge 494 158 Table 2. The me hane po en ial o di e en biomasses used in his s udy.159 Biomass CH4 po en ial Uni Re e ence Biowas e 107 m3 CH4/Mg FM [33] Sewage sludge 163 m3 CH4/Mg TS [33] Ca le manu e 19 m3 CH4/Mg FM [33] Pig manu e 10 m3 CH4/Mg FM [31] illus a ed in Fig. 8, whe e he op imal biogas plan loca ion was de ined based on 6259 a ms, o ganic was e om 7 municipal sou ces, 1 ou ism cen e, and 2 g oce y s o es.260 Li es ock manu e was he la ges sou ce o biomass o he po en ial biogas261 plan s. Howe e , in Ala us and Kuo ane, a signi ican inc ease o me hane po en ial262 was achie ed h ough combining manu e wi h biowas e and po en ial RCG cul i a ion263 in cu away pea lands. The la ges me hane ene gy p oduc ion po en ial is in he wes e n264 pa o he egion, whe eas no po en ial loca ions o biogas ins alla ions we e iden i ied265 in he Soini municipali y (Fig. 6). In pa icula , la ge-scale a ms (o e 50 heads o 266 ca le) ha e an impo an ole in u u e biogas p oduc ion in he s udy a ea. Only abou 267 20 % o he dai y a ms ha e o e 50 heads o ca le in Finland and he numbe o small268 a ms is cons an ly dec easing [41]. Fo example, in Denma k, he la ges a ms we e269 also iden i ied, in mos cases, as ele an and i al o u u e biogas p oduc ion; a e age270 a m size in Denma k has inc eased om 131 heads o ca le o 238 heads pe a m in271 en yea s ( om 1999 o 2009 [10]). Ag icul u al esidues, including slu ies and c ops272 p oduced o ene gy, ha e also been ound o be impo an biomass sou ces in o he 273 egional GIS-based biogas analyses, ep esen ing om 50% o o e 90% o o al biogas274 po en ial [13˗14] in s udied u al a eas.275 No ably, he cul i a ion o RCG on p io cu away pea lands was con i med as a276 po en ial eeds ock sou ce o a biogas plan in an a ea o in ensi e pea ex ac ion. In277 Ala us, he e we e wo a eas whe e cul i a ion o RCG on p io cu away pea lands278 could inc ease local biomass esou ces (Fig. 6). Howe e , he cu away pea lands a e279 usually loca ed o e 10 km away om a ms, which can make he logis ic a angemen s280 di icul (Fig. 7). Also, ce ain limi a ions o cu away pea lands mus be add essed, such281 as he di icul y o cul i a ing ag icul u al c ops in hese a eas because o ypical high282 wa e le els [42]. Al e na i ely, o es biomass could be conside ed on emo e cu away283 pea lands, and in ac landowne s gene ally p e e o es y as an a e -use al e na i e284 ins ead o ene gy c op p oduc ion [30].285 The ac ha sludge gene a ed in was ewa e ea men plan s and biowas e286 gene a ed in municipali ies and ou is cen es a e o en loca ed a away om la ge-287 scale a ms complica es he loca ion o he biogas plan . Howe e , in Kuo ane, he288 inc ease o ene gy po en ial is achie ed om combining he join me hane po en ial o 289 biowas e ( om one ou is cen e, g oce y s o es, and municipal collec ion acili ies)290 and la ge-scale a ms nea he own cen e (Fig. 8). Finally, he GIS ools used in he291 p esen s udy alloca ed biomasses acco ding o easonable anspo a ion dis ances (10292 km) and helped o plan biogas plan s sizing.293 294 Figu e 5. S udied eeds ocks and hei g oss biogas po en ials (MWh) in he s udy295 egion as box-and-whiske s plo s.296 297 Figu e 6. Feeds ock p oduc ion si es and hei di ision in o clus e s (gi en as numbe s)298 o he 13 po en ial a m biogas plan s (la ge ci cles) (>100 kW) and eigh po en ial299 cen alized biogas plan clus e s ( illed wi h colou ) (>300 kW).300 301 302 Figu e 7. Dend og am p esen ing cen alized biogas plan clus e s acco ding o a303 anspo a ion h eshold o 10 km in he s udy egion. Agglome a i e clus e ing based304 on comple e linkage was applied o combine biomass p oduc ion si es as clus e s when305 h eshold dis ance was no exceeded. Rec angles a e used o indica e clus e s, and306 symbols indica e ype o biomass. Biomass poin s (n = 189) and clus e IDs ( o al 43)307 a e indica ed in he bo om le co ne . Me hane po en ials a e gi en in blue.308 309 310 Figu e 8. Example o he sel -p og ammed op imiza ion ool o iden i ying a sui able311 loca ion o a powe plan by minimizing anspo a ion dis ance when biomasses a e312 spa sely dis ibu ed in a po en ial biogas p oduc ion a ea (clus e 19). The model313 minimizes he sum o o al anspo a ion needs. The po en ial plan loca ion is314 illus a ed as an as e isk (ene gy po en ial indica ed below he as e isk in MWh/a). The315 biomass sou ces a e deno ed wi h symbols explained in he legend o Fig. 7.316 3.2 Wood e minal scena io317 The op imal loca ions o wood e minals in he s udy a ea we e de e mined based on318 ke nel densi y analyses along wi h oad ne wo k da a (Fig. 9). The denses wood319 esou ces a e loca ed in no he n pa s o he Soini municipali y and on he bo de o he320 Äh ä i and Ala us municipali ies (Fig. 9). The oad ne wo k co e s he la e a ea qui e321 well, especially conside ing ha a class 1 oad (i.e., a highway) c osses he a ea (Fig. 9).322 In he case o Soini, he wood esou ces a e loca ed nea a oad ne wo k o lowe quali y323 class. Howe e , all he oads in he s udy a ea a e s ill sui able o uck anspo a ion324 o wood biomass.325 Wood e minals in hese p io a eas (Fig. 9) could be conside ed i wood326 p ocessing a in e media e e minals becomes popula , which would imp o e he327 balance o he wood supply and inc ease he need o wood s o age capaci y. The328 calcula ed wood e minal loca ions, howe e , we e no equal o eal exis ing e minals329 (Fig. 9). The e minal loca ions in Ala us, a he Äh ä i (Myllymäki) ailway s a ion,330 and in he cen e o Soini municipali y we e no cong uen wi h eal wood a ailabili y.331 Fu he mo e, exis ing e minals loca e in good logis ical si es (nea ailways and ucks)332 o p omo e la ge-scale wood u iliza ion wi hou conside ing o es esou ces. E en so, i 333 may be a ional o es ablish small-scale e minals o se e mo e local bioene gy plan s334 in he spo s ound in he p esen s udy (e.g., [43]).335 Fu he , in eali y, di e en wood p ocu emen o ganiza ions do conside he336 well-being o he oad ne wo k and en i onmen al limi a ions [45˗46], which he337 applied me hods in his s udy do no au oma ically conside . When linking limi ed338 model calcula ions and eal wood p ocu emen oge he , expe knowledge and339 consensus solu ions can be used in decision-making (e.g., [46]). Fo example, in la e340 sp ing ime, he e can be weigh limi a ions on local oads, and he d i e- h ough o 341 imbe ucks is o bidden.342 343 Figu e 9. Ke nel densi y map o wood esou ces ( ee s and olume in m3 ha-1) in he344 s udy a ea. Da ke colou s indica e g ea e densi y o wood esou ces ( o es in en o y345 da a [37]; oads [39]; municipal bo de s [38]). Po en ial wood e minal loca ions a e346 loca ed in a eas wi h dense wood esou ces nea he highes oad classes (in da ke 347 blue). Colou ep esen s ela i e densi y and no speci ic uni s.348 349 3.3 Feasibili y o he me hods o de ining he loca ions o bioene gy plan s350 The p esen s udy de eloped and assessed me hods consis ing o ou e op imiza ion,351 hie a chical clus e ing, loca ion op imiza ion, and ke nel densi y es ima ion o 352 iden i ying biomass p ocessing o s o ing loca ions in cases o mul iple eeds ock such353 ha anspo a ion dis ances a e minimized. The me hods op imize biomass354 anspo a ion om he collec ion poin o non-p ede ined powe plan loca ion, which355 shows he p og ess oge he wi h p e ious s udies using di e en GIS on bioene gy356 plan planning as summa ized in Table 3. The goal was o achie e he highes po en ial357 bioene gy p oduc ion and plan size wi h sho anspo a ion dis ances om collec ion358 poin s o all o he loca ions in he oad ne wo k. The esul s show ha hese me hods a e359 sui able o alloca ing biomass o bioene gy in u al a eas and he me hods can be360 conside ed as decision-making ools o help plan powe plan size.361 The op imiza ion me hods applied in his s udy p omo e he use o GIS ools in362 bioene gy planning. The same kind o R analyses ha e no p e iously been used in363 biogas plan planning while e.g. ke nel densi y analyses we e used in loca ion biogas364 plan s in Sou he n Finland (e.g., [14]).365 In u al a eas, i is impo an o include in he model he oad ne wo k and no 366 only Euclidean dis ance because geog aphic obs acles such as lakes and moun ains can367 a ec he s uc u e o he oad ne wo k in many cases. Fo example, in he p esen 368 s udy, he oad ne wo k conside ed he lakes, which o ms app oxima ely 7% o he369 o al s udy a ea, and only a ew o hem can be c ossed by using b idges [35].370 Consequen ly, he s uc u e o a oad ne wo k has an essen ial ole in anspo a ion371 cos s.372 Table 3. Selec ed GIS based decision suppo models s udied o di e en bioene gy373 applica ions.374 GIS me hod The me hod can be used o Re e ence Ma ko chain model Fo ecas ing he spa ial dis ibu ion o Danish li es ock in ensi y and u u e biogas plan s [10] Mixed in ege linea p og amming model Bio e ining plan loca ion op imiza ion by emo e sensing and oad ne wo k [16] GIS – Analy ical Hie a chy P ocess – Fuzzy Weigh ed O e lap Dominance (GAF) model Decision suppo on sui able loca ions o biogas plan s [12] Ke nel densi y and p-median p oblem Pinpoin ing a eas wi h high biome hane concen a ion (Ke nel densi y). Whe eas p- median p oblem is applied by choosing acili ies such ha he o al sum o weigh ed dis ances alloca ed o a acili y is minimized [14] Modi ied p-median p oblem E alua ing biomass supply ca chmen s (an ex ension o he p-median model) [9] Modi ied Dijks a algo i hm A sys emic app oach o op imizing animal manu e supply om mul iple small scale a ms o a bioene gy gene a ion complex including concep ual modelling, ma hema ical o mula ion, and analy ical solu ion. [15] A Mul ic i e ia Spa ial Decision Suppo Sys em in eg a ed wi h GIS/ELECTRE TRI me hodology Add essing eal-wo ld p oblems and ac ual in o ma ion (e.g. soil ype, slope, in as uc u es) in biogas plan s si e selec ion. [11] The analy ical hie a chical p ocess (AHP) Decision suppo p ocess, which cap u es quali a i e and quan i a i e aspec s o in o ma ion (such as en i onmen and economy) in o GIS en i onmen o he si ing o anae obic co-diges ion plan s [13] 375 The me hod desc ibed in he p esen pape can be use ul o municipal-le el376 business de elope s and o p omo ing business ac i i y in u al a eas. The me hod377 helps o ecognize ene gy po en ials by clus e ing he eeds ocks and by inding378 ho spo s wi h ke nel densi y analysis. In pa icula , he biogas plan op imiza ion379 scena io was use ul o iden i ying po en ial a eas o bioene gy p oduc ion gi en380 mul iple po en ial eeds ocks. Fu he , he sel -p og ammed ool can help o op imize381 biogas plan loca ions by minimizing anspo a ion cos s, especially in si ua ions when382 candida e biogas plan loca ions ha e no been de ined in ad ance. Many GIS ools,383 such as e.g. Closes Facili y and Loca ion-Alloca ion in A cGIS, equi e such candida e384 poin s. One clea ad an age o his me hod is also ha he con igu a ion o biomass385 sou ces can be easily changed and he analysis can be e- un i some a ms decide o386 lea e ou om p oposed coope a i es.387 The assessed op imiza ion model can make loca ion de e mina ion easy when388 cen alized biogas plan s a e planned. Di e en ne wo k analyses and adjus ing he389 anspo a ion h eshold limi (10 km) lowe o highe could p o ide di e en 390 alloca ions o logis ical solu ions o biogas plan loca ion. Fo example, by adjus ing391 he h eshold limi o 12 and 15 km, he numbe o po en ial clus e s inc eases o 9 and392 11, espec i ely. The biogas plan s a e o en placed nea he spa ial mean o biomass393 sou ces, because in many cases he e a e se e al a he la ge biomass sou ces. In hese394 cases, he anspo a ion dis ances would s ill be less han 10 km, because he dis ances395 om biomass sou ces o he cen ally loca ed plan a e usually smalle han he396 maximum dis ance be ween he biomass sou ces. Also, i is possible o balance397 biomasses be ween clus e s a e wa ds o each an e en mo e e en dis ibu ion o 398 loca ions conside ing biogas po en ial among all clus e s.399 Acco ding o he applied biogas plan loca ion op imiza ion me hod, he simples 400 anspo a ion si ua ion is in hose la ge a ms (a leas 4,500 Mg o cow manu e pe 401 yea ) which a e conside ing he cons uc ion o a m biogas plan (>100 kW o g oss402 powe capaci y). In p ac ice, his means app oxima ely 200 dai y cows o abou 300403 bulls a a m. In hese cases, i may be easy o b ing addi ional eeds ock om smalle 404 a ms, because he manu e quan i ies in hem a e smalle and he eby anspo needs405 along he oads a e minimized. Acco ding o he op imiza ion model, he biogas plan 406 localisa ion si ua ion is pa icula ly demanding i he e a e 2-3 equal size a ms wi hin407 he po en ial clus e , and he a m’s own p oduc ion o manu e is no high enough o a408 [27] Na ional Land Su ey o Finland. A ea size by municipali y,588 h p://www.maanmi auslai os. i/ ie oa-589 maanmi auslai okses a/o ganisaa io/ ilas o ; 2017a [accessed 2 Augus 2017].590 [28] S a is ics Finland. Popula ion by Region,.591 h p://pxne 2.s a . i/PXWeb/pxweb/ i/S a Fin/S a Fin__ m__ amuu/001_ amuu_ 592 au_107.px/? xid=82ce39 0-3202-4b33-b cd-3ece427997ad; 2017 [accessed May593 2017]594 [29] Winquis E, Luos a inen S, Kässi P, Pyykkönen V, Regina K. Maa ilojen595 biokaasulai os en kanna a uus ja kas ihuonekaasujen pääs ö ähennykse [The596 easibili y and g eenhouse gas emission educ ion o a ms scale biogas plan s].597 Helsinki, Finland: Na u al Resou ces Ins i u e Finland, Luonnon a a- ja598 bio alouden u kimus; 36/2015.599 h ps://juku i.luke. i/bi s eam/handle/10024/486091/luke-600 luobio_36_2015.pd ?sequence=4 [accessed 6 June 2018]. In Finnish.601 [30] Laasasenaho K, Lensu A, Rin ala J, Lauhanen R. Landowne s’ willingness o602 p omo e bioene gy p oduc ion on was eland − u u e impac on land use o 603 cu away pea lands. Land Use Policy 2017;69.604 doi:10.1016/j.landusepol.2017.09.010.605 [31] Rasi S, Leh onen E, A o-Heinilä E, Höhn J, Ojanen H, Ha ukainen J. F om was e606 o a ic uel: P ojec s’ inal epo . Finnish case egions. Jokioinen, Finland:607 MTT Repo 50; 2012.608 [32] VAHTI; Finnish En i onmen Ins i u e. The loading in o ma ion o Finnish609 was ewa e ea men plan s, h p://www.syke. i/ i-610 FI/A oin_ ie o/Ympa is o ie oja jes elma ; 2017 [accessed 17 May 2017]611 [33] Laasasenaho K. Possibili ies o u ilize biogas in small u al communes: A case612 s udy a Soini commune. Mas e Thesis. Jy äskylä; 2012: Depa men o 613 Biological and En i onmen al Science, Uni e si y o Jy äskylä, Finland. In614 Finnish.615 [34] O’Shea R, Kilgallon I, Wall D, Mu phy JD. Quan i ica ion and loca ion o a616 enewable gas indus y based on diges ion o was es in I eland. Appl Ene gy617 2016;175:229–39. h ps://doi.o g/10.1016/j.apene gy.2016.05.011.618 [35] Na ional Land Su ey o Finland. MapSi e online map,619 h ps://asioin i.maanmi auslai os. i/ka apaikka/?lang=en; 2017b [accessed 17620 May 2017].621 [36] Na ional Fo es In en o y. To al wood olume pe hec a e,622 h p://ka a.luke. i/openda a/ alin a.h ml; 2013 [accessed 2 Feb ua y 2018].623 [37] Na ional Land Su ey o Finland. Municipal bo de s, PaITuli - Spa ial da a o 624 esea ch and eaching, h ps://a aa. da a. i/web/pai uli/la auspal elu; 2017c625 [accessed 25 Janu a y 2019]626 [38] Finnish T anspo Agency. Digi oad da a; 2018 [accessed 15 Janua y 2018 wi h627 license CC 4.0]. In Finnish.628 [39] R Co e Team. R: A language and en i onmen o s a is ical compu ing [so wa e].629 Vienna: R Founda ion o S a is ical Compu ing, Aus ia; 2017 h ps://www.R-630 p ojec .o g/ [accessed 23 Janua y 2019]631 [40] Csa di G., Nepusz T. The ig aph so wa e package o complex ne wo k esea ch.632 In e na ional Jou nal o Complex Sys ems in Science 2006;1695. h p://ig aph.o g.633 [41] Niemi J, Vä e M. edi o s. Suomen maa- ja elin a ike alous 2018 [Finnish634 ag icul u e and ood indus y in 2018]. Helsinki: Luonnon a a- ja bio alouden635 u kimus; 34/2018. h p://juku i.luke. i/bi s eam/handle/10024/542199/luke-636 luobio_34_2018.pd ?sequence=3&isAllowed=y [accessed 8 Oc obe 2018]. In637 Finnish638 [42] Picken PT. Land-use scena ios o Finnish cu -o e pea lands – based on he639 mine al subsoil cha ac e is ics Bulle in Geological Socie y o Finland640 2006;78:106–19.641 [43] BioHub p ojec . P ojec desc ip ion, h ps://www.bo nia-a lan ica.eu/abou - he-642 p ojec s/p ojec -da abase/biohub; 2018 [accessed 6 June 2018].643 [44] Uusi alo J. 2003. Me sä eknologian pe us ee [The basics o o es echnology].644 Helsinki, Finland: Me säleh i; 2003. In Finnish.645 [45] Lauhanen R., Ahokas J, Esala J, Hakonen T, Sippola H, Vii imäki J, Koskiniemi E646 e al. Me sä oimihenkilön ene gialaskuoppi [The ene gy calcula ion handbook o 647 o es o icial]. Seinäjoki; Publica ions o Seinäjoki Uni e si y o Applied Sciences648 C. Teaching ma e ials 6; 2014. In Finnish.649 [46] Kangas J, Lauhanen R, S o e R. Kunnos usoji us aih oeh ojen esis ö aiku us en650 asian un ija-a us einen a ioin i ja lii äminen pää ösanalyysiin [Summa y:651 Assessing he impac s o di ch ne wo k main enance on wa e ecosys ems on he652 basis o he expe knowledge and in eg a ing he assessmen s in o decision653 analysis]. Suo 1996;47:47–57. In Finnish.654 [47] Cal e K, Mabee W. Geo-In o ma ion Spa ial Analysis o Biomass Resou ces655 wi hin a Socio-Ecologically He e ogeneous Region : Iden i ying Oppo uni ies o 656 a Mixed Feeds ock S eam 2014:209–32. doi:10.3390/ijgi3010209.657 658 659 660 661