マルチエージェントシステム(MAS)を用いた大規模商業施 設の影響評価に関する研究

日本語概要

日本においては、大規模商業施設の郊外立地などにより中心市街地の商業機能などが衰 退しているため､多くの都市において様々な都心再生政策が実施されてきている。しかし、

都市システムに内在する不確実性と複雑性のため、大規模商業開発や都心再生政策の影響 を評価することが難しい。

本研究は、マルチエージェントシステム(MAS)を用いて、商業施設の立地に対する規 制誘導政策を評価することを試みた。具体的には、商業施設の立地と世帯の購買行動を反

映することができるシミュレーション・モデルとして、Shopsim‑MASモデルを開発した。

このモデルを用いて、商業施設の立地に対する規制誘導施策に対応して、都市圏におけ る世帯の買い物行動をシミュレーションすることにより、商店と世帯の相互作用から発生 した買物行動について予測することができ、政策の影響を示し、評価することができる。

第 8 章

StudyonDevel叩mentandApplicationofMASfbrImpactAnalysisof Large‑scaleSh叩pingCenterDevelopment

1．Introduction

ThecommercialenvironmentofmanylocalcitiesinJapanisexperiencingdeclinesolocal govemmentshavedevelopedallkindsofcitycentergenerationpoliciestoconstrainthistrendand revitalizethecentralcitycommercialenvironment.HowerverDitisdifficulttoevaluatethepotential impactofcurrentpoliciesonthefUtureofacityduetotheuncertaintyandcomplexityinheremm anurbansystem,whicharosefromcomplexcomponentsoftheurbansystemandcomplicated interactionsbetweenthedifferemelements・Therefbre,tools,whichcanprovideinsightintofUmre impactofplanningpolicieswhileembracethecomplexityanduncertaintyoftheurbansystems,are inanimperativeneed.IthasbeendemonstratedthatMulti‑agentSystem(MAS)ispowerfillin

e x p l o r i n g t h e u n c e r t a i n t y a n d c o m p l e x i t y i n h e r e n t i n a n u r b a n s y s t e m ' ) ' 2 ) . R e c e n t l y i n J a p a n , S o m e

researchesbeganusingMAStoanalyzethephenomenaoflocalcitycenterdeclineinJapan.For

e x a m p l e , Y o s u k e A n d o e t a l . 3 ) c a r r i e d o u t a r e s e a r c h o n c i t y c e n t e r v a c a n c y i n w h i c h t h e y s i m u l a t e d

theemergenceofvacantbuildingsandtheeffectofemptyspaceoncommercialspaceusingagent

basedmodel.

ThepurposeofthissmdyistoexploreMAStosimulatethepotentialimpactofcitycenter generationpolicies・Differentfrommost@game‑playing'MASmodelswhicharebasedonpure agentgbehaviordisregardingurbanplannmginstitution,thisstudyintroducesthereallanduse zoningandplanningregulationsasconstraintsfbragent'sbehavior,thisisabigstepfbrwardfbr usingtheMASsimulationfbrplanningpractice.InthiSpapel;thepoliciesspeciallyrefertothe developmentregulationsconcerningthelocationsitesandupperlimitationoflargescaleshopping centers,whichattractmuchattentionoflocalgovemmentssincetheirclosingincitycemersand shiftingtoout‑ofcenterlocationarecommonlyrecognisedasoneofthemajorreasonsfbrthe declineofcityCenters.TheMASmodelinthisstudyiscalledShopsim‑MASwhichisdeSignedto simulateinteractionsbetweenindividualsthataffectedbythedevelopmentregulationsandto investigatetheirglobaleffectsoncitycenterJcommercialenvironment.Ⅷthinthismodel,a planner,adeveloperlhouseholdsandshopsareregardedasagents.Theshopping‑marketspatial pattemsemergingfifominteractionsbetweenshopsandhouseholdsareusedfbrexaminationand exhibitingtheimpactofdifferentdevelopmentregulations.

2．Method

Givenourpurposeoutlinedintheprevioussection,itisourfirststeptoextractdevelopment regulationsusedfbrregulatingthelocationsitesandfloorspaceoflarge‑scaleshopcenters (hereaftercalledB‑shops)fifomtheUrbanPlanningLawofJapanandalocalcity'scenter

revitalizationbylaw.Basedontheseregulations,differentpolicyscenarioscanbeeasilydefinedby usersusmgShopsim‑MAS.Threepossiblepolicyscenariosareusedtoillustratetheapplicationof Shopsim‑MASfbrimpactanalysisofdevelopmentregulations・Inthesescenariosdifferent B‑shops'locationsitesandfloorspaceupperlimitationsareprescribed.

Thesepolicyscenariosarecarriedoutinavirtualcitywhicharerepresemedbyadigitalurban space・Thisspaceisconstimtedbycellswhichareheterogenousintermsoflandusezoningstams thatarerepresentedbyasetofcodesinthemodelandassociatedwithvariablesindicatingwhether thereisanexistinglarge‑sCaleshopornot・The@citizens'ofthisvirtualcityareagentsthatare thoughtrelatedtothepolicyscenarios,comprisingofaplanneragent,adeveloperagentandshop agentsandhouseholdagents.Theplannerrepresentsthelocalgovernmentwhoprovidesplanning infbrmationandinitiatespolicyscenarios;thedeveloperconstructslarge‑scaleshopsunderpolicy constraints;newbuildingsoflaIge‑scaleshopsintensifythemarketcompetenceandhouseholds makedecisionsregardingwheretogoshoppingaccordingtotheirownpreferencemeasuredby maximumexpectedutilities.Theshopping‑marketspatialpatternsgeneratedfifomlocalimeraction betweenshopsandhouseholdsineachpolicyscenarioarecomparedandanalysed,thustheimpact analysisofthedevelopmentregulationsarerealized.

3.DevelopmentregulationsfOrB‑shops

DevelopmentregulationsfbrB‑shopsaredifferentbetweenmunicipalitiesintermsofpossible locationsitesandfloorspaceupperlimitationandarewritteninvariousfbrmats・Inthissmdy regulationsaboutB‑shops'slocationsitesdescribedinUrbanPlanningLawareapplied,andlocal regulationsaboutB‑shopdevelopmentinKanazawaCityarealsoappliedfifom2002(seeTable‑l and2).IntheurbanplanningareaofthiscitylrestrictionsonB‑shopgsIocationandfIoorspace upperlimitationaredifferemaccordingtotheplannedzonningtypesdefinedinCommercial EnvironmentPlanning.Table‑Ishowsrulesoflocationcandidatesites,floorspaceupperlimitation ofB‑shopsthatarestipulatedinthiscity'sbylaw.InthispapeLtheseregulationsareinteperated intopossiblepolicyscenariosthatwillbelaunchedinthevirtualcity.

Table‑1Bylawfbrpla皿ningB‑shoP'slocationinKanazawaCity

Locationcandidate sites

Centralarea

RailwayStationarea

Cultural preservation

Zone

Subcentralarea

Requirementsoncandidatesites

C B D

Improvementareasalongmain

road

Otherimprovementareas

Areasalongthemamroad connectingtostationandother m"ortransportfacilities

Areasalongmainroad

Otherareas

Areasalongmamroad

Floor limitation

(m2)

Nolimit 20000

3000 10000

3000 1000 5000

spaceupper

Otherareas

Neighborhood Areasalongmainroad

commercialareas Otherareas

Residentialareas Areasalongmainroad

Otherareas

Industrialareas Areasalongmainroad

Otherareas

mble‑2PlanningregulationsonlocatiOnOftheB‑shop

Urban

pl^an、

area

●

lng

Urbanization

Promoting

Area

Urbanization ControlArea WhiteLand

Landusezone

lstlow̲riseexclusiveresidentialdistrict 2^、。low‑riseexclusiveresidentialdistrict 1^Stmid‑highexclusiveresidentialdistrict

2^ndmid‑highexclusiveresidentialdistrict lstresidentialdistrict

Exclusiveindustrialdistrict Commercialdistrict

Quasi‑industrialdistrict hdustrialdistrict 2^､。residentialdistrict

Quasi‑residentialdistrict

Neighborhoodcommercialdistrict

−

−

1000 3000 1000 3000 1000 3000 1000

Permitting

X

○

▲ 0

XB‑shopsarenotpennittedtolocateintheselandzoningdistrict OB‑shopscanbepermittedtolocatedmtheselandzoningdistrict

State

▲InprincipleanydevelopmentareprohibitinUrbanizationControlArea.

4｡FramworkofShopsim‑MAS

ForexaminingtheimpactofB‑shopdevelopmentregulations,fburtypesofagentsaredesignedin thismodel.Twotypesofagentsdirectlyaffectphysicalinfiastructureofthevirtualcitylincludinga planneragentwhomakesplanningdecisionssuchaslandzoning,developmentstrategies,anda developeragentwhobuildsthenewB‑shopwhosescaleandlocationmustbeconsistentwith

p l a n n i n g p o l i c i e s . O u r m o d e l e m p l o y e d t h e D e c i s i o n T a b l e 4 ) t o d e s c r i b e t h e d e c i s i o n ‑ m a k i n g

processoftheplanneragemanddeveloperagent.Besidestheplanneragentandthedeveloper agent,therearetwokindsofagentslivinginthisvirtualurbanspace,whichareshopagentsand householdagents.NewB‑shopsdevelopmentwillaffectthesupplyofgoods,marketcompetition andhouseholdshopchoice.Householdagentsdecideautonomouslywheretogoshopping accordingtotheirindividualpreference.Briefdescriptionsoftheshopagentandthehousehold agentaredoneasfbllowsbecausetheimpactofdifferentdevelopmentregulationsonthecity commercialenvironmentcanemergefifomthelocalinteractionbetweenshopagemsandhousehold agentsasdifferentspatialpatterns.

4.1Shopagent

Shopagentsarefilrtherclassifiedintotwotypesaccordingtotheirfloorspaces:B‑shopagentand

S‑shopagent・S−shopheremeansthesmallandmedium‑sizeshop.Thedeveloperlocatesanew B‑shopwhenhegetsthedevelopmentpennission廿omtheplanneragent;S‑shopsconcentratein thecommercialareaandnonewS‑shopiscreatedinthissimulation.

S‑shopsareassumedtohavehomogeneousattributes,i.e.theyhavesamegoodspricesandfIoor spaces.HouseholdsandB‑shopshoweverareheterogeneous.ExistingB‑shop3fIoorspacesand pricesaregivenexogenously.ThenewB‑shop'sfloorspacearesetaccOrdingtolocalplanning regulationswhenitisopenedbythedeveloperagent.Actuallylthecompetitionstrategiesofanew S‑shopinarealsocietywillnotopentopublic,thusweassumethatthenewB‑shopwilllauncha pricestrategytocompetewiththeexistingshopsinsimulationasequation(1),inwhich competitiveimpactsoftheS‑shopsintermsofpriceandloactionandattractivenessofother existingB‑shopsintennsoffloorspaceareconsidered.

R=K*EXPeb*d｡n)+Rnd*(Sn‑Se)/a

(1)

Where:

ThepriceofthenewB‑shopisP";parameterKisaconstant,equaltothepriceofsmallshopsin citycenter;parameterbisthepricedeclineindex,whichisgivenexogenously;variable必"isthe distanceofthenewB‑shopfiFomthecityCenterO;R"cjisanumberbetweenOandlgeneratedby computerfbllowingtheunifbrmdistribution,representinguncertainpartofpricederivedfifomthe influenceofdifferencebetweennewB‑shop'sfloorspaceS@andallexistingB‑shopgaverage floorspace此aisaexogenousconstantandhereitsvalueissetequalto500.

4．2HOuseholdagent

Thehouseholdshavedifferentestimatesaboutdistance,price,andshop'sfloorspacewhenthey decidewheretogoshopping.Throughsimulation,themarketsharesbetweenB‑shopandS‑shop willemelgeasspatialpatteminthesimulationworld・ThechangeofspatialratioofS‑shopg marketsharetoB‑shopgcouldbeusedtoassessscenariosinitiatedbytheplanneragem.

W e a d o p t a s t a n d a r d r a n d o m u t i l i t y f i f a m e w o r k 5 ) f b r h o u s e h o l d s h o p c h o i c e . I n e v e r y s t e p ,

householdscomparetheexpectedshoppingutilityofoptionalshops,andchoosewheretogo shoppingunlesstheirdemandshavebeensatisfied.Thismodelisusedtoestimatethemarket sharesofS‑shops,B‑shopsandthenewB‑shopinsimulation.

Weassumedthat:

(1)Thegoodssoldinallshopsarehomogeneous,i.e.thehouseholdgoestobuythesamegoodsat alltheshops.

(2)Eachhouseholdhasaconstantdemandfbrgoods､Whenthetotaldemandsofallagentsare satisfied,thesimulationprocesswillbeended.

(3)Eachstep,ahouseholdwantstobuyaunitofdemand.

(4)Ahouseholdonlyconsidersshopswithinacertaindistance,hisreachabledistanceγ、

(5)Theshopwiththehighestutilityissupposedtobechosen・Whenavailableshopsareunderequal

conditionsintermsofutility,thehouseholdchoosesonefromthemrandomly.

Theexpectedutilityofhouseholdiassociatingwiththealtemativeshop/isgivenby

(2)

UM=Vi+EU,

(3)

V リ = Z f j ( X ! ｡ j ) ‑ T C り ，

k

町● Ｘα ●●

世

β九

一一Ｊ一一町

Ⅸｑ

ＥＴ

(4)

(5)

仁 , ‑ x " + 6 , ‑ y , ア

Where

VariableZ/isthedeterministicpartoftheutility;Xk"isthe肘hattributeincludingpriceand floorspacedescribingstoreノpresentedtohouseholdi.;"isafilnctiontoevaluatethe attractivenessofthek‑thattributeofshopノtohouseholdI;TCI/isameasureofthedisutility oftravelbetweensiteofhouseholdiandsiteofshopﾉ;βりisaspecifictasteweightofthe householdiwithrespecttotheattributekofashop;l/isaparameterreflectingtheattimde ofthehouseholditowardthecostoftravel,hereitissetequalto‑1;uisanexogenous constantthatrepresentstheunittravelcost.Variableeダistheunobservedrandomcomponent ofutilitythatisusedtocaptureuncertaintyofshoppingbehavior.

4.3Spatialpatternsandmarketshares

Basedontherandomutilitymodel,spatialpatterns,generatedfifomhouseholdagents'shopping behaviors,arerepresentedbydifferentcombinationsOfshop'smarketshares.Ashop'smarket shareismeasuredbyafifactionofcellswhereshoppmgrateofahouseholdagentintheshopis morethan80%・Ifwithouttherandomcomponentintheutility)eventhoughhouseholdagentshave differentestimatesofXII/,thespatialpatterncanbefiguredoutinthefirststepofthesimulation withaclearboundarybetweenS‑shopandB‑shop'smarketshares・Therandomcomponentcan destroytheclearboundaryandcreatedifferentspatialpatternsasshownintheFigure‑l.However, whenS‑shopsandB‑shopsarelocatedatthesameplace,theboundarywillnotexsitandarandom spatialpatternwillemerge.AccordingtothesimulationresultsofthecasethatB‑shopandS‑shop havedifferentlocatiOn,thesmallerrandomcomponentis,theclearertheboundarylineappears.

Thus,differencesbetweenhouseholdagentsexpressedbytherandomcomponentwillcause differentspatialpatterns・ForcheckingtheboundarybetweenS‑shopandB‑shop'smarketshares, thesmallerrandomcomponentcandeliverexpectedresultsbecausetheimpactfromthedifferences ofhouseholdagentsinthispaperisnotimportam.Hence,therandomvalue500fbrtheutility modelofbothS‑shopandB‑shopareemployedfbrfUrthersimulation.

5.Policyscenariosevaluation

InordertoillustratehowMASmightbeusedtoanalyzeimpactofB‑shopdevelopregulationson

innercityregeneration,hreepolicyscenariosarefbnnulated,whichdiHeremintennsoflocation sitesandfloorspaceupperlimitationofB‑shops.Thesescenariosaresupposedtobeperfonnedin avirtualcitylwhicharethemamcomponemsoftheurbanspaceofShopsm‑MAS

＃・摩・︾弓卓ｂ日■且■■■回■せ■■■■．︒．﹄︾Ｐ曝狸・ロ︒﹄蚕．葱ロユ牢一両旧櫃空皿︑皿函函︑牢函函叩加皿■回訓．ロ？・今．︲︸．．︸鍾口●ぜ︑Ｉ衝ロ首ｎ厘画風画画興ロ：回健ロロロ薮．．γ腰・︽．．．︲■は■宙■■ｐ■古■■ロｐ■■■■■■■■賃回■■■■■．■■■ざ

・ ら 寺

﹄

■

■

■ Ｐ 虎

■

．

■

■

■

■ こ

■

■ 虞 回

■

■

■

■ 陣

■

■ 買 再

■

■

■

■

■

■

■ 画 励

■

︒Ⅱ︒■■■■ロ回■■■・ｐＯら瓦⑨斡祠再歩廊卿■毎輯■■吋■■■■陣■勅■■■野呼詣窮諾詣別私叩諾認謡詔謡諾謡窪龍躯鵲繁

ゞ型︽雫牢●唾︑四四ｍ函坤エ咋師争訓匪叩︑叩牢加函函血函率叩幽醍々胆恥醒函︑亜卵卵叩叩皿土

芯謡呼認識忠鵡繩醗翠唖蹄謡︾孵卵蹄識唖諦蹄魏唖

■■■■■■■回■■エ■■■岳■■■創■■日ロロＰ６ｂ全国両■■口■輯■■■■■酉■●■■■■■■■■■お■■■■■■餌■■︲ロロビ■■■画■卸回畢月巧賦ご韓■■■■＆■■■■■■■■■回■︑︒．︽戸０６回■ｐ■■■回■億■■ｐ■■■■■箪■卿■側■凹む■■町載■■■■１１６■回■■■■衝↑卸個■■■は■■■■■■■■■■■■■■■■■■■ｎ■■ｎ吋屯■■画回︑■陣■聞凶画■ｎ両■頁■■■■■■■■■■■■■■■同■■■■Ｐ８ｒ■■且■■■■刺■卿穆■■

■

■

■

■

■

■

■

■ ｎ 口

■

■

■

■

■ 艮 回

■

■

■

■ ロ

■

■

■ 一 回 脚 冑 胞

■

■

■

■

■ 画

︐ 両 回

■

︑

■■■■■■■■頁■白■■■■貸画■．■画■■田口侯で昼貫ロロ鍵靭■鰹馳■種■敬宅■■唾包鳳■■画■■■■■■■■画■■■両は■何と■回■宮ロ盟丘毎回■■■や

◆戸■■■■■■■■ｐ■律■■側■■■■再■■■け．Ｑ・・側■■■■■■■■回脚皿十■■■■■■■員■■■■■■■画■■■■■■■■誼訂■御■ロ■卸■■■由■回．︲・ワロ■■■宙■■■便■■■■■■︑■■■日田冑■■■■■同色■■画■ｎ吋■■やｑと■■■旬■■■■■■■■■■■■■■暁■■斡■■岡■■伺回■︑■■■即宙︒．■■■■■■■■■■■■■■■■■■■■■■■■刺■■■■皿再■■側凹■

暉 邑 部

＆

■

■

■

■

■ 催

■

■

■

■

■

■

■ 竃 願 麓 ロ

■ 画

■

■ 便 回 埠

■ 阿 回 回

■

■ ロ ロ

■

■ ｎ

．

□ｆ回■■■■再■■■■■■劃■■■■■■■■■■ｂ制同■■ロ■■凹国昼︽﹂■■■色■■■■■■■■■■■Ｕ■は目巳佳■■■目釘ロ■■周■■４才句︲︒■■■■■■唇■■■■■笹田■風■■■町■同■釦■■首■︲．§︲︲今■■■■■■■■■■■■■■■側■■■■■■回■固■■側︐

Ｌ■脾凹員■■■■■厳■■■働伺■■鱒Ｅ■屍■降向■■回■■■日■■■算■■■■■回■■■■回回■■■■■■■回■■ｑ■■■■

■ 戸 ■

■ ■ 唾 ■ ■

■■■■

一・■

Ｐ４回

■

Ｆ■■■回何画■■■■■■ロロ■算■■■■■回■■■■回回■■■■■■回■■■ロ画■■■■円■■■■■■■

■

■ 旦 垂 ■ ■

B‑shopandS‑shopmdifferempositions

ム リ t l U ﾄ ' 1 1 0

＋HH十十什H十・I出Et監拙_ロ

蓑蕊鑿鍵蕊

I■ﾛ■己国

■

隙

F 軍

■妙

！

^{千 ■}

鍵

迩 醗 謹 蝿 畠 群 §^■■■

鍵蕊

^翌

＝ 一 旬

識 鵜

一 一

■■■

■■■

■■周■■国

鍵難難

￨ ￨ ﾛ Ⅱ Ⅱ ■ 且 ■ B 1 ■ ￨ ￨ ￨

^{謹 耀 騨 蕊 雛 蕊騨}

両 J W i 宮 、 ■ ■ ■ ■ 画 ■ ■ ■ ､ 拝

購熊輔 十 雑

＋I P F ＝ ■ ■ 冑 冑 戸 ■ ■ ｰ E ■

ユ

■S伽ppmgratemS‑s加ps>80%mShoppmgratemtheB‑shop>80%

Rs=5000,Rb=10000Rs=000,Rb=2000Rs=500,Rb=500 Figure‑1Thespatialpaiternsofshoppingbehaviours

X

必▲典

．:凸,岸,｡ 哩哩､遇凹凸:限.L『型Lu註｣営四唖劉1回＝曇f,』涯郵iL伽0．』‑F,．、』‑掴−1．−.ざ▲己｡.｡．..X ,弧l蝿哨蝋 制副猶融p駐刷灼El淵型師崗謝剛.．;.&'g L, 13.,：1』かl司・,､， ｡． ,．,;!▲・・:。，。 ．‐ ･X X X X x 話 罰 龍 ﾕ 翻 腫 呼 武 ; 唖 氏 函 卜 雅 1 画 圃 醗 画 圏 聞 閲 倒 圃 駒 扇 甜 ； ･ ' 、 . ： ▲ . ． X

蕊

iHl,甥f,;唾X劉醐 jK‑F呂、浬也;×帰 一野趣g 闘刺x 1jDz:.H澤擢鯉司罰 .i:1.1弘製踊剖胡 蹄圏鍬;酎幽鯉醐醐 :1』】切剛睦喝側慨

；IIL稚垣』鯉8.j卿 .､Ifl23Kq蝉14翻唖 .；ん ;.2,.蝿:踊麺動

．＆1四園哩恥副淀

#鰯:議唱

､も･色』X：、皿

1

Main

, . ； ･ q p f p 1 ■ ･ ■ ｡ ■

･ 句 ▲ ■ J

， 甑 圃 燭 醐 閏 況 凹 圃 魁 幽 園 圃 画 四 唖 瞳 掘 騒 蜘 圃 届 國 圏 蝋 圏 割 閥 圏 劇 砲 聴 圏 賎

&。．1．−．．．1回醒圃駅蝿騒×湿画圏』‑6j，､国凹飼晦呵8m砺閥旧刷駕劇．!．．・：，奇．．､図掴蝿馴廟忌 :,，．.,.．‑号 ,:1．鴎醐創醜欄園×圃圃間脳 ﾛ :｡:@部一・闘磁圏削圏:..；．【1，『、､、 44．．，．．．J，閣図剛噛謝榴

． 轍 聞 面 倒 罰 × 鞭 幽 蜘 幽 四 ・ ・ ・ 糎 閏 協 糊 甑 伽 聴

． . 、 腿 I ｮ 咽 附 × 隅 1 吋 騨 騎 珊 翻 ． ， 、 個 哩 型 圃 閣 ﾛ ﾊ 画 悶 絶

．.−．．．．．．．．．，，．．歩:.隅魍穆到×幽幽蹴踊画題蝿1−−．，治I:;凶.,I‐.蝿囲闘脳脳国鯛醜思画圏 4．【9 ;・・心:ゞ$堤曽 砿i顧印愛瓢X鰯剛1剴幽窺睡圏翻'.:､.．L'1?..9.,J.‐}.．.！：1圃醐醐鍾窮皿溺副撹麓閣題 劉 鰯 罰 蝿 瑚 簡 翻 薗 × 班 剖 嘘 測 隠 耐 個 闘 圃 。 四 四 U Ⅲ 醐 獅 圏 躍 掴 閉 園 翻 圏 I E 罰 騨 固 魍 園 画 面 圃 耐 m x 画 胃 刷 璽 阿 輻 幽 圃 謝 画 . 閏 罰 園 翻 禰 嗣 I 刊 覗 圃 圃 嗣 刺 副 掴 鞄 跨 雷

． ， 、 白 幽 奴 鴎 蜘 捌 馴 圃 幽 烈 個 圏 × 副 画 魎 堀 閉 幽 掴 閃 照 腰 哩 、 4 劉 岨 鰹 I . 』 幽 回 劉 捌 凶 伺 幽 盤 殴 囲 編 蝿 麓 盟

溝 xXxxxF..'!！

￥.皐1.i1ﾕX砂 , :剖単AI咽×

灘 x X X 箕

▲ ×

×

;篭r§辱蒸祇

鯏 届 ． 擁 KXX穂 曲 盗 ▲ ぬ 進 些 ＆

6 6

X浜 灘斡 篤 漢韓溝渕

閥謝鵬別刷腐訓漢浪

gomma･cialdistrict

N e i g 1 b o r i n g c c m m c r c i a l d i s t r i c t

2ndreside,Itialdistrict

Q u a s i ‑ i n 珈 愚 t r i a l d i s t r i c t

Industrialdistrict

罫

Q u a s i ‑ r e s i d e m i a l d i s t r i c t

Otherlandnsedistricts

U C A 圏 U 翌 A

Railway MainRoad

B‑sh"

S‑目加p

Railway誠ation

Ccmma･cialcent"cfthecentralcity

１２２４３６７

□▲×鯵▲⑬四 鬮圃囿魎圃圃□

(b)LandusezonmgmUPA Figure‑2Thehypotheticalurbanspacespatialstructure (a)Urbanspace

5.1ThehypotheticalurbanSpace

Thissmdyconcemsahypotheticalurbanspaceof2500cens(50X50)whereeachcellmearsures 500mX500m.TheurbanspacecomprisesacentralcityandtwoneighboringcitiesThemodel assumesthatlhecentralvirtualcilyhasthetypicalcharacteristicsofKanazawacityinJapan, namelywithatraditionalconnnercialcemerlocatedmtheheartofthecity;withurbanplannmg area(1230cells)dividedmtoUrbamzationPromotingArea(UPPA)andUrbanizationControlArea (UCA)andwithdemedlandusezoneswithinUPAAllthesesplannmg・mginfbrmationisassignedto eachcellofcentralvirmalcitybytheplanneragem・Theurbanspacespatialstructureisshownm

F i g u r e ‑ 2 ( a ) a n d ( b ) . T h e l s t ‑ 6 t h t y p e s o f l a n d u s e d i s t r i c t s i n F i g u r e ‑ 2 ( b ) a r e z o n e s w h e r e B ‑ s h o p s

canbepelmiltedtolocate.

5.2Processofsimulationandpolicyscenarios

TheshoppmgbehaviOrsofhouseholdagentsasdescribedinsubsection4.2aresimulatedfbr producmgmarketsharesofshopagentsmurbanspacemordertovisualizetheimpactofthenew B‑shop.Thesimulationprocesscanbeimplememedasfbllows:

●TheuserofShopsim‑MASdefinesapolicyscenariotobeimplemented.

●Theplanneragentsetsthespatialstructureandinitiatesthisscenario.

●S‑shopagentsandexistingB‑shopagentsarecreatedintheurbanspace.Householdagentsare createdanddistributedtothewholecentralcityurbanplanningarea.

●ThedeveloperagentplacesthenewB‑shopinurbanspaceaccordingtodefinedscenarios.

●Theusersetstheinitialvaluesofparameterincludingunittravelcost,reachabledistanceof thenewB‑shop.

●Householdsthendecidewheretogoshoppingasdescribedinsubsection4.2untiltheir

demandsarefulfilled.

BasedonthedeVelopmentregulationsdescribedinsection3,threescenriosarefbmulatedfbr locatinganewB‑shopinthevirtualcity．l)CenterActivation(CA):Tbreversethedecline, encouragelarge‑scaleshoptolocateinthecentercommercialareawithoutupperlimitationfbr floorspace,butstrictlyrestrictout‑centerlocation.2)RailwayStationDevelopment(RSD):In ordertodeveloptherailwaystationareaintoacomprehensivebusinessarea,B‑shopcanbeopened

n e a r t h e s t a t i o n , w i t h a n u p p e r l i m i t a t i o n o f l O O O O m 2 . 3 ) N e i g h b o u r i n g C O m m e r c e P r o m o t i o n

(NCP):Tbimprovementcommunityconvenience,encourageB‑shoptolocateinneighbouring

c o m m e r c i a l a r e a , w i t h a n u p p e r l i m i t a t i o n o f 3 0 0 0 m 2 .

5.3Modeltest

Asdescribedinthehouseholdshop‑choicemodel,themainexogenousparametersaffecting deteministicutilitiesofhouseholdagentsareunittravelcost(c),reachabledistance(Y)andfIoor space(S).Inordertoexaminetheoperationalcharacteristicoftheshopsim‑MAS,asensitivity analysisfbrparametervalidationisconducted・Hereeachparameterisexaminedrespectivelyandit isarguedthatifoneparameterisprovedvalidinaffectingshoppingbehaviol;itisselfevidentthat allparameterscanworktogethertoimposesuchaffection・Thesimulationiscan･iedomunderthe settingasfbllows.Householdsinonecellaresupposedtobehomogeneousandareregardedasone agent;hencetherearel230householdagentsinthesimulation・Eachhouseholdagenthas50 demandsthatmdicatesthenumberofshoppingtimesinonemonth.TherearetwoexistingB‑shops andseventeenexistingS‑shopinthecitycenter.ThefIoorspaceofthefirstB‑ShopissetaslOOOO

m 2 a n d t h e s e c o n d B ‑ s h o p ' s i s s e t a s 2 0 0 0 0 m 2 r e f e r r i n g t o t h e f I o o r s p a c e s t i p u l a t e d i n p l a n n i n g

regulations.InNationalSurveyofPrice(www.stat.gojp),theHoorspaceofasmallscaleshopis

u n d e r 4 5 0 m 2 . H e r e , t h e s m a l l s h o p s i n t h e c e n t e r o f K a n a z a w a c i t y a r e c o n s i d e r e d t o b e s m a l l e r a n d s e t a s u n d e r 3 0 0 m 2 . T h e p a r a m e t e r s h a v e g o o d e x p r e s s i o n i n t h e s i m u l a t i o n a s s h o w n i n

Figure‑3(a‑c).Consequendly,saleamountsofshopsaresignificamlyaffectedbytl･avelcost, reachabledistanceandfloorspacethatinfluentsshoppingutilitiesofhouseholds.

ForcalibrationofShopsim‑MAS,theunittravelcostissetas20en,averagebusfarefbronecell space500m;corl･espondentlythereachabledistanceissetas30fbrl5000m・CommercialStatistics Serveyinl985,DigitalNationallnfbrmation(http://nlfip.mlit.gojp)isemployedfbrthistesing.

Figure‑4(a)showstheshops'spatialdistributionofKanazawacityinl985.Themesheswithmore thanl50shopsareidentifiedasthecitycentel;whichaccommodates2006S‑shopsand4B‑shops.

T h e t o t a l f l o o r s p a c e o f S ‑ s h o p s i s l 9 2 4 4 5 m 2 a n d t h a t o f B ‑ s h o p s i s 2 6 4 8 3 m 2 . F o r m o d e l t e s t , t h e c e n t e r a r e a i s m a p p e d i n t o 3 6 c e l l s w i t h 3 6 S ‑ s h o p s ( t h e f I o o r s p a c e o f e a c h S ‑ s h o p i s 3 0 0 m 2 ) a n d 1 B ‑ s h o p ( 1 5 0 0 m 2 ) , w h e r e t h e r a t i o o f t o t a l S ‑ s h o p 3 f l o o r s p a c e ( 1 0 8 0 0 m 2 ) t o B ‑ s h o p J i s 7 . 2 ,

almostsamewiththerealfloorspaceratioofS‑shopstoB‑shops.ThesettingofHouseholdagents issamewiththatinparameterexamination・Figure‑4(b)showsthevirtualshops'positionsbased ontherealcityandthesimulationresult.ItcanbeseenthattheperfbnnanceofS‑shopssurpassed muchthatofB‑shopsbecauseoftheirobviousadvantagesinnumberandfloOrspace.Table‑3 showsthecomparisonbetweentherealdataandthesimulationresult,whichareinconsistentwith eachother・ConsequentlylwecanconcludethatShopsim‑MAScanbeemployedasapromising tooltosimulatespatialpatternsofmarketshares.

０００００ ００００００００

００００８６４２

﹈匡．○Ｅくの市切

205080100120150180200250300 UnitTravelCost

Y=30,S1=10000,S2=20000 (a)Unittravelcostandshop‑saleamount

０００００ ００００００００

００００８６４２

﹈匡．○Ｅくの石︑

0 5 1 0 1 5 2 0 2 5 3 0 ReachableDistance

c=20,S1=10000m2,S2=20000m2

(b)Reachabledistanceandshop‑saleamount

80000 60000 40000 20000

﹄匡二ｏＥく①万⑩

0

3 5 8 1 0 1 2 1 5 1 8 2 0 Z 5

FloorSpaceoflstB‑shop(unit11km2)

c = 2 0 , Y = 3 0 , 8 2 = 2 0 0 0 0 m 2

(c)Floorspaceandshop‑saleamount

‑ S ‑ s h c p 一 一 B ‑ S h o p

Figure‑3Parameterssensitivityexammation

蝋蝋臘蛎捌蝋轆

９１６９口■■・Ｆｑｑｑ

■９６６０

■０ⅡⅡ０Ｆ■■

■

■

■

■

い狸詠い範肺獣砺里齢・認泌韓郡諄鮒紳帖神林排肺躯諦艶

t

N ：

A

^{軌 ■ ■ ● ■G a G 白 白}｡P●●●

■ ■ 心 O G B G 匂 い ● △ め わ B C F Q ● B 弓 ■ 0 屯 ● ■ 伊 ■ ■ ● 凸 ■ p b U ●

●■

！

9● ^{蕊 ■■●０画■Ｉ●ｂ①６口︒００１} I ムクロロ■○凸●■ＤＣ︒ｑｏＱ４

I 、｡

／

NIImb"Ofgllol)9 ^6■

■ ｡

0‑100 101 150

＞150

｛

ロ唾︒

、 い 争 ■ や ● の

024Kilollleters

l l − l ̲ ＝ L

( a ) T h e d i l o n O f r e a l S h o P S : : M : : ; 職 謡 劉 瞬 脇

(1985)

(b)Simlllationresultsinthevirtualcity Figure‑4Realshops'distributionandvirtualsimulatiOn

ZZz6ﾉe‑3Co77qp""〃QfIrq/e"zoz"加7℃α/cz7zdv"r"αIce"たγ Floorspace(nf）

MeshsizeShop(number) ^Saleamount Marketshare

Reahty

(1IQn2)

S‑shop(2006)¹⁹²⁴⁴⁵ 186(bmionen) ^89％

B‑shop(4) ²⁶⁴⁸³ 22.4(binionen) ^11%

隙 臘 , ）

^{S‑shop(36) 10800} 55475(demand) ^90％

B‑shop(1) ¹⁵⁰⁰ 6025(demand) ^lO%

４

・ 咽

・

・ 戸

﹄

▲

︾や・劃謁溌等創〆色９●●◆ａ■

令■■勾配学叱

ヰロヴ●冬凸■

紅訓釧郡諦謹皿

叩鋤鯨岬吋﹄即咄解坤酔距伽

呵ｑ４４０句■梱■ＣＧＯ■−０︐●９００口Ｑ０ｇＤｐ自可■旬坤甸酋■◆句印ｑ■■ｆ▽◆４◆→４もｅｆＧ４４

や宇白︑寺︒００６◆◆◆＆鳴協縄輯叩舘胤

時■ｐ﹄■わわ岳口の②●

識

問●】g唾9 1 L

謡詑謹︒昂認斡部︒や．Ｊ１尾翼詞拠詩帥盟開ｕ計測陛遍．﹈△橇顧鱗雛鐸撫瀞

●■ｂ阜凸卜一日■●﹄■■ｎＰ的■曲ロｅ●ｑ●①●■難輔撒謝灘

ｑＱｏＪｐｑ０９０凸６曲４●ｑ４Ｑ■毎︒９９●０．８Ｇマゥ９０ＤｂＤｇ００Ｏｄ０ｄ００■日

輔句旬■▽○１１句．旬■■旬■側Ｑき■４︒■■ｑ●■純鈴謝詫４０面塊測帥脇艸祁叩︑

ｐ●︒︒●◆◆︲４．１▽守り白け申凸宅︒◆︒■●毎．

●０●●︒︑︑上ｄｂｆ９９■ｂＤｂｐ０り００６０９．輔い︒●ずｂｐｂＰＬ■やｐ■■■●■ｎ口■■Ｈ毎・け列記謝躯晶側冊帥別開制闘︲・

蕊鐵蕊織鍵

け４４４９中ＧＨＬ８や４０６９９４０１◆！・錘錘躍鰯艤聯滞︑叔銅蝿弐

1 1缶鼎聯瞬秘

呼翠牽認︒恥や覗川

９▲０９４血ｑ０４Ｆ４１心Ｚｒ小︒

＄

︐０才０申〃◆Ｏ９４ｃ０中０９ＣＤ＆０９◆りり・甲凸寺ｏ◆︒■●毎ｄｂｂｂｂＬｒ０色？ＤＯＱ０Ｂ

旬●㈲ｎ口■■Ｈ毎・け９０００００４４勺０６周

りｏｇ０ＰいいＯＧゆ

︑ 恥 鴨 剥 卵 餌 函 那 凸 鉱 枇 認 誌 対

↑

■ｑ０ｂｏｂｊ■Ⅱ︒■４●︒■仏画禺口号４０１◆いも︒

Ｆむち︑ｆ則︐１ｑ︑・Ｌ

S h o p s p o s i t i o n F l o o r s p a c e = 3 0 0 0 m 2

^{5000m2 10000m2 15000m2}

EShoppingratemS‑shops>80%EShopp加gIatemexistingB‑shops>80%DShoppmgrateinthenewB‑shops>80%

Figure‑5ThespatialpatternofshoppingrateofCAscenario

5.4Comparisonofdifferentscenarios

AcmallyjtherearesomeB‑shopsandS‑shopsmthecemerclosedbecauseofoperatmgdeficit andmanyB‑shopsloQatingoutsidethecitycemerofKanazawaafterl985HoweVer,itisdifficult totakemtoaccountallrealshopsinthesimulationbecauseoflimitedsystemcapacityLHence scenariosanalysisiscon血ctedundersimplifiedhypotheticalconditions.

Threescenariosareshnulatedwiththesameparametersettmgsasmtheparametersensitivity examination.Theexistmgshop3positionsareonlysetasFigure‑2shnilartothelocationof B‑shopsintheeasternandsomhernKanazawaCityThespatialeHbctsofCAscenarioisshownm Figure‑5and6・SpatialpatternsmFigure‑5arerepresentedbymarketsharesintermsofshopping rate.Itcanbeseenthat,marketsharesofcemershops(mcludmgS‑shopsandthenewB‑shop)and existmgB‑shopsareseperatedbyaboundarybbmmarketsharesconcernmgS‑shopsandthenew B‑shopappeararandompatternsmcebothofthemhavethesamelocation.S‑shopsarefacedwilh fiercecompetitionafferestablishmemofanewB‑shopinthecitycemerandtheylostmostoftheir marketshare.FromFigure‑6,itcanbelearnedlhatthelargerthenewB‑shopisthemoremalket shareoflilecemershops,butS‑shops'marketsharedecreases.ThisindicatesthatCAscenariodo haveeffectinimprovmgthemarketperfbrmanceofcentershopsasawhOle,butmaycausesevere harmtothecemerS‑shopsatthesametmeifthereisnolimitationonB‑shop'sscale.

Tbcompareinfeescenarios,1hedeveloperagemismadetodevelopanewB‑shopofsamefloor

s P a c e , 3 0 0 0 m 2 , i n a l l t i l e s e s c e n a r i o s ・ T h e c o m p e t i t i o n f a c e d b y S ‑ s h o p s m i l l e s c e n a r i o R S D a n d

NCPisnotasstrongasthoseinscenarioCA,bmmarketstatisticsrenectthatbothofRSDand NCPprovidenoanyhelpmpromotingthecemercommercedevelopment"gure‑7).Inlatertwo scenariosthelossofmarketshareofS‑shopscausedbythenewB‑shopismorelhanmCA scenario,asshowinFigure‑8.ThisfilrtherindicatesthattheCAscenariomightbeaneffective measuretoimprovetheactivityofcemercolmnerceifthereisareasonablelimitationonB‑shop's floorspace.ComparmgwithcommercialzonesmmasterplanofKanazawacityBCAandNCP scenarioareacceptedatthesametimeintheplannmgconcepts,whichmightfindasolutiontothe conflictbetweentheamuenceofthecommericalcemerandtheexpansionofthecity.Ifthemodel capacitycanbeextendedtoftfbrthecomplexofrealityiShopsm‑MAScanbeemployedasa

promismgtooltoanalyzethedynamicprocessregardmgB‑shopplanning

40000 35000 30000 25000 20000 15000 10000 5000 0

﹄匡．○Ｅくの雨の

3 5 8 1 0 1 5 1 8 2 0

FIoorSpaceofNewB‑Shop(unit:1km2)

老一S‑shop−■−ExistingB‑shop‑←NewB‑shop−←Center

Figure‑6ThesalestatisticsmCAscenario

300UC Z5000 20D0O 150DD ユD00p 5000 0

営三ｐＥ証里ｍの

調 _鰯 ^霧議鯛 _鯛｜￨鰯

BaSe C A R S D N C P

回S‑shops ■Centershop三

W h e n n e w B ‑ s h o p ' s f I o o r s p a c e = 3 0 0 0 m 2

Figure‑7Thesalestatisticsofshopsmthecitycenter

r 画

■■

￨確率壷

4 1

。｡：

○I aq口●I

●●1

●GI

。●●1 巴のd ﾛ ● ● 申 I Oe●1 .■●0や1

｡●●●I

●●Geq

■●0．1 p●○○1

心0●｡

●Ood C■｡■

ロ 旬 e q qd

1，､｡『,.；

■ ● コ マ 甲 ﾛ 臥

短溌卜・‐

■ 0 ● □ ◆ ● ● 』 もDeO●●｡①●

●■●申申●①｡■■

●巴の●●●①｡●■ 照

rF' J ･ ･ q

ク･･･::

鯉

漁』

閑

1，・

●

1，

．4

６００６Ｌ

Ｐ■Ｕ４ｄ

Ｄ０■

00

鯉 マ畢●●◆Ｏ︲？︒●︒●■ｂ写り◆︒●ＯｂＤ●も●●●ｐ０Ｏ◆◆●◆●◆今●６●■■０︒◆︒●●００．●心口ｅＯやＣｅのＤＤｃｅ●﹄︑︒Ｏ●●︒◆Ｐ●●◆●ＧＢ︽■①●中の︒釦伽伽■叩幸い釦酌却■吟

ｑｇＤ０尋●■︒●︒■り︾●Ｇ◆●●申︒◆令９０マ■●■ｑＧ９ｇＢ■ＯＤ

Ｐ︒●申伊◆︒●台申︒●ご●●申９５６●ｄｅｅ守り●凸○●︒︒●◆９

守り心申？ｏ４●●︒■一

●ｐ司恥舶伽印加●

■０ｑ申●Ｑ○●○○︸■■争争●︒●０●

■ｅｂ■Ｇｄ︒■︑︒●◆●●●Ｇ９．Ｕｄ０■０９ 可６Ｆ４０Ｇ匹４０●▲

︲４５

口■ｇ■

ｑＦＩ

Ｆ

●

？

■

︾◆︒●ＯＯｐ守室々︾●ｅ００ｇｐＰ◆◆●︒４０句■︒■■ｑｐｂｐＱ９●●︒

︾◆ｇｏ︒○中●■一︒︒●■○句●①●一○●●■争企︒︒●■▲一旬●■ゆ●●いら●■トー申●■■︒●●◆︒

胆

蕊

CAscenario RSDscenario NCPscenario

■S加ppmgratemS‑shops>80%DS加ppmgratemB‑shops>80%

Fe‑8Spatialpattel･nsofshopp加grateindifferentScenarioS

( n e w B ‑ s h o p f l o o r s p a c e = 3 0 0 0 m 2 )

Figure‑8

6．Discussionandfurtherresearch

Inthispapel;theuseofMASfbrmpactanalysisoflargescaleshoppmgcemerdevelopmem regulationsisproposedandillustrated:First,byintroducingrealurbanlandusezoningtofbrm agem'sbehaviourconstraints,theShopshn‑MASsimulatethevirmalurbanspacema'more

practicalwayinthecontextofurbanplanning.Second,theoperationalcharacteristicofthe Shopsim‑MASisexaminedthroughparametersensitivityanalysisandtestwithcommercialsurvey data(1985)ofKanazawa.Duringthisprocess,Shopsim‑MASexhibitshowmarketspatialpattems emergefifomthesalingandshoppingprocessindirectlyaffectedbydevelopmentpolicies,andhow thesespatialpatternscanbeusedtoanalysetheimpactofdevelopmentregulations・Theimpactof thenewB‑shoponthecommercialenvironmentofthecitycenterarevisualizedaccordingtothe threescenarios,whichindicatethattodevelopnewB‑shopinthecitycentermightbeaneffective measuretoimprovecommercialactivityofcityCenteraslongasthereisareasonablelimitationon B‑shop'sfloorspace.

Asshowinthispaper,Shopsim‑MASprovedtohaveagoodperfbrmanceinanalyzingthe impactoflargescaledevelopmentregulationsfifomaspectsoflocationsitesandfloorspace. Howevel;itdoesnotnecessarilymeanShopsim‑MAScanbereadilyusedinpracticethoughwe believeitwouldbethecase.Therearestillmanychallengesremainingfbrfilrtherresearch・For example,theimpactsofthedynamiccompetitionbetweenS‑shopsandB‑shopsonspatialmarket patternsarelefiasanunsoIvedproblem.Themostimportantchallengeishowtodealwithlarge amountdatathatrequiredbymicro‑simulationandcalibrationinMAS・Itisclearthatareasonably

c o m p l e t e u r b a n s i m u l a t i o n m o d e l w i l l n e e d b n o r m o u s a m o u n t s o f d e t a i l e d d a t a , n o t o n l y i n c l u d i n g

landuse,householdsandtheircharacteristics,butalsoenvironmentalandsocial‑economicfeamres. DatafbrplanningaregenerallyavailableinGISfbrm,andmaybereadilyintegratedintothe databasefbrmodeldevelopment・InfilturewewillintegratetheMASmodelwithGIS,andthen moresatisfyingoutcomescanbeexpected.

References

l)MasanobuKii,KerjiDoi(2005),｢Multi‑agentland‑useandtransportmodelfbrthepolicy evaluationofacompactcity｣,EnvironmentandPlanningB:PlanningandDesign,Vol.32,

485‑504.

2)MichelleChabrol,DavidSarramia,NikolayT℃hernev(2006), methodologyjlnternationalJoumalofProductionEconomics」

｢Urbantrafficsystemsmodelling

,Volume99,Issuesl‑2,156‑176.

3)YosukeAndo,KeViYokotaandTbhruYoshikawa(2005), 「Urbansimulationbyamultiagent systemconsideringemergenceandagglomerationofvacantbuildingsinurbandeclineprocess andeffectofemptyspaceoncommerce｣.JournaloftheCityPlanninglnstituteofJapan,No. 40‑1,52‑59(inJapanese).

4)T.Arentze,A.Borgers,H.Timmermans(2000), 「Aknowledge‑basedsystemfbrdeveloping

retaillocationstrategies｣,Computers,EnvironmentandUrbanSystems24,489‑508.

5)Baltas,GeorgeandPeterDoyle(2001), 「RandomUtility Surveyう」JournalofBusinessResearch,Vol.51,115‑125.

ModelsinMarketing・Research:A

ドキュメント内 塞蕊毒 (ページ 91-106)