规划师主体模型：一项低碳城市形态规划支持的工具

1、1 城市模型与规划支持系城市模型与规划支持系统统1.1 规划支持系统在城市规划中的应用探索1.2 多尺度的北京城市空间发展模型1.3 规划师主体模规划师主体模型型：一一项项低低碳碳城城市市形形态态规规划划支支持持的的工具工具1.4 囊括方法、软件和模型的规划支持系统框架体系1.5 面向空间规划的微观模拟2 大模型与定量城市研大模型与定量城市研究究2.1 大模型及中国应用案例2.2 基于OpenStreetMap和兴趣点数据的地块特征自动识别2.3 地块尺度中国所有城市的空间扩张模拟2.4 中国PM2.5的人口暴露评估2.5 利用北京公共交通刷卡数据的若干定量城市研究2.6 当前定量城市研究的四

2、项变革3 规划设计响规划设计响应应3.1 数据增强设计：新数据环境下的规划设计回应与改变3.2 街道城市主义3.3 城市规划实施评价：针对中国城市的分析框架3.4 基于人类活动和移动数据的城市增长边界实施评价3.5 中国收缩城市及其研究框架城城市市模型及其模型及其规规划设划设计计响应响应3.6 历史上的北京规划Applied Urban Models and Their Applications in Urban Planning & Design3.6 Historical city plans in Beijing1 Urban Models and Planning Suppor

3、t Systems1. Planning support systems in urban planning2. Beijing urban spatial development model families1.3 Planner Agents: A toolkit for support planning a low carbon urban form4. An applied planning support toolkit including quantitative methods, software and models in China5. Urban micro-simulat

4、ion for spatial planning2 Big Models and Quantitative Urban Studies1.Big models: Several fine-scale urban studies for the whole China2.Automated identification and characterization of parcels (AICP) with OpenStreetMap and points of interest3. Simulating urban expansion at the parcel level for all Ch

5、inese cities4. Estimating population exposure to PM2.5 in China5. Bus landscapes: Analyzing commuting pattern using bus/metro smartcard data in Beijing6. Four changes on quantitative urban studies in the big data era3 Applications in Urban Planning & Design1.Data augmented design (DAD): Planning

6、 & design in new data environment2. Street urbanism3.Evaluation of urban planning implementation: An analytical framework for Chinese cities and case study of Beijing4. Evaluating the effectiveness of urban growth boundaries with human mobility data5. Shrinking cities in China and the research a

7、genda1 INTRODUCTIONLand use pattern (urban form), or land use layout, is a key part of physical plan (master or detailed) Spatial distribution of land use and density Hard to predict by a planning support system (PSS) Proxy of carbon emission and energy consumption Land use pattern scenario analysis

8、(LUPSA) most are parcel-basedCUF (Landis 1994)What if? (Klosterman 1999)INDEX (Allen 2001)iCity (Stevens et al. 2007)Other papers regarding land use layout optimizationPlanners in LUPSA tools Less attention was paid on the behavior of urban planners Our research question: How do planners compile lan

9、d use pattern? What are rules (preferences)? How to identify these rules? Are these rules varying among planners? Could we develop a PSS for “simulating” land use patterns using the identified rules? Is the pattern associate with low carbon emission?Building city in a vitroHatner and Benenson, 2007,

10、 EPBThe entropy of LEGOCrompton, 2012, EPBIn this research, we will identifyplanner rules by Questionnaire What one planner will do Mining plan drawings What one planner has doneThen we willdevelop a PSS (Planner Agents),and simulate land use pattern using identified rules.Planner APlanner CPlanner

11、BIdeally and hopefully to save planners time and promote plan compilation efficiency Support planning a low carbon urban form E.g. A plan area Identified rules of 20 planners Generate 20 patterns in one minute by using PlannerAgent The principal investigator chooses a perfect one All 20 planners foc

12、us on it and propose the final drawing Planner Cs work associates with low carbon2 PLANNER AGENTSPlanner types Non-spatial planners Infrastructure, transportation Not directly with land use pattern Spatial planners Responsible for preparing land use pattern Chief planner Confirm the final plan schem

13、eSpatial planner: the general process1.Totals in area For each type of land use (e.g. residential, commercial and industrial) From decision makers or forecasted by macro models2.Constraints Geographical context: slope, eco space Institutional constraints: development restrictions3.Negotiating with n

14、on-spatial planners (factors)Assume planned facilities, roads, city centers, CBD, etc., are ready prior to plan a land use patternWeight factors4.Negotiating with citizens (public participation process)Not accounted in our current researchSpatial planner: simplified rules The taste (weight) of each

15、land use on factors is different. The weight could be calibrated using questionnaire or data mining on existing plan archives (land use with the highest probability would be selected for a parcel). E.g., industrial parcels tend to be located along main transportation network, commercial parcels arou

16、nd amenities.b巳Questionnaire surveyQ Virt叫real 阳曰Real modelPlanning rules (requirements and preferences): Public facility NeighborNatural reserveParcel size, form BlockSubway siteLocal conditionsFormulating comprehen sive constraintsPlanning law Planning guideline Physical geographic statusLand use

17、pa“ern 2 by SPA2Land usepa“ern 1M芷 e: t he rmt hxt for m.I at i咱 l ar略usepct t em 283 i s t he sar. as t f回 f 町 I and use pct t em 1.Land use pattern 1constraintsPhysical geographic statusN ot e: t he m et hod f or m ul at i ng l and- usepat t er n 2& 3 i s t he sam e as t hat f or l and- use pa

18、t t er n 1.Evaluating land use patternsThe ideal land use patternSPA2 SPA1 SPA3between SPA 1 and EPAsRAN ot e: t he st eps, f or w hi ch t he backgr ound i s gr ay ar e consi der ed i n t hi s ar t i cl e.RA evaluationLand use pattern evaluation Spatial autocorrelation (MoransI and LISA) Landscape m

19、etrics using FRAGSTATS FEE-MAS model (Long 2011) Calculating potential transport energy consumptionLand use pattern 2 by SPA2Land use pattern 3 by SPA3CPACoordinating land usepatterns3 BEIJING APPLICATIONThe central city of Beijing1085 km218 big zones Beiing Detailed Plan (-2020) Land use plan in ea

20、ch zone has been exclusively designed by a responsible planner, in 2007 A perfect data for applying Planner Agents21 factorsZone 12 as an exampleLand use typeParcel distributionNumberArea (km2）PercentageR11443.850.41C9744.410.41M40.470.004O12118.940.18Total336107.671.00ConstraintsExtracted from Urba

21、n Containment Plan of Beijing See Long et al 2011 for detailsIdentified rules using multinomial regressionParameterWeightRCMIntercept C21.70203*.59824*2.24992*.108661.78990*1.50529*C221.69092*1.98993*1.48453*C25.27165*.63531*1.50131*C3.54465*.53033*.09401C4.19670*.20072*.34227C51.01238*.71570*.37010

22、C6.59667*.83476*.57046*CBD3.13736*.73107*7.74911*Exit.77072*.81033*.21059G.06680.14353*.52322*Gov.22590*.11004.78724*Hwst.08708.28315*.95491*Newcty8.33651*.010481.21120Railst.29179*.14296.79214*Rd062.09906*1.19993*1.10308*Rvr.26074*.71772*1.32691*Subst.36312*.57882*.41520*Tam.522991.24361*39.32950*X

23、zl.31318*.52759*1.24840*Yizhg91.77109*101.64079*33.57548*Zgc1.49658*.1689123.24940*Rules of the same planner, by questionnaire2.3 Location16. CBD0.330.520.15Total 20 planners surveyed in BICP (planners) and PKU (plan students) Comparison to be conducted17. Town centers0.400.470.1318.Development zone

24、s0.200.290.5119.Rivers and wetlands0.430.250.323. Parcel properties20. Current land use type0.360.310.3321. Parcel area0.290.300.41CategoryPIFWeightRCM1. Basic topography1. Elevation0.320.310.372. Slope0.300.320.392. Accessibilities2.1Transport3. Airportsfacilities 0.260.310.434. Rail stations 0.260

25、.370.375. Highways 0.230.250.516. Main roads 0.300.340.367. Subway stations 0.430.430.138. Bus stops 0.420.400.192.2 Public facilities9. Government departments 0.390.350.2610. Entertainment facilities 0.490.350.1611. Amenities (such as supermarkets) 0.500.320.1912. Medical and health institutions 0.

26、570.210.2313. Educational and research institutions 0.580.210.2114. Banks and insurers 0.360.420.2215. Parks and attractions 0.550.290.1622. Land price0.330.320.354. Socioeconomic23. Population density0.360.410.23characteristics24. Employment rate0.300.370.325. Environment25. Air quality0.460.340.21

27、26. Traffic noise0.560.280.1727. Vegetation coverage0.490.280.2328. NIMBY facilities0.460.360.18Comparison of mined and surveyed rulesWhat has done and will do are generally different,in terms of taste of each land use on various factors.Three scenarios by different plannersABCLand use typeParcel nu

28、mber (scenario A)Parcel number (scenario B)Parcel number (scenario C)R163157130C116146182M1178O462616Total336336336Evaluating scenarios A framework for energy consumption and carbon emission based on planned land use patterns 1.1 Population synthesis for generating resident and business agents (fini

29、shed in BUDEM2) 1.2 Building reconstruction for planned parcels (in progress) 2 Activity- and agent-based impact simulation (challenging) 3 Impact accounting Commuting section finished Long Y, Mao Q, Shen Z, 2012, “Urban form, transportation energy consumption, and environmental impact integrated si

30、mulation: A multi-agent model” in Zhenjiang Shen (ed.) Sustainable Development and Spatial Plan, Springer-Verlag Berlin Heidelberg. A low carbon scenario is possible to be identifiedIdeas could be borrowed from other talks, e.g. Yang Jiang Proposal of a Modeling Approach to Assess Urban Energy Consumption and Carbon Emissions based on Spatial Structure and Form Tony Hargreaves Estimating the building stock from regional model forecasts and its low carbon potential Z