title THE interactive SPATIAL ARRANGEMENT and 分析的 model for ACHIEVING SUSTAINABLE 布局优化 Casestudy: hongkong socail housing building EN abstrac 1 This paper is motivated by the fact that in Hongkong, approximately 460,000 units are demand for Complexity and high-density social housing from 2015 to 2025, and Hongkong Authority upgrades strategies for adopting a new library of Modular Flat Design for mass customization in social housing.however traditional space layout methods rely heavily on the static and sectoral approaches, which can't fully fulfill the satisfaction of fast Variety generation and Optimization of complex and extensive adjacency requirements. To this end, we developed an interactive 3d spatial layout method for the fast automatic social housing space planning layout and optimizing tool can help architect and citizens to collaborate on complex statistical adjacency requirements for fast prototype generation. users can modify both architectural elements and topology relation iteratively based on real-time visualization of the sustainable performance. It can be used to Different geometries and social conditions to fit different projects. For the generation of 3D social housing layout, the designer can start the planning layout process by entering the spatial model data recorded in excel and the architectural framework designed by the architect, and then defining the interconnection of various model spaces And generate the 3d prototypes by the physical model which simulating the analogy of force based on architecture’s adjacency requirements. At the same time Offering real-time space feedback on spatial performance. A novel index Spatial proximity average distance was jointly proposed to evaluate the obtained results. The system produces the output within 10 seconds, which is drastically faster than the conventional manual workflow.The algorithm is designed for Grasshopper, physics engine Kangaroo, spiderweb, all running in the Rhino 3D environment. The algorithm was designed for Grasshopper, the physics engine Kangaroo, and spiderweb, all running in the Rhino 3D environment. Finally, we demonstrate how the designs can be compared based on various performance analysis measures and discuss the limitations that arise when our approach is applied in Hongkong. The results indicate that the tool can quickly generate various spatial prototypes and reduce the distance of modular with adjacent needs. The interaction of the analytic and generative process reduced the time and cost of the initial design stage The project helped architects and non-professionals to work in synergy to optimize strategies for the arrangement of social housing units in a way that coordinated a top-down activation of the urban regeneration process at the neighborhood scale Abstract版面2 版本2 the arrangement of Modular Flat Design for mass customization in social housing is a complax task The synergy between the top-down design strategies of architects and the unique bottom-up spatial adjacency requirements of the public takes time and costs. Fast integration of space layout and performance in the early design stages facilitates design enhancement. However, the classical approach is independent rather than holistic, and it is based on a very limited number of observations and sectoral approaches. To this end, we developed an interactive 3d spatial arrangement method for the fast automatic social housing space planning and analysing The tool helps architects and citizens collaborate to count the adjacency requirements of complex spatial relationships for fast generation People can interactively modify architectural elements and topological relationships to suit different projects based on real-time visualization of space performance. It can be used in a different geometric and social context For the generation of 3D social housing layout, the designer can start the planning layout process by entering the spatial model data recorded in excel and the architectural framework designed by the architect, and then defining the interconnection of various model spaces And generate the 3d prototypes by the physical model which simulating the analogy of force based on architecture’s adjacency requirements. At the same time Offering real-time space feedback on spatial performance. A novel index Spatial proximity average distance was jointly proposed to evaluate the obtained results. The system produces the output within 10 seconds, which is drastically faster than the conventional manual workflow. The algorithm is designed for Grasshopper, physics engine Kangaroo, spiderweb, all running in the Rhino 3D environment. a case study is Hongkong Authority upgrade strategies for adopting a new library of Modular Flat Design for mass customization in social housing. Finally, we demonstrate how to compare designs based on various performance analysis measures and discuss the limitations that arise when we apply our method in hongkong The results indicate that the tool can quickly generate various spatial prototypes and reduce the distance of modular with adjacent needs. The interaction of the analytic and generative process reduced the time and cost of the initial design stage The project helped architects and non-professionals to work in synergy to optimize strategies for the arrangement of social housing units in a way that coordinated a top-down activation of the urban regeneration process at the neighborhood scale Introduction版本2 promblem Despite being regarded as a thriving international city, housing crisis remains a serious problem for Hong Kong. As one of the most densely populated cities around the world, the population density of Hong Kong has reached 7,140 people per square kilometer in 2020. There are about 3.3 million people as 44.7 percent of the population living in the social houses. According to the report by the LTHS Annual Progress Report 2015, 460,000 units are estimated to be required from 2015 to 2025. Currently, there are about 720,000 families as 30 percent of the population living in the rental housing units. The targets of Hong Kong Housing Authority are providing approximate 15,000 new rental housing units per year, and maintain the waiting time around 3 years for the applicants on the Waiting list of the rental units. (2011 Hong Kong Housing Authority) With the increasing demand for housing, in 2008 the Authority adopted a new design strategy based on design principles of “Functional and Cost Effective”, which is massy customizing public rental housing. ![](https://www.filepicker.io/api/file/0fGVUIVRv2Jw95GPwGIW) This new library covers a whole spectrum of small modular flats (1-Person / 2-Person and 2-Person / 3-Person flats) and family modular flats (1- Bedroom and 2- Bedroom flats). Individual project architects are to adopt these standard modular flats to make up their building blocks to suit individual site configurations. however the arrangement of Modular Flat Design for mass customization in social housing is a complex task The synergy between the top-down design strategies of architects and the unique bottom-up spatial adjacency requirements of the public takes time and costs. fast generation space layout with the performance during the early design stage can be beneficial to design enhancement and help architect and Non-specialists The sustainability factors, such as social relations, public demand, environmental factors, economic, and sustainable mobility are challenging to be simultaneously considered with space layout planing requirements together to make the smart city decision but classical approaches are based on very limited observations and sectoral methods. There has been much research and discussion regarding which computer automatic space layout program in 2d, but lack of interactive 3d methods which not only can fast generate space layout also can optimize prototype space connection based on Real-time visualization of space performance object To this end, we developed an interactive 3d spatial layout method for the fast automatic social housing space planning layout and opitimzing in initial design stage object is that The tool helps architects and citizens collaborate to count the adjacency requirements of complex spatial relationships for fast generation People can interactively modify architectural elements and topological relationships to suit different projects based on real-time visualization of space performance. It can be used in a different geometric and social context the model attempts to combine the gnerating supported by phical parametric model and visulazing Real-time sustainable performance indicators as an Integrated tool for fast generating prototypes and 优化 The more detailed objectives are 1 A collaborative platform allows architects and non-professionals to record the data of social housing quantities and different module types by excel and automate the generation of geometries. It is also possible to statistically process the proximity and correlation of the geometry to the spatial topology 2 generate space layout by stimulating in automatic and an interactive way which designers can not only modify the geometry but also the relation of space adjection 3 calculate and visualize in real-time the average distance of all spatial modules to all other modules which show the accessibility and spatial adjacency back ground2 automatic Space layout planning for complex issue The automatic spatial layout of computer methods plays an important role in the solution of complex problems during the design process. The complexity of spatial planning has been one of the major issues, which has plagued researchers for a long period. The complexity of design and construction is increasing while the duration of the initial design step is significantly decreasing, which change the activity of the space planning design (Donato 2017). #complexity Space layout planning can itself be broken down into two problems, determining topological or qualitative properties, and determining geometric or quantitative properties (Jo and Gero, 1998; Flemming, 1989). The traditional urban design methods have a significant drawback, which heavily relies on the static and sectoral approaches, which cannot fulfil the complex and dynamic SLP requirement. However, optimization of the traditional method is important (Miao, Yufan,2017). #traditional aslp can solve complex social relations problems There is a strong relationship between social relations and space plans. There are 3 sustainable facets that are connected tightly with the space plan: social, economic, and environmental. Hillier and Hanson (1983) presented “space syntax” as a concept that is associated with social relation. #social relation aSLP is not only deal with the relationship between geometrical and topological but also includes merging the co-working units, minimizing material costs and transport distances, improving productivity, satisficing the residents and maximizing the daylight using. #susbility fator add to slp Monjur MOURSHED PhD said the optimization of SLP aims to minimize the consumption of the energy used (the sum of heating, lighting and cooling energy requirements) and increase the compactness of planning. He also mentioned the implementation of the genetic algorithm aims to optimize the size, scale and the relative position of the units. A mature algorithm should minimize energy consumption and satisfice the requirement of topology and geometry in the design process. In recent years there has been a lot of research and discussion on 2D computer-automated spatial layout procedures for solving complex problems with abstract models. The spl tool is difficult to find a layout that satisfies the condition, real-time performance enables us to develop the possible solution in a dynamic way. #complexity FAST gnerating space layout with the performance during the early design stage can be beneficial to design enhancement. visualization of SPL space adjacency connection space adjacency and closer proximity was Extensive used to analyzing and assessing the space relationship performance Based on Boutabba, H. and Farhi, A. (2011), importance is incorporated into the process of generating the spatial configuration and building space and is necessary for its social organization and intrinsic components. During schematic building design, an architect develops an overall aesthetic motivation for the building, while trying to arrange individual spaces so they meet the adjacency and area requirements specified in the functional program@1999Arvin, Scott A. adjacency and requirement This article explores the utilization of Grasshopper (a Rhino plugin) as an analytical tool to graphically represent a three-dimensional analysis of adjacency requirements in space layout planning. @2015Boon, Christopher Christopher Boon 2015 made a unique system of analysis that can create an extensive range of otherwise unexplored solutions to space planning when faced with complex adjacency requirements and a large number of programmatic elements. SPL generating part and analyzing topological space relations performance part is individual and static. The analysis is based on a simple factor such as the position or function of the outcome for optimization. However, in a real architectural design process, physical and non-physical influences such as building space requirements, building function, socio-spatial relationships connect, building geometry morphology, and architect strategy are all acting on the outcome. Interactive 3d methods fast generate space layout also interactive ways optimization of the scheme prototype based on Real-time visualization of space performance \ The visualization of a dynamic, real-time, complex system of influences is the subsequent work to be done SPL by an interactive physical model interactive physical model as a parametric tool connect the SLP fast generation part and Dynamic modification based on real-time feedback of space agency performance Although researchers have developed many researched to implement automating design process, most of these researched lack the direct interaction with aimed users and go against the design language that architects used. 交互很重要 #interaction Physically-based space planning is a means to automate the conceptual design process by applying the physics of motion to space plan elements. Baraff uses that method to model the realistic behavior of rigid bodies in resting (non-colliding) contact (Baraff, 1989) and building objects by dynamic constraints (Barzel & Barr, 1988). Liggett and Mitchell, 1981; Flemming and Chien, 1995). 都提出了approaches to automated space layout planning use an iterative process of constructive initial placement, in which spaces are positioned one at a time however In these approaches, an ordering function is needed to determine which space to position first. Arvin & House work. They applied the mechanical metaphor technique to the generation of architectural floor plans by using the analogy of mechanical springs and dampers to model a variety of design objectives (Arvin & House, 2002) The tool introduces an algorithm that simulates the analogy of mechanical springs through an interactive simulation of a parametric model.@2016ElSayed, Mohamed automatic multi-stories space planning tool that helps low-income citizens to modify their prototype housing provided by the government. The above cases lack consideration about Integrity space requirement. The bottom-up demands of the residents part and architect strategy of generating part and analyzing spatial relationships performance are individual and in a step-by-step approach. it is a challenge to build a holistic platform to address the full process of architectural design methord frame the space arrangement framework has been designed in 4 modular components, namely (a) input, (b)design and decision components, (c)space layout planner (d) performance evaluator ![](https://www.filepicker.io/api/file/C4zxZZSSW2FexObQXvpQ) software implacation The theoretical approach that authors use is the simulation of the analogy of mechanical springs and boxes collision to reorganize and amend architectural spaces proximity. The authors developed the algorithm in the generative design tool Grasshopper and the live physics engine Kangaroo,Topological and spatial syntax visualization tool spiderweb, all working within the Rhino 3D environment “SpiderWeb”enables the use of Space Syntax in parametric modelling and its pursuit in computational design 2013 Richard Schaffranek ![](https://www.filepicker.io/api/file/lBf9DXZZTwGwF1s6I959) 这个我截图的要模仿的和文字没有关系 input Prototype information modelling Qualifying inputs for the main building structure and boundaries. Determination of the operation areas start by input the geometry in Rhino (a)Determination of the operation area. the main structure remains the structure gathering point will be set as a gravitational origin such as a building core, which later modular will make automatic space planning layout and they will be arranged around it. (b)Prepared a 3d volume initial position modular. The range of distribution of the prototype space, the initial position will affect the final arrangement result (c)Result. After calculating the 3D arrangement volume. The tool will automatically fill in the shapes with a three dimensional, 3d box grid that can also be adjusted to match the main structure and boundary as desired. The demation of 3d box gird will be defined as the smallest multiple of a constituent modular. Prototype information modelling Utilizing the lunch box in Rhino and enter the parameter specifying the design intent in Excel. Parameter transforms information into a 3d physical model. Modifications appear instantly on the software screen as a 3D-colored model, users can change the parameters which are Rome name, Function, Spaces information, Number of charts, 3d scale (Area and height can be defined by designers) （which should be considered dependent on minimax factor of 3dgrid ）The module space will be modelled automatically and randomly assigned within the volume. The designer can adjust the initial position of the module according to a vague intention, in this step the architect can optimize and adapt the spatial prototype to any shape and initial position as long as the scale modulus of the prototype matches the 3d gird. design and decision components Adjacency relations input to physical models of mechanics. Macroscopically, the architect needs to input factors such as the main structure of the building, sunlight, orientation, boundaries, main gathering points, etc. into the linkage and gravitational weights of a particular spatial unit. First, the design requirements and sustainable elements will be abstracted into a rigid body of suitable objects static, which will then attract dynamic units with specific weights. The result of the operation will be close to the side of the old people's room that gets more sunlight (similarly cases such as commercial units close to the main street, greenery units close to public spaces, etc. can be freely defined and taken into account in the design). The physical gravitational relations of the system are defined according to the adjacency relations and specific design requirements. The algorithm is based on the GH platform of the universal modelling software rhino which specifies the links and adjacency weights between units, which facilitates specific model modifications for architects faced with specific needs for various building situations. The authors divided the spring forces link into groups to achieve the hierarchy in the desired space relations. (a) Micro-relationships refer to the adjacency between the unit and the unit (b) Macro-relationships refer to the adjacency of all unit groups to the main body of the building and to the environment. (a) Microscopic adjacency between units and unites This program adjacency matrix was introduced to assist in defining complex relationships and weights between units and unites. The architect can complete the matrix based on client interviews and smoothly define complex unit-unit relationships. The numbers on the charts represent the various needs of the client. The matrix forms a hierarchy of space in terms of priority and ranks the adjacency requirement of each element in relation to all other elements in the chart. There are three different levels of importance indicated by tones. (b) At the macro level, the architect needs to input factors such as the structure of the building body, sunlight, orientation, boundaries, major gathering points, etc. into the linkage and gravitational weights of a particular spatial unit. First, the design requirements and sustainable elements will be abstracted into a rigid body of suitable objects static, which will then attract dynamic units with specific weights. The result of the operation will be close to the side of the old people's room that gets more sunlight (similarly cases such as commercial units close to the main street, greenery units close to public spaces, etc. can be freely defined and taken into account in the design). space layout planner Simulation and interactive adjustment Run the simulation. Based on Kangaroo, simulation can be implemented of the analogy of mechanical springs and unit collision to fast generate the unit combination by inputting spaces proximity relationship. The definition creates a line connecting the units for every adjacency requirement. A higher value attractiveness indicates closer proximity and more optimized adjacency. The result is caused by Various gravitational forces in space for spacesuit attraction and repulsion. users can modify the result in a ineractive way by changing the architecture elements or adjacency requirementThe simulation goes mainly in two steps. a. Topological adjacency operations unit and unit archive the space proximity relations by using ‘spring force’ mechanical simulation to generate the initial location of individual units and affects how units relate to each other. The forces were abstraction into axial forces， one point at the centre of the unit connect to another point with the line, the physics engine Kangaroo simulates the spring force physics. box-collide was defined to make sure the results of the operations are closely linked and orthogonal. b. Calculations between groups For the global all-unit attraction by the building space, meeting the objective of filling the building boundary. By using another Kangaroo engine. The vertical traffic space is generally set up as a core of attraction to ensure the realization of traffic in the multi-storey arrangement. The centre points of the unit connect with architecture geometry centre point. architecture volume boundary plays a role as a big box-collide as a container to full will with units. The design of the built form will be translated into rigid volume boundaries and spaces will meet unique spatial forms while satisfying topological adjacencies. Users can add or delete any design factor which can be defined as a geometry for interacting with units, such as sunshine, wind direction, preferences of floor level, etc., which can be abstracted as physical gravity models. c. Parameter requirements The range of force value designed as a slider bar component of grasshopper from 0-1000N.users need to consider the weight of various demand factor to make a decision. ![](https://www.filepicker.io/api/file/gSkZhUxdRVuscnT0ZfI1) performance evaluator 空间的 Spatial layout modification and real-time visualization of average path distance performance average distance and spcae adjecnt value were jointly proposed to evaluate the obtained results. Space proximity fitness（realtime） The definition creates lines between program units connecting each unit for every adjacency requirement. the line's total distance will add up and average processing, which can be weighted to the average distance A lower value indicates closer proximity and more optimized adjacency. users can consider and try to minimize this total distance to improve fitness. Shortest Path Average Distance (static) When the space layout is determined, the detailed spatial model is represented. The system automatically calculates the average distance of the shortest path from each space point to each space point on the plane Showing accessibility of each space .low value represent a better accessibility and publicness Utilizing the plugin named spiderweb in GH, which provides functionality for preforming calculations on graphs. The library is created with a special focus on the integration of graphs and graph-related theories (e.g. SpaceSyntax) in parametric design environments. develop an algorithm to generate a visualization of sustainable performance factors. ![](https://www.filepicker.io/api/file/kf2rzE2QkmkXXRYybJht) ![](https://www.filepicker.io/api/file/kf2rzE2QkmkXXRYybJht) Space Syntax can be used to predict, qualify and quantify aspects of human behaviour of different design proposals and the impact on their neighbourhoods. Evaluation and Revised design After stimulation is completed, users will get resulting solution with visualization performance showing spaces proximity value, accessibility. If the prototype is accepted by considering the performance of space character in Realtime, users can bake the model to Rhino. If not, users can modify inputs factor by 1 changing the modular or architectural elements position 2 changing the force value of different step string 3changing the architecture form volume. If the simulation step was wrong, the system will send warning messages to guide them. Agent specific physical model After calculating the layout of the unit boundry, designers make a choice of plan, the designed modular with detail will be proxied to the appropriate location. ![](https://www.filepicker.io/api/file/TzQU8KbQT6iK6w7X5d1V) hongkong case study our tool was the application to test application high complexity in spatial relationships and This Archiford LLC archiford.com architectural studio designed social housing concept design and merge the data according to Hongkong authority specifications in 2018. With this expertise, we have used data source of Hongkong client interviews about spatial relationships and social-spatial unit requirement posted by Archiford 2018.6. using x units with different characters shows in excel and adjacency matrix for Hongkong social housing project (Table 1), a site in Hongkong with an irregular boundary volume and five height floor which is the part of social housing was chosen to test interactive space arrangement tool in physical engine Kangaroo. Figure ![](https://www.filepicker.io/api/file/AOAdK8kSrKEwGIu95yUH) This Archiford LLC archiford.com architectural studio designed social housing concept design and merge the data according to Hongkong authority specifications in 2018. With this expertise, we have used data source of Hongkong client interviews about spatial relationships and social-spatial unit requirement posted by Archiford 2018.6. using x units with different characters shows in excel and adjacency matrix for Hongkong social housing project (Table 1), a site in Hongkong with an irregular boundary volume and five height floor which is the part of social housing was chosen to test interactive space arrangement tool in physical engine Kangaroo. Figure Figure X-x depicts the Grasshopper definition for this specific case and generates the calculation procedure, using the steps outlined in Section 2.2. programme的参数定义了units 属性，引力相邻关系 ，图 ![](https://www.filepicker.io/api/file/H5TXd0PgSaKdkDd9ZROO) ![](https://www.filepicker.io/api/file/2GKmUlW0QrGPqAM01Cs0) 图 运行过程 ![](https://www.filepicker.io/api/file/NUaAqqqSQuToVFGYu2BA) 图 一个方案 ![](https://www.filepicker.io/api/file/1wG1aHkbQK9fKT85uSkH) 平均距离图 多个方案 。 space distance value reslut x shows the x ![](https://www.filepicker.io/api/file/Dqo3nbyMSlu6HyQyOego) Recursive Shadow Casting - Preforms recursive shadow casting on a visual graph grid and outputs the visual graph. (Turner, Doxa, et al. 2001) The definition creates a line connecting the origin points for every adjacency requirement. A lower value indi- cates a closer proximity and more optimized adjacency. The definition adds up the total distance of all adja- cency lines, which can be weighted to create greater attraction or repulsion in order to prioritize certain adja- cencies over others or ensure two program elements are isolated from one another. The Galapagos evolutionary solver will try to minimize this total distance to improve fitness. 作为该工具的延展 把楼全局做了户型得到相应的方案 ![](https://www.filepicker.io/api/file/EhBw1zhzTeWdfFczXJ2P) ![](https://www.filepicker.io/api/file/xuWkiffqTROn6XfBwJ2g) Discussion and future work We feel that our prototype system provides a convincing demonstration of the attractiveness of the responsive design approach and the use of physically-based methods in implementing this approach. its spend 20s for generating Interactive 3d methods can not only fast generate space layout but also interactively optimize the prototype solution based on Real-time visualization of space performance The fact that the physically based objectives do most of the work allows one to easily explore many design possibilities Design solutions are created automatically from a small set of design objectives, and the system is very flexible in that design objects can be easily manipulated, allowing new designs to emmerge. Our current implementation is highly interactive and results are produced in real time, but whether or not this will hold for more complicated designs is an open question. From the results shown in the case study, the requirements from the social housing in Hongkong project can be fulfilled with our tool and showing the muti performance parameter in real time. The arrangement layouts can be automatically generated based on the ajanctive need and units characters record in excel . Moreover, the designers can interactively modify the space arrangement layout by changing individual unit to achieve satisfying results. advantage 1.The advantages of this tool, particularly with social housing, are that the system uses multiple layers of the gravitational system as a holistic simulation result. This allows critical spatial relationships to be arranged in an adaptive way to simultaneously consider the requirements of multiple factors solutions. For example, the elders' space in figure x always is placed based on rules of Closing to the stairs and closer to the south side for the light at the same time. The adaptability of the system can help designers solve the more complex actual situation People from multiple directions can collaborate are that the system is simulated with multiple layers of gravitational system holistic results 2 Regular and orthogonal space-filling calculations are implemented in literature such as xxx and xxx. In this system figure x shows irregular physical forms can also be filled in, the logical of filling logic need to be defined by manual to fulfil the vertical traffic, but this system makes a contribution for free automation of built form space arrangement. even for complex building designs, in the reasonably near future. 3.the major advantages in using this system for space arrangement for social housing is the fact that the visualization of the result performance would be modified and decided for the architect in the conceptual design stage when they are facing a large number and complexity of initial problems. From section x to x ，designers can generate the final space prototype and the plan characteristics in a few seconds. 4 People from multiple fields can collaborate based on Building information model. The model not only records the physical geometry but also the information such as the weight of adjacent space, space dimensions, complex unit characters and relationship, and the performance corresponds to the generation of each programme. Using general language for cross-platform collaboration，especially in hongkong case using excel as the input of data, it can save time costsand for enhancing management of the lifecycle. limited and future work In this case, a lot of manual work is required There are several limitations to this script. In some results, solutions were loose and not compacted in setback (Figure 4:b, c, d and f). This script currently treats units as dynamic gravity element in Kangaroo during the Calculation process. For many cases, too much mechanics act together and the whole system does not reach a steady-state to give results. This could lead to units gradually become disconnected with each other at a long time Mechanical interference. Further development of the script would be necessary to put more mechanics control structural optimisation on programs for generating a stable result. Some spaces were not accessible (Figure x ，because the unappropriate values were set for the spring constants. In this case, a lot of manual work is required to adjust the force value to fix to unit single-unit accessibility. In future work, finding a range of mechanics values that apply to more range design situations is necessary. For a systematic mechanics value range, data experiments need to be carried out in large numbers The building design is concerned with satisfying various assessment objectives, which are sometimes in conflict with each other. The performance data showed in figure x space performance with the good shadow casting but a long traffic flow, when faced with a large number of this kind of units, its a challenge for People to globally optimize. Performance of various indicators, therefore, needs to be assessed as a whole rather than in isolation in the near future step. The palapas multi-factor optimisation systems or ai data learning would be a scalable direction for global holistic assessment of space to calculate fitness rankings conclusion From our case study, it has been proved that our physically-based tool can efficiently and effectively implement automatic space arrangement and performance feedback process in practice. Cooperation of designers in different fields can benefit from the interactive 3d prototyping space arrangement which saves a lot of manual work and requalify the synthesis of 3D units arrangement destined and Complexity in 3dspace to improve social cohesion with a view to sustainability and densification. However, it has also been found in Hongkong case study, that the physical gravitational structure and range of gravitational values have their own limitations to implement the needs of More accurate space arrangement the problem. Although all errors can be adust by manual optimisation as intermediate solutions, a better mechanics hierarchy and parameters range is required to solve the fully automatic problem. In future, more systematic mechanical architecture and accurate weighting parameters for optimization methods and holistic evaluation strategies should be developed to better fulfil the space arrange synthesis. We have introduced the concept of responsive designdeveloped. We have also demonstrated, with a simple prototype system, that the concepts of physically based modeling provide concrete principles around which to construct such tools. O Acknowledgments This work was supported in part by an architecture design company archiford ltd . 图 ![](https://www.filepicker.io/api/file/ogfqhleRmeNVHdKvslC9)