Computational approaches For Cities urban patterns The case study of Sale Medina MASTER THESIS 2020
1
Curriculum Vitae BASIC & CONTACT INFO
Jaouad AKODAD Tel : +212 6 32 81 37 11 Instagram : Jaouadakodad LinkedIn : Jawad AKODAD 08/02/1996, Nador
Nationalité
Marocain
Jaouad.akodad@e.enarabat.ac.ma
LANGUES
HOBBIES
français | Anglais Arabe | Amazigh
Dessin | Music Photographie
SKILLS DESSIN TECHNIQUE : • Autodesk Revit • Autodesk Autocad M O D E L I S AT I O N 3 D • Autodesk 3DSmax • Rhino/Grasshopper R E N DU / P O ST- PR ODU C T I ON : • V-ray • Corona • Lumion • Adobe Photoshop • Adobe Indesign CAD TO CAM / PRODUCTION : • Mach 3 • Autodesk Fusion 360
32
Naissance
ÉTUDES
2014 - 2017
2017 - 2018
2018 - 2020
Ecole Nationale d’Architecture de Fès
Fès, MAROC
ENAF - Lisence
http://archi.ac.ma/
Faculté d’Architecture et d’Urbanisme
Mons, BELGIQUE
FAU-UMONS - Mobilité
http:// web.umons.ac.be
Ecole Nationale d’Architecture de Rabat
Rabat, MAROC
ENAR - Master
http://archi.ac.ma/
• Workshop - Rénovation de Dar El Baroud en
Meknès, MAROC
WOR K SHOP S
2016
04 Fev
-
13 Fev
Maison d’Architecture - Ordre de Meknès 2017
09 Nov 13 Mar
-
16 Nov 20 Mar
• Workshop - Conception de la cité du Clos de l’Age d’Or - société Centr’Habitat
Haine-St-Paul, BEL-
• Workshop - Réaménagement des 7 placettes de
GIQUE
la médina de Fès - EuroMed Fès 2018
05
Juil
-
12
Juil
• Workshop - Etude de la réhabilitation du
Fès, MAROC
complexe thermal Sidi Harazem - Mme Aziza Chaouni 2019
27 Mai - 01 Juin
• Workshop - Réalité virtuelle - Réalisation d’un
film en réalité virtuelle
Sidi Hrazem, MAROC
EXPÉRIENCE PROFESSIONNELLE
2015
01 Aou - 31 Aou
• Stage chantier - Rharrit Architecte
Meknès, MAROC
2016
11 Juil - 11 Aou
• Stage Agence - GEA Architects
Madrid, ESPAGNE
2017
01 Juil - 15 Aou
• Stage agence - Archimad studio
Meknès, MAROC
2018
01 Aou - 07 Sep
• Stage administratif - Ministère des l’équipements publics
2019
17 Fev - 17 Mai
• Stage agence - Archimad Studio
Meknès, MAROC
2020
01 Sep - 04 Mar
• Stage Agence - Said Berrada Architecte
Rabat, Maroc
01
• Stage Laboratoire de recherche - TIC Lab - UIR
Salé, MAROC
jui
-
31
Dec
34
COMPUTATIONAL APPROACHES FOR ANALYTICS AND URBAN DESIGN
A
bstract. This project focuses on the use of computational tools to develop a data driven approach for an analytical study about different urban systems. This “framework” examines urban Big Data of the historical medina Sale in Morocco. The computational tools are more effective to provide in( ENA )of Ecole Nationale d’Architecture sights within complexity, becoming a key to generate more efficient solutions Rabat, Maroc throughout the design process. The findings of this study highlight the poten( TIC - Lab ) Technologies de l’information et de la communication, UIR tial of a data driven approach to explore analytical aspects and move further to Sale, Maroc generative designs, using algorithms. December, 2020
Keywords. Parametric, Computation, Big Data, Urban Analytics
Project Name : Numerical Identity, Case of Sale Medina
Tutors: Mr Mounir Ghogho ( TIC-LAB ) Mr. Mohammed Bakkali ( TIC-LAB ) Mme Mouna Mhammedi ( ENA ) Jury: Mr. Nabil Rahmouni ( ENA ) Mr. Mohammed Akazaf ( ENA ) Mme. Sebti Dilar ( ENA )
54
56
76
Content : Introduction Methodology
9 11
Part I : Computational Approach for Urban Analytics Phase I : Data mining and the city
13
Phase II: City modeling using Computational approach Phase II : Urban Network Visualisation Phase II : Energy Modeling Phase II : Thematic Classification of urban activities Phase II : People Flow simulations:
15 17 19 22 24
Phase III : Generative Design Process for Tensile Structures
28
Part II : Concept of Emergence of the Islamic Geometric Patterns :
30
Possibilities of emergence using shape grammar and rules Infolding the polygonal structure of a single pattern Results as Graphical design Concept of Emergence for building typologies Rules definition and possibilities of typology connections Negative space generation from the emergence process Step by step process for the master plan generation Growth process / Densification Emergence variations Generative master plans Conclusion
32 33 34 26 38 39 40 42 43 44 48
78
I
ntroduction: The technology of information and communication (TIC) had an important impact in the development of technology as well as the design practice. Technology is always redefining new boundaries by the introduction of new tools that simulates architects and designers’ creativity, allowing them to explore new concepts and innovative methods. The introduction of the TIC in the 60’s for the design field had a significant impact by replacing the hand drawing tools with computational aided design (CAD). Afterwards, these tools had been improved as a base for architectural design, interior design, urban design and so on and so forth. They have a strong ability to produce efficient results and add a fresh perspective to these fields. The integration of this technologies on the design practice remains largely underexploited, presumably because technology is advancing faster than the construction industry. In order to benefit from the true potential of computation, this has to be implemented from the early stages of the design process, not only as a drawing tool, but introducing new possibilities in the use of computation in both the analytical phase and the design process and by developing a very strong bond leading to a generative design where analytical data is exploited smartly. The possibilities derive mainly from the increase of data control and creating new computer-oriented approaches that benefit from the computer processing power.
98
910
M
ethodology : This project provides an urban study of the historical medina of Sale in Morocco. Through computation, urban data will be implemented in a networked system, generating insights about how the city’s related systems respond and perform. The study will start by setting the sources of data to be used (data mining), and how they are managed in a dynamic way using algorithms and visual programming (VP). The following chapter will focus on the construction of relationships between data and the city, and will examine urban patterns through the network of data analytics. It provides a form of creating smart city based on computer and human interactions, in which urban patterns are investigated to inform a set of solutions using computational modelling, and parametric design.
11 10
PHASE I /
Data Mining :
Data of Sale Medina, Morocco
PHASE II /
Visual Analytics
3D modelling Mobility Network Visualisation Energy Modelling Thematic Classification of urban activities People Flow Simulation
PHASE III /
Generative Design
Tensile Structures for public spaces Urban Design of mixed housing
11 12
D
ata mining and the City: Nowadays, cities become more complex and more dynamic to be apprehended making it unbearable to manually process its informations. At the same time, Big Data is increasing considerably, becoming more available and elaborated, and providing massive information about all types of behaviours and patterns within cities. Information extracted from Big Data will enhance the knowledge and the level of accuracy in decision making regarding the city. From the perspective of architects, urban designers, and planners, and considering the tools with which they work, many aspects of data can be supported in the design process. With the support of new spatial technology platforms and big data, this work will tend to understand, evaluate, and manage urban patterns and evolution
13 12
018
-6.8
4
.8 o -6 68 T
34
.01
82
Elevation Data To 3
4.0
50
9
Shuttle Radar Topography Mission (SRTM) data are used to reconstruct the topography of large regions, which consist of Digital Elevation Models generated by NASA and the National Geospatial-Intelligence Agency. SRTM topographic data consists in 90-meter (295-foot) pixels spacing for elevation’s information of many part of the world. The figure illustrates the generation of topographic levels of Sale city that correspond to (N34 W006). Within the visual programming platform of Rhinoceros, many scripts are provided to help convert Hgt files into geometrical surface, and then, applying horizontal contours in order the visual the topography curves.
Morphology Data Two dimensional (2D) urban morphology was provided from CAD files and managed into a Shapefile to contain the elevation data. The idea is to extrude each plot with the corresponding elevation value in order to construct a 2.5D model.
Mobility Data OpenStreetMap in a non-profit organization providing geographic diverse user-based data of addresses, places’ names, and route planning. Once the data has been collected, it is entered into an algorithm that manages the infrastructure network.
Energy Data The generation of graphical data such as the sun path, sunlight hours, radiation studies, and more are proceeded by providing specific inputs. In one hand, Energy Plus with weather data (EPW ) of the city of Sale. In the other hand, the geometry of the city, which is the 2.5D model of the medina. In order to achieve more effective results, a projection will be applied to the geometry geo-located at the surface of the topography levels.
13 14
Phase II: City modeling using Computational approach
Building shape data
Building elevation data
Topography data
14 15
15 16
Phase II : Urban network visualisation OpenStreetMap in a non-profit organization providing geographic diverse user-based data of addresses, places’ names, and route planning. Once the data has been collected, it is entered into an algorithm that manages the infrastructure network. This figure visualizes how mobility is distributed from highways to pedestrian pathways.
16 17
17 18
Phase II : Energy Modeling Recently, many tools allowed new approaches of energy modelling and thermal simulation. Ladybug is an open-source plugin implemented in Rhinoceros/Grasshopper to help designers explore relationships between environmental data and buildings geometry models. The process of making energy models within the design platform is more effective to support complex geometries, i.e. the urban morphology of a historical medina. The using of energy modelling to endorse environmental analysis within architectural practices has been noticed to reveal more useful solutions when coupled with design processes. The generation of graphical data such as the sun path, sunlight hours, radiation studies, and more are proceeded by providing specific inputs. In one hand, Energy Plus with weather data (EPW ) of the city of Sale. In the other hand, the geometry of the city, which is the 2.5D model of the medina. In order to achieve more effective results, a projection will be applied to the geometry geo-located at the surface of the topography levels.
19 18
Wind Rose ( m/s ) Location > Rabat/Sale Range > All Months
19 20
21 Juin 6:00 > 19:00
21 December 6:00 > 19:00
1 0 Shaded Area Visualisation Per 1 Hour
20 21
12 11 10 9 8 7 5 4 3 2 1 0.00 Sun Hours Visualisation 21 December / All Day
21 21
Phase II : Thematic Classification of urban activities Decomposing and analysing parts of the medina is significant to reveal patterns of its urban systems. In this process, the traditional techniques may not be capable to evaluate such complex activities neither to deal with different parts acting simultaneously. Meanwhile, using computational tools can process such complexity, and manage human activities into thematic classification to form six main themes: Retail, Artisanal, Tourism, and Culture.
22 22
23 23
Phase II : People Flow simulations: Thus, the evolution of informal activities through time is related to city patterns, that are specific to a domain from where data is collected. By integrating the distribution of the thematic activities into the algorithm, a zone of people moving in and out can be predicted by connecting all the locations of each thematic classification. The people flow visualization is based also on the existing urban roadways that provide information about several hierarchical levels of the mobility network. Furthermore, the people flow will serve to determine the zone where informal activities self-evolve, while public places are considered as the centre of growth of each thematic activity. Whenever a block level is null and its size is larger than 200m2, the urban typology is referenced as public place. The investigation can also examine the potential of each public space to become the centre of a thematic activity. Through computational processing, the visual analytic explores how the city is performing by setting relationship between the public places, the people flow, and the location of classified activities.
Path lengh : 24 24
13 956 m
12 901 m
10 769 m
10 774 m 25 25
Phase II : People Flow simulations: This chapter focus on identifying interlinked relationships between the various sources of data to explore how the city is performing, by integrating them into a networked system based on algorithms. According to this view, the city is seen as a set of complex interactions between its different urban components, which is mainly incorporated in the concept of the Smart City. The urban Big Data become a tool to better inform planners about how the city is responding by giving insights from very large and dynamic datasets. The proposed method proposes consecutive phases, gathering data coming from different sources, and implemented into an algorithm. They are summarized in three main phases: • • •
Definition of thematic activities. Investigation of public spaces. Determination of people flow.
Data Inputs
Mobility network
26 26
Public places
Local activities
Thematic classification
Investigation of PP
People flow simulation
27 27
Phase III : Generative Design Tensile Structures for public spaces
The next stage on the design process consists of the generative process based on the network system output and additional parameters, such as the energy modelling that will serve mainly to evaluate the thermal simulation on the path of people flow. The application of energy simulation in this process aim to study the solar shading in public spaces, in order to optimize long term outdoor thermal comfort. Therefore, the existing building and topography must be considered when creating shading outdoor areas. In addition, the EnergyPlus provides Data ( EPW format) from exact locations to study the solar path of all periods of a day, month, or the whole year. The analytical study results on areas that necessities sufficient shading while avoiding excessive shading that are already created from the surrounding environment. Finally, the resulting areas that need to be shaded, are converted to mesh surfaces, and implemented to a plugin (Kangaroo) within grasshopper platform, to apply physics of textile. The surrounding buildings will serve as anchors, by connecting the mesh surface to the closest building corners. Additional parameters are also taken into account such as gravity force, and the spring strength. Tensile structures are the result of the generation process, which are applied to each thematic path, generating a multitude of shapes mainly related to environmental parameters. visual analytics served to specify areas that necessities a design intervention, rather than applying the design to the whole site.
28 28
Step By Step Process for tensile structures generation
Sun Hours
12 11
10 9 8 7 5 4 3 2 1 0.00
Tensile Structure variations Parameters of variation : Area to shade / Builsing corners / Strengh
29 29
PART II: Concept of Emergence in the Islamic Geometric Patterns : For centuries, Muslim artists used a compass and a rule as tools to explore visual math inherent in geometry to achieve aesthetic qualities. The construction techniques remain unclear to nowadays. Few texts survived about the process behind, and others remain secret. Many construction techniques have been unrevealed, but a heritage succeeded between generation of artisan families. Islamic geometric patterns are served in cover wall with ceramic, zellige. Sculpted wood in doors or stuck in ceilings. Construction techniques can be recreated by analysing the underlying structures of existing ornamentations. Such aesthetic quality is achieved with conformity with geometrical and arithmetic relations, which is related to the scientific apogee in the middle east and central Asia during the 8th an 9th century, based on the translation of ancient scientific manuscripts of the Greeks and Sanskrit.
Figure 1
30
Figure 2
El-Khwarizmi, in the early 9th century, wrote the earliest document related to geometry (mohammed,2000). His work played an important role in Algebra, Arithmetic and Algorithms. Many researchers argue that the starting point of any Islamic geometric pattern is a circle. From a symbolic perspective, the circle refers to unity, wholeness. Because of circle has neither starting point nor ending point, it represents eternity which is related to the Islamic idea of the Creation, The one God, as a generator of all forms. (Critchlow, 1976; Akkach,2005). According to this view, Scientists suggest a classification of Islamic geometric pattern according to the circle subdivision. While applying tessellation on a circle, star shapes emerge, by connecting vertices of constructive polygons: A 6-point star emerge from hexagonal polygon, 8-point star from the octagon... Since it is not possible to cover all types of Islamic pattern, this paper attend the study of the 8-point star, or as known in the Islamic culture as : ‘‘Khatam Suleiman”. The orthogonal grid can be used to create designs with eight-point stars. Repeating the same process of tessellation. a whole pattern emerges. The results is highly affected by the infinite variations in the grid, and the process of tessellation.
Figure 3
Rules definition of the 8-point star emergence using shape grammar: A shape grammar is a set of shape rules that apply in a step-by-step way to generate a set, or language, of designs. (Knight, 2000) In this section, Shape grammars are used to analyse existing designs. The goal is to describe isometric transformation within the pattern, such as repetition, rotation, and isometric scaling. The shape descriptions will serve as rules to generate new designs according to their variations: 1. Repetition: The major characteristic of the Islamic geometric pattern is the concept of ‘‘Repetition’’. It can be achieved with at least one or two basic polygons using a grid system. The orthogonal grid consistently filled with 8-point star shape like regions results in very highly visual symmetric patterns. 2. Isometric Scaling: Star geometric shapes can be determined from a circle, from which come the full polygons, including series of roots and proportions. Isometric scaling happens when proportional relationships are preserved as size changes. Therefore, the square includes the root of 2, the hexagon includes the root of 3, and the root of 5 in the pentagon, which is the golden ratio that can be found in the Fibonacci sequence. 3. Rotation: Rotation, on the other hand, is based on repetition around specific points that play the role of centres of rotation. The 8-point star can be rotated on the orthogonal grid, while adding two directions with 45° diagonal lines. four axes of rotation are defined through the square vertices of the grid. When a design is rotated around a centre, the design as a whole remains unchanged and would assume rotational symmetry.
In complexity theory emergence “refers to the property of a collection of subunits that comes about through the interactions of the subunits and is not a property of any single subunit” (Flake, 1998, p. 449). The geometric pattern emergence can be seen as a three-stage process:
Figure a • The first step consists in the definition of an octagonal grid, to set the proportion of the initial 8-point star shape. The final form of the grid is not predefined but has infinite variety that emerges as a result of different shapes generated.
Figure b • The emergence results from intersecting shapes composed of one set of 8-point stars, but in the combination isometric scaling of these shapes, there emerges an infinite range of complex geometric forms.
Figure c • Finally, Rotation transformation rules are applied to the emergent design, which result to more complex result.
31
Possibilities of emergence using shape grammar and rules
32 32
Infolding the polygonal structure of a single pattern
33 33
Results As Graphgical design Supporting the generative design as a process, the result explores new geometric territories through the power of computation. The algorithmic procedure as demonstrated, in the previous chapter employs the same variety of the 8-point star shape, as were used historically and is therefore a contemporary form of an ancient methodological practice. Supporting emergence as process, the result explores new geometric territories through the power of computation.
34 34
35 35
Concept of Emergence for building typologies Grid system for pattern generation
Infolding the structure to define the Typology units
T1 / 400 m2 1st Generation
T2 / 200 m2 T3 / 100 m2
T4 / 200 m2 2nd Generation
T5 / 100 m2 T6 / 50 m2
3 Generation rd
4th Generation
36 36
T7 / 100 m2 T8 / 50 m2
T9 / 50 m2
Detailed plan for a single pattern unit
T6
T5
T4
Typological Unit definition
T3
T6
T1
T2
T5
T8
T4
T7
T9
37 37
Rules definition and possibilities of typology connections
Isometric transformations between parts
38 38
Negative space generation from the emergence process Community Space / Public space
39 39
Step by step process for the master plan generation
40 40
Site delimitation
Urban network integration
Main road extensions
T1 / Residential
T2 / services
T2 / Residential
T4 / public facilities
Public square
²
41 41
Growth process / Densification
42 42
Emergence variations
43 43
Generative master plans Using shape grammars for building’s aggreagation
44 44
45 45
46 46
47 47
Conclusion : Analyzing the literature covering the data-driven approaches in urban planning, we noticed that the use of Big data and algorithmic process are proving valuable knowledge of the city performance. Cities which ensure connectivity of their resources and infrastructures, while producing, distributing, controlling, and monitoring their Big Data are called smart cities. This study represents the findings derived from the use of these data-driven approach applied to a historical context. Besides gathering visual analytics of urban Big Data in order to understand the city behaviours, the findings serve furthermore as inputs to generate urban designs, as final output of the method. The historical context of the Sale medina demonstrates the relevance of this approach to reveal the complexity found within its urban systems. The early stage of the design consists in data visualisation of urban pattern such as the topography, mobility network, urban typologies, and human activities. It is essential to take into consideration those urban patterns in every stage of the urban design, and to inform the design generation using data and computation. What is more is the capacity to increase control of the data collected, while integrating them into a networked system, within the Rhinoceros/Grasshopper platform. The key to a more efficient designs combines the computer processing constructed to manage connections between the information collected from the analytical study and inform the design formation. The visual analytics served to specify areas of design intervention, rather than applying the design to the whole site.
48 48
49 49
Jaouad AKODAD Tel : +212 6 32 81 37 11 50 Instagram : Jaouadakodad LinkedIn : Jawad AKODAD