Algorithms-Aided Sustainable Urban Design: Geometric and Parametric Tools for Transit-Oriented Development

Algorithms-Aided Sustainable Urban Design: Geometric and Parametric Tools for Transit-Oriented Development

Fernando T. Lima, José Ripper Kós, Rodrigo Cury Paraizo
DOI: 10.4018/978-1-5225-0029-2.ch035
(Individual Chapters)
No Current Special Offers


This chapter is part of a research on algorithmic approaches to sustainable urban design. It focuses on the use of computational tools to provide quick and dynamic assessment while planning and discussing interventions in urban areas. The objective is to address the use of algorithmic systems to formulate effective strategies for sustainable urban projects, guided by Transit Oriented Development (TOD) principles. TOD is an urban development model that considers geometric principles and measurable parameters for designing sustainable cities. It advocates the creation of mixed-use neighborhoods within walking distance to a variety of transportation options and amenities, so that basic urban needs are easily accessible. In addition to establish a theoretical framework connecting algorithmic-parametric concepts and geometrical features of TOD, this chapter describes an experimental employment of algorithmic models working on TOD principles, in order to enhance a systematic and dynamic testing and subsequent argumentation for sustainable urban projects.
Chapter Preview


The application of mathematical models as a mechanism for optimization of urban development has long since been studied and proposed by experts. Dantzig and Saaty (1973) advocated the use of mathematical models to evaluate improvements in urban development by optimizing infrastructure in high-density neighborhoods. Salingaros (2010), in turn, demonstrates possibilities of generating urban morphologies from the application of planning legislation as generative code.

On the other hand, Mitchell (1991) understands that the contribution of computational revolution to the design process lies on the replacement of the human intellectual skill by machines that processes information. However, Oxman (2006) considers digital design the one in which the process of shaping the object is highly mediated by digital technology and, yet, the role of the designer remains central. Thus, it is important to consider the process, as it is to highlight it in the designer’s leading role for decision-making.

To Menges (2006) design as subject is a manner of abstracting and assessing possible alternatives of settings, scenarios and concretions without necessarily achieving each possible solution physically. In this sense, relating parametric and algorithmic resources, and their capability to dynamically manage data, can support an interactive design approach. These tools have the necessary flexibility to explore numerous possibilities, allowing the comparison of different options and the choice of the most suitable solutions, according to the adopted performance parameters and criteria.

Computational design has become widely accepted in architecture. However, few approaches use computational resources supporting urban design, in order to develop adaptable masterplans. The algorithmic thinking applied to urban design has it fundamentals on the argument that parametric systems enable the prompt generation of different alternatives of composition from the simple alteration of values of one or more specific parameters, allowing obtaining different scenarios that can be assessed in a way to steer the decision-making. It is a panorama open to interdisciplinarity, to collaborative work, due to the fact that modifications can be easily made and so can evaluations, all this leading to better performance.

The parametric paradigm applied to urban design constitutes a new possibility that is based upon the utilization of patterns and design rules. The aim of this approach is to facilitate the dialogue between different participants of design process, allowing the elaboration of flexible proposals. Parametric systems introduce the possibility of constantly altering models during the whole design process, allowing the generation and testing of several versions within a controlled design environment, just by changing values of a specific parameter. For that, parametric software have become a vital tool since they make possible the visualization of either the ensemble of a project or the intervention and updating of the parts. Computational design and algorithmic-parametric procedures are contributing to change design methods, as they provide ways of exploring multiple solutions – including optimization tools to indicate solutions with better performance.

In the meanwhile, the matter of sustainability currently assumes a central role in the observation of contemporary urban centers and in the dimensions of its development. The socio-environmental picture that characterizes contemporary societies illustrates that the impact of humankind on the environment is becoming more complex, both quantitatively and qualitatively.

It is a fact that cities no longer encompass the current development model. It is vital to think about self-sustainable cities development by means of integrating urban planning, urban design and architecture. In this context, it becomes highly recommended, apart from a new culture of cities, the existence of new approaches of design, which might enable a global vision of the processes that comprise the city and its interrelations. These approaches can help amplifying the cognitive operators of those who think and design cities, potentiating the comprehension and processing of information that should help interventions.

Key Terms in this Chapter

Algorithm: A set of rules that precisely defines a sequence of operations for solving a given problem. It starts from an initial input of instructions that describe a computation that proceeds through a finite number of well-defined successive states, producing an output and a final ending state.

Generative Systems: Computational applications that use algorithms, parameterization, simulation and performance optimization techniques. These applications guide a composition from the evaluation of different solutions through simulation and optimization techniques.

Genetic Algorithm: An optimization resource that uses interactive procedures to simulate the process of evolution of possible solutions populations to a particular problem. The process of evolution is random, but guided by a selection mechanism based on adaptation of individual structures. New structures are generated randomly with a given probability and included in the population. The result tends to be an increase in the adaptation of individuals to the environment and can result in an overall increase in fitness of the population with each new generation.

Evolutionary Algorithm: A computational optimization tool inspired by biological evolution mechanisms, such as reproduction, mutation, recombination, and selection. Potential solutions to the optimization problem play the role of individuals in a population, and it´s evolution is obtained after the repetition of the above operators, according to a fitness function.

Grasshopper®: A visual algorithm editor integrated with Rhinoceros ® software.

Walkability: The extent to which a neighborhood supports walking, in a measurement of how inviting or un-inviting this area is to pedestrians. Density, diversity, design, destination accessibility and distance to transit are key factors that influence and promote it.

Rhinoceros®: A commercial 3D computer graphics and computer-aided design (CAD) application software that focuses on producing mathematically precise representation of curves and freeform surfaces.

Visual Programming Language: Any kind of language that allows coding by manipulating elements graphically, rather than by specifying them textually.

Optimization: The identification of the best solution (with regard to some criteria and from a set of available alternatives) to a given problem.

Galapagos: A Grasshopper ® add-on for application of Evolutionary Algorithms on a wide variety of problems by non-programmers.

Complete Chapter List

Search this Book: