Genetic Algorithms: Stages of the Study in the Expert System

Genetic Algorithms: Stages of the Study in the Expert System

DOI: 10.4018/978-1-5225-8958-7.ch004

Abstract

This chapter includes the second phase of the Re-Coding Homes Project, which has been conducted as a TUBITAK (The Scientific and Technological Research Council of Turkey) research project with the title “A User-Centered Model Research Towards a Flexible Interior Spatial Design for Mass Housing Units: Urban Renewal Housing.” In this phase, which will be presented in this chapter, the design model and an expert system have been developed with the aim of improving the interior spaces of mass housing projects in accordance with different user needs and providing solutions that will increase the flexibility of apartment interiors. The expert system that can operate the flexible modular system proposed in the housing units has made it possible to achieve a large number of spatial variations by means of “multi-parameter layout design.” “Mass-customization” approach was used in order to generate satisfactory results for users' spatial needs and life styles.
Chapter Preview
Top

Background

Interior modularity and modular systems have been studied for different typologies of spaces for many years. Ericsson and Erixon (1999) defines the modular design approach as creating a diversified set of products that is shaped by a strategic and flexible mentality, rather than looking for the optimal product that leads to the optimal solution. In this case, new products that respond to changing needs can be developed and these solutions can be integrated into the system without breakdown of the whole. In architecture or construction fields, a well-thought-out modular system will increase flexibility by reducing the negative impact of changing needs and facilitate the management of the system (Ericsson and Erixon, 1999).

Modular design allows a plurality of different products to come together in different combinations within the framework of repetitive units and hierarchical relationships of these units. And this can be considered as a situation that increases the potential for responding to flexibility and change. Also according to Baldwin and Clark (2000), a modular system is open and flexible to develop at the module level. In this sense, the modularity can respond to new needs through the exchange of necessary parts of the system without changing the whole system. Modularity also aims to increase efficiency or value of use by reducing complexity. Product structure plays a key role in managing complexity. In this sense, it can be said that the modular design is an approach which supports flexibility in the product basis.

In this work, while creating the design model, the use of modular design principles has been effective in the sense of avoiding the complexity in creating spatial combinations and obtaining product variety that meets different usages. An expert system has been created that provides alternative interior space layouts for different user types and usages by using the method Genetic Algorithms.

Genetic Algorithms have been used for optimization of design, spatial arrangement and architectural form research in the field of design. Genetic Algorithms, which is a generative design approach, is used to solve non-linear design problems. This approach has been inspired by the evolutionary process of nature and it simulates long-term natural selection in a short time within a computer environment. Designs evolve in computer environment according to the fitness function determined by the designer. Design alternatives are obtained after its evolvement due to the target function (Gu et al., 2010).

The objectives of the genetic algorithms can be summarized as follows (Rosenman, 1997; Carranza, 2005):

Key Terms in this Chapter

MOGA: Multi-objective genetic algorithms.

Spatial Potentials: Features and facts that can be interpreted to increase the performance and quality of space.

User-Centered Design: Design that evaluates user needs and preferences at all stages of the design process.

Mass Production (MP): Producing standart and big amont of products for the clients.

Modular Design: Design based on interchangeable components in order to increase flexibility.

Multi-Parameter Layout Generation (MPLG): A design approach that is used to solve complex and multicriteria design problems.

Design Parameter: A variable that is evaluated during the design process determining the characteristics of the final object or system.

Expert System: A computer program that uses artificial intelligence (AI) technologies to simulate the judgment and behavior of a human or an organization.

Genetic Algorithms: A computational and generative design approach which is used to solve non-lineer design problems.

Flexible Interior Design: Interior design that can be used for different activities and purposes at different times.

Mass-Customization (MC): Producing products that are customized due to the clients’ requests, characteristics, etc.

Complete Chapter List

Search this Book:
Reset