A neural network based urban growth model of an Indian city

Abstract

The aim of the study reported in this paper is to demonstrate that the subjectivity in urban growth modeling and the calibration time can be reduced by using objective techniques like Artificial neural network (ANN). As a case study, the ANNbased model was applied to simulate the urban growth of Saharanpur city in India. In the proposed model, remote sensing and GIS were used to generate site attributes, while ANN was used to reveal the relationships between urban growth potential and the site attributes. Once ANN learnt the relationship, it was then used to simulate the urban growth. Different

Neural networks-based simulation of land cover scenarios in Doon valley, India

“…Maithani sites and orchard) and achieved an overall accuracy of 83%. Similar work on ANNbased urban growth modelling was also reported by Pijanowski et al (2002aPijanowski et al ( , 2002bPijanowski et al ( , 2005, Thapa and Murayama (2012), Arora et al (2004), Maithani (2009) and Maithani et al (2010).…”

Simulation of land use changes for the planning of a metropolitan area

“…Yeh and Li [7] had identified seven spatial variables to represent the site attributes of each cell for the simulation of urban development viz., (i) distance to the major urban areas; (ii) distance to suburban areas; (iii) distance to the nearest road; (iv) distance to the nearest expressway; (v) distance to the nearest railways; (vi) neighbourhood development level; and (vii) agricultural suitability. Maithani [8] had identified that urban growth is defined as a function of the following three factors, also called as causative factors which are (i) accessibility to roads (connectivity is a major factor affecting the urban growth process); (ii) accessibility to the city core, (most of the higher level facilities are located in the city core); and (iii) accessibility to infrastructural facilities. Almeida et al [9] had used an extensive list of twelve variables for determining urban land use change through simulation viz., (i) area served by water supply; (ii) mediumhigh density of occupation (25% to 40%); (iii) existence of social housing, (iv) distances to ranges of commercial concentration; (v) distances to industrial zones; (vi) distances to residential zones; (vii) distances to peripheral residential settlements; (viii) distances to isolated institutional use; (ix) distances to main existent roads; (x) distances to the service and industrial axes; (xi) distances to planned roads; and (xii) distances to peripheral roads.…”

Modelling urban growth: Integrating both physical and human dimensions in developing world context

“…Substantial amount of studies have adopted these theories to explain the complex system being represented in the model (Batty and Xie, 1994;Li and Yeh, 2002a;Maithani, 2009;White and Engelen, 1993). The incorporation of neural networks, cellular automata and artificial intelligence implicitly implements these theories in urban studies (Manson, 2001).…”

“…China, between 2008 and 2012, an increasing interest in urban modelling using CA has been observed in other Asian countries, notably Japan, Korea, Malaysia and Nepal (Guan et al, 2011b;Kim, 2009;Maithani, 2009;Naimah et al, 2011;Samat, 2006;Thapa and Murayama, 2011). Applications of CA in USA, Europe, or China reflect more on the leading research clusters rather than the urban growth challenges faced by major cities in the world.…”

“…For this task, in addition with assessment against land cover map of JMA in 2005 by DPU (2012), the accuracy assessment was also done by comparing the retrieved land cover maps with images from Google Earth. We used the most common assessment procedure which is the pixel-by-pixel validation using confusion matrix (Maithani, 2009). …”