Page 57 - Çevre Şehir İklim İngilizce - Sayı 4

Hacı Abdullah Uçan - Tayfun Dede - Sinan Nacar

Introduction

In this study, in order to make inferences and forecasts for the future urban
transformation strategy of our country, economic data such as the amount of
construction performed in 2002-2020 period, the amount of square meters
of construction permitted and the quantity of manufacturing, gross national
product, national income per capita, inflation figures, real interest figures
based on loan usage etc. were used. Considering that the economy is the most
important factor affecting the speed of construction and urban transformation
along with the legal regulations in the country, in the light of these economic
periodic data, it is aimed to predict the future of Türkiye’s urban transformation
process by analyzing historical data using algorithms based on artificial neural
networks (ANN). With the findings obtained, it will be possible to model the
urban transformation process and the future of urban transformation in the
country by taking advantage of the data and past experiences related to
urban transformation and construction sector in our country until this process
is completed and to help determine policies accordingly. As an independent
variable in the study, social and economic indicators that directly or indirectly
affect urban transformation practices were used as inputs, provided that they
belong to a certain period. By comparing the data obtained as outputs with the
processing of these inputs in the software to be used in the study, the accuracy
of the prediction models achieved by the performed process was confirmed.
The obtained forecast models can be used to calculate the specific social and
economic cost of the urban transformation application that is planned to be
carried out in the future. The prediction models of the urban transformation
process were examined with traditional regression-based and ANN-based
modeling. In order to get more effective results from ANN-based modeling, an
up-to-date an popular algorithm, the Gray Wolf Algorithm, has been included.

Literature Review

Many aspects are taken into consideration when determining the urban
transformation implementation strategy in the world. Factors such as
population, economic situation, education, cultural structure affect the urban
transformation process. Some scientific studies related to the modeling of
urban transformation planning using ANN information processing technology
are expressed below.
An urban simulation was performed by Yeh and Li (2003) using ANN cellular
automation for land use planning, and urban planning was presented based
on geographical information systems according to the results. This simulation
contributes to decision makers and planners on smart and sustainable urban

46 The Journal of Environment, Urban and Climate