Page 668 - NGTU_paper_withoutVideo

Modern Geomatics Technologies and Applications

2. Materials and methods
2.1. Study area
Regarding malaria statistics, three provinces of Sistan and Baluchestan, Kerman, and Hormozgan have been introduced
as the most susceptible areas for malaria prevalence during the last ten years [15]. The county of Saravan was selected as the
study area due to being one of the malaria-endemic regions in Iran. In this county, about 654 cases of malaria were reported
during the years 2011-2016. The geographical location of Saravan county was shown in Fig.1.

Fig. 1. Study area.

2.2. Data and pre-processing of them
In order to investigate the effects of the aforementioned environmental factors, the environment (i.e., a network of cells
[16]) of our agent-based model was made up of a variety of data as well as satellite images. In that way, the meteorological data
including the monthly average of air temperature as well as relative humidity at all twelve synoptic stations of Sistan and
Baluchestan province were gathered and then interpolated for each month of the year 2013. The normalized difference vegetation
index (NDVI) was calculated in ArcGIS 10.3 software using a combination of Near-Infrared (i.e., Band 5) and Red bands (i.e.,
Band 4) of Landsat 8 satellite images [17] for each month of the year 2013. The Shuttle Radar Topography Mission (SRTM)
data was used to apply as well as investigate the effect of the digital elevation model (DEM). Besides, to investigate the impact
of distance from water sources, permanent rivers in Saravan county were considered as the only available water sources for
mosquitoes to lay eggs. The Euclidean distance analysis was performed on the permanent rivers of Saravan county. The data of
Saravan county’s boundary, the boundary of villages and cities of Saravan along with their population, as well as statistical data
on the malaria cases in 2013 were also used to pursue the aim of this study. All these data were processed and converted into
ASCII format in the ArcGIS 10.3 software.

2.3. The proposed agent-based model
2.3.1. The implementation of the proposed agent-based model: To implement our agent-based model, the NetLogo 6.0.4
platform was used [18]. The implementation process of the proposed model was performed in three main steps including (i)
creating the environment of the agent-based model, (ii) creating agents and distributing them, and (iii) modelling the movements
as well as interactions of the agents.
The first step of implementation whose objective is creating the environment of the agent-based model, was performed
in two phases as follows:

2