The Alternative Housing Portal
This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA
Overview
Download & View The Alternative Housing Portal as PDF for free.
More details
- Words: 3,423
- Pages: 15
The alternative housing portal
Spatial Analysis
Blok 1, 2007
University of Copenhagen Department of Geography and Geology
Thomas Andersen (exam no. 15) and Sabrina Rothausen (exam no. 1) Supervisor: Thomas Balstrøm
The alternative housing portal
Spatial Analysis 2007
Background Copenhagen is a centre for workplaces occupying over 400.000 persons just in the city and around further 310.000 persons in the environs, which overall is proportional to 26 % of the Danish labour force (Statistikbanken). As other European capitals, the square meter prices in Copenhagen are some of the highest in the country. The average price for houses in Copenhagen is 27.000kr/m2, which is approximately the double of the country average (Realkreditrådet). Consequently, the housing market in the city is very expensive and hard to get in on. Especially for first time buyers and families, it can be extremely difficult to find a house or apartment in Copenhagen with enough space to an affordable price. Within the last 6 years a rising degree of people have been moving away from Copenhagen choosing alternative housing on Sealand and in Malmo (Politiken). Many settle down outside the Copenhagen area to get lower prices, more space, and surrounding nature, while people at the same time often do not want to loose the advantages of living in a city. The biggest challenge is finding a place to live outside the city without loosing easy access to good infrastructure and services. When buying a house/apartment the location is of great importance. At the moment most of the real estate market only provide information about the house but have deficient information on the surrounding area and distance to services and nature. The area the house is located in can be unknown to the buyer wherefore a simple illustration of the facilities of the area is needed. What if a potential house buyer could quickly and easily locate areas matching preferences indicated by the buyer himself?
Objective This paper will illustrate how several factors relevant for choosing a house can be collected in a database and through spatial analysis help find the optimal place to live. The goal is to create an alternative housing portal where people can search for places to live outside Copenhagen. It should in practice work as a homepage available for all. The portal is aimed at people working in Copenhagen and commuting every day. It will provide the tools for locating areas and houses of interest for the buyer. Different factors are specified in the portal and can be valuated by the user, working in such a way that potential house buyers give in their preferences for acceptable distances to the centre of Copenhagen and to other services. The areas complying with the certain specified criteria are then pointed out. Within these areas the houses for sale are shown and various pieces of information about the houses will be given in tables. Next step is then to set up search criteria for the home. The final result also contains information about the distance to different nature types. The
Page of 15
1
The alternative housing portal
Spatial Analysis 2007
output will be presented as a map showing houses for sale where the nearness of nature in the areas of interest can be displayed after request.
Software and Data The software used for this spatial analysis is ESRI’ ArcGIS 9.1 – ArcView and Spatial Analyst extension. The GIS application Model Builder is used for creating the flow of analyzing processes giving the result. Further, Network Analyst has been used for defining test- and service areas. The data used is shown in the Table 1. Most of the data is vector data except the nature layer that are in grid files (raster).
Table 1: Applied data
Data
Feature
Source
Danish Square net - raster
Cells
University of Copenhagen
Road network - vejnet_dk
Lines
University of Copenhagen
CPH_C
Points
Digitized
Stations
Points
University of Copenhagen
Schools
Points
Geomatic aps – CVR register
Shopping
Points
Geomatic aps – CVR register
Homes for sale
Points
Boligtorvet
Nature: woods, ocean, lake,
Polygon
KMS - TOP10DK
beach with rescue equipment
Converted into raster
The road network data, from which the service areas are created, contains information of road type, speed limits, and one-way streets. This makes it possible to calculate the real travel distance by car to the city centre. The housing data shows the location of houses/apartments for sale with attached information about cash price, address, house/ground area, number of rooms, and construction year. This data is acquired from a national housing portal called “Boligtorvet” that among other things provides an overview of all house and apartments for sale in Denmark.
Because of limited time, we have made the assumption that the raw data and the created data inputs are accurate and valid for our model. If we have had the time, the data should have been tested and verified continuously throughout the process of making and running the model.
Page of 15
2
The alternative housing portal
Spatial Analysis 2007
Method The alternative housing portal is meant to be running as a homepage where users can find houses and apartments of interest online. We have therefore chosen primary to work with simple vector data and a delimited test area. Large amounts of raster data require more storage capacity than vector, which could limit the accessibility for users and reduce the speed for getting a result. Especially, more data layers or an expansion of the test area would take up more computer space and thereby increase the expenses for the model (Longley et al. 2005;181-186). Additionally, vector data is good for this type of model because input data is based on merged service areas, which contain a limited amount of polygons. The size of the data is then relatively small, which makes it fast to work with. As mentioned above network analyst is used to identify firstly; the test area, secondly; the service areas. With the help of linear referencing these areas can be pointed out within a specified distance from the chosen origin(s) (Longley et al. 2005;188). Access to nature is represented as layers of raster data. Raster is chosen to give a simpler grid structure e.g. with the usage of scales. Furthermore, it will not slow down the operation of the model because the data is not used as input in the analysis but only for visualizing (Longley et al. 2005;181).
The model for finding alternative areas to live outside Copenhagen is created in Model Builder. With this application it is possible to create, edit, and manage GIS models by specifying input data, the functions to be used, and the output data. It is also possible to have a changing parameter, which is essential in this model because the distance to certain areas is specified by the user (ESRI 2007). The tools used in the model are selection and clip, which are relatively easily operating functions. The selection function extracts features from an input layer and stores them in a new output. The selection in this model is based on a Structured Query Language (SQL) expression. SQL may be used directly via an interactive command line interface (Longley et al. 2005;225), working as a statement or expression that command the database to perform a function. It could for example be a selection of houses over (>) 100 sqm. In this model data manipulating language is mainly used. The clip function cuts out a piece of a feature using one or more of the features as overlay (see figure 1). In this model it is a very useful function because most the output dataset can be preformed as clips of the specified service areas. Moreover, the clip function is a simple procedure which makes it fast. This is important since these functions are meant to be active (should be defined by the users) while offering quick operating.
Page of 15
3
The alternative housing portal
Spatial Analysis 2007
Figure 1: Illustration of the clip function (ESRI 2007)
Analysis The first step was choosing a test area for running our model. We choose only to work with the area of a 70km travel distance from Copenhagen central station to limit the amount of data. This area is also within the most attractive distance to live outside Copenhagen when commuting. The test area was identified through Network Analyst and is shown in Figure 2. Figure 2: Test area
Page of 15
4
The alternative housing portal
Spatial Analysis 2007
The data on infrastructure was prepared for the model. In Network Analyst service areas were created by loading all the points of interest to the application. From these points distance intervals were set and drawn as polygons. The polygons from all the service areas were exported to a new shape-file, which is used in the model. Table 2 describes the type of points and defined intervals of the points of interest. Figure 3 shows an example of a service area in this case shopping. All the supermarkets in the test area are presented as points. Every point has 1,2,3,4,5,6,7,8,9,10 km service areas, shown as lines creating polygons, around it. If any of the service area overlaps, the polygons are merged together.
Table 2: Service areas created in Network Analyst
Point of interest
Type
Intervals
CPH_C
Central Station
5,10,20,30,40,50,60,70 km
Stations
IC-, Re-, S-, and Metro-train
1,2,3,4,5,6,7,8,9,10 km
Schools
Elementary
1,2,3,4,5,6,7,8,9,10 km
Shopping
Supermarket
1,2,3,4,5,6,7,8,9,10 km
Figure 3: Example of a service area (shopping)
Page of 15
5
The alternative housing portal
Spatial Analysis 2007
Data on houses and apartments for sale on Sealand was lent to us by Boligtorvet; this data contains the basic real estate information such as address, price, number of rooms etc. For loading this file into GIS we had to geo-code the addresses in X-point (program available at Geomatic aps) for assigning them with X and Y position. Once the geo-coding was executed the tabular file was loaded into ArcMap with the help of the add XY data tool in ArcMap. Finally, the file was exported to a shape-file.
Model 1 – Locate homes in areas of interest The next step was to create a model that could locate homes for sale in the area of a given interest. The service areas were loaded to Model Builder and the select and clip functions were applied. Annex 1 shows the complete model as a flowchart. The preferred maximum distance to the different services are then meant to be specified by the user. In the flowchart parameters that can be applied by the user are shown in a light blue colour. The service layers are continuously clipped together and the output will be houses/apartments within the areas of interest.
Figure 4: Example of selection in Model 1
Figure 4 shows an example of the selection in Model 1. The distance to areas of interest (km) can be specified in the fields. The expressions are simple SQL statements that select areas with a maximum distance to Copenhagen Central, train stations, schools and shopping. If the distance to one of the areas does not matter, the statement can be deleted and will not have any influence on the output. When the distances have been specified, the model will run the selection and the clips. The last clip is in the layer with all the houses and apartments for sale. The final output is the areas of interest, wherein the houses and apartments for sale are shown as points.
Page of 15
6
The alternative housing portal
Spatial Analysis 2007
Model 2 – Specify criteria for the home This model will help the user in closing in on the houses of interest by specifying certain factors about the home. The model is exclusively using the select function with the parameters listed below •
Price: max kr
•
Size: min sqm
•
Rooms: min number
•
Age of house: max age (calculated from construction/restoration year)
Annex 2 shows the flowchart of model 2. After executing the model only houses/apartments that apply the choices/wishes of the user in the areas of interest will be shown.
Figure 5: Example of selection in Model 2
Figure 5 shows an example of selection in Model 2. The parameters of importance for finding a house or apartment, in the area of interest, can be specified in model 2. When all the parameters are specified, the model selects and shows all the houses and apartments within the specified criteria in the area of interest.
Page of 15
7
The alternative housing portal
Spatial Analysis 2007
The nature factors The distance to nature can be displayed as background on the output. Four nature types can be showed – sea, lake, forest and beach with rescue equipment. These nature types will appear as raster layers showing the distance in the form of a colour code. Low distances are coloured red and orange, where a long distance to nature will appear blue on the map (a hot-cold scale). A legend according to the nature types will indicate the scale of the colouring. Figure 6 demonstrates the test area with the beach layer displayed. The layers can be clicked on and off after desire of the user. This makes it straightforward for users to divide between nature types in case they e.g. prioritise nearness of forest above other nature types.
Figure 6: Distance to different nature types
The total search system with both models is attached as a CD-Rom and Annex 3 is a guide in how to run the search.
Page of 15
8
The alternative housing portal
Spatial Analysis 2007
Results The purpose of the alternative housing portal is to gather all factors that could influence on the choice of place to live, on a web-based portal. ArcMap should be seen as an Internet browser where the homepage of the housing portal is open to all. The search options are illustrated by the models in ArcToolbox and the layers in the left part of the figure are the nature types, which users can click on and off to see the distance to certain nature themes. The map to the right visualizes the results of the search using the models. Figure 7 shows an example of a search on the “webpage”. The distance parameters input for this specific search was: •
CPH C <= 10 km
•
Schools <= 3 km
•
Stations <= 1 km
•
Shopping <= 1 km
Figure 7: Example of a search on the alternative housing portal (Model 1)
The figure shows the output of the search where the area of interest is the brown layer and the houses/apartments of interest are the points. There are 3695 houses and apartments within these specific search criteria. Some of these homes are shown in the attribute table (Figure 8).
Page of 15
9
The alternative housing portal
Spatial Analysis 2007
Figure 8: Attribute table of a specific search
Figure 9: Example of a specific search on the alternative housing portal (Model 2)
Page of 15
10
The alternative housing portal
Spatial Analysis 2007
For reducing the number of houses/apartments the user can specific criteria for these by using the second model. An example on a search is shown in Figure 9. The input parameters here were: •
Price <= 2 million kr
•
Size >= 70 sqm
•
Number of rooms >= 3
•
Age of house – expression was deleted because it did not matter
The output of this more specific search is 85 houses and apartments. So by running Model 2, the number decreases considerably and helps the users to find the right house/apartment in areas of interest.
Discussion The alternative housing portal offers a new approach in the search for a place to live. Here several factors are gathered in one database. Infrastructure, services in the area, the home and nearness of nature are all factors that influence on the choice of place to live. By joining all these variables in a single portal the search for a new house/apartment can be more efficient and focused. At the same time the search builds on an individual valuation of the different factors, which shapes the result after the users own personal preferences. In the design of the models and the layout output, we have considered the importance of a simple, well-arranged application that is easy and fast operating. This is significant if the portal should work as a homepage. On top of that it is uncomplicated to update the data for the home to current houses for sale.
There are ways to improve or expand the portal. The selected nature types do of course not represent all types of nature, but gives a good picture of the nearness of the most common nature types. Landscape topography and the slope of the site are also factors that would influence on the amenity value of an area. South-facing gardens or a great view over the surroundings could increase the amenity value considerably. Although, amenity value has great importance, a representation of it would demand great amounts of data and processing of this. A way to do it in a housing portal like this could be as a weighted average in raster cells between the different natural amenities. This would be very difficult to calculate because; firstly, there are many factors to be considered. Secondly, it is hard to make an objective valuation of subjective things such as forest or a view. Thirdly, when we tested the weighted average function on our data it seemed to run slow, so if the
Page of 15
11
The alternative housing portal
Spatial Analysis 2007
function should run online it had to be optimized quite a bit. However, the amenity value has a huge impact on the price of a house.
Beside the nature factors that can have a positive effect on the house prices, there are also features that can have a negative effect. These are environmental issues such as; •
soil and air pollution
•
noise from big roads, windmills or airports
•
industry
•
waste disposal site or waste water treatment plant
Alternatively, a weighted average of the amenity value should include these negative factors. This would complicate the calculation and data processing further, but could be of great importance for the buyers.
At other housing portals demographic factors for the neighbourhood area are shown when searching for houses. Here data concerning the age distribution or the average income or education levels and other demographic factors are available for the user to study. These factors could be displayed as raster layer similar to the nature factors. The frequency of crime and types of crime in an area are also factors that might be interesting to consider displaying. On this type of portal where travel distance and distance to services is of great importance the most obvious thing to incorporate is travel time, both with personal and public transportation. Users could then specify a more precise timeframe in which they would be willing to travel per day. At the homepage www.rejseplanen.dk users can calculate travel time for travelling with public transportation when the addresses from and to are given. If this system with specific addresses and public transport could be incorporated, it would be a big asset for the model. The model could then give more exact distances (address to address) which would have a wider focus than on the people with a car.
Page of 15
12
The alternative housing portal
Spatial Analysis 2007
Conclusion In this project we have created an alternative housing portal for people looking for a place to live within the range of Copenhagen. We have managed to collect various important factors influencing on the choice of home. The models have shown to be well functioning and fast in the process of helping people find houses/apartments in areas of their interest. Therefore, we can conclude that Model Builder with the right data can be a very useful tool for creating a model (portal) for people searching for a new home. Furthermore, we can conclude that the model is dependent on the amount of data available. The more data available, the more specific and aimed toward the individual user, the output can be.
Page of 15
13
The alternative housing portal
Spatial Analysis 2007
References ¾ Longley, P.A. et al (2005): Geographic Information Systems and Science. 2nd Ed., Wiley. ¾ ESRI (2007): ArcGIS Desktop Help http://webhelp.esri.com/arcgisdesktop/9.1/index.cfm?TopicName=welcome ¾ Politiken: Middelklassen flygter fra København (4.4.2007) http://politiken.dk/erhverv/article278763.ece ¾ Realkreditrådet: www.realkreditraadet.dk ¾ Statistikbanken: www.statistikbanken.dk
Data ¾ Boligtorvet ¾ Geomatic aps
Page of 15
14
Spatial Analysis
Blok 1, 2007
University of Copenhagen Department of Geography and Geology
Thomas Andersen (exam no. 15) and Sabrina Rothausen (exam no. 1) Supervisor: Thomas Balstrøm
The alternative housing portal
Spatial Analysis 2007
Background Copenhagen is a centre for workplaces occupying over 400.000 persons just in the city and around further 310.000 persons in the environs, which overall is proportional to 26 % of the Danish labour force (Statistikbanken). As other European capitals, the square meter prices in Copenhagen are some of the highest in the country. The average price for houses in Copenhagen is 27.000kr/m2, which is approximately the double of the country average (Realkreditrådet). Consequently, the housing market in the city is very expensive and hard to get in on. Especially for first time buyers and families, it can be extremely difficult to find a house or apartment in Copenhagen with enough space to an affordable price. Within the last 6 years a rising degree of people have been moving away from Copenhagen choosing alternative housing on Sealand and in Malmo (Politiken). Many settle down outside the Copenhagen area to get lower prices, more space, and surrounding nature, while people at the same time often do not want to loose the advantages of living in a city. The biggest challenge is finding a place to live outside the city without loosing easy access to good infrastructure and services. When buying a house/apartment the location is of great importance. At the moment most of the real estate market only provide information about the house but have deficient information on the surrounding area and distance to services and nature. The area the house is located in can be unknown to the buyer wherefore a simple illustration of the facilities of the area is needed. What if a potential house buyer could quickly and easily locate areas matching preferences indicated by the buyer himself?
Objective This paper will illustrate how several factors relevant for choosing a house can be collected in a database and through spatial analysis help find the optimal place to live. The goal is to create an alternative housing portal where people can search for places to live outside Copenhagen. It should in practice work as a homepage available for all. The portal is aimed at people working in Copenhagen and commuting every day. It will provide the tools for locating areas and houses of interest for the buyer. Different factors are specified in the portal and can be valuated by the user, working in such a way that potential house buyers give in their preferences for acceptable distances to the centre of Copenhagen and to other services. The areas complying with the certain specified criteria are then pointed out. Within these areas the houses for sale are shown and various pieces of information about the houses will be given in tables. Next step is then to set up search criteria for the home. The final result also contains information about the distance to different nature types. The
Page of 15
1
The alternative housing portal
Spatial Analysis 2007
output will be presented as a map showing houses for sale where the nearness of nature in the areas of interest can be displayed after request.
Software and Data The software used for this spatial analysis is ESRI’ ArcGIS 9.1 – ArcView and Spatial Analyst extension. The GIS application Model Builder is used for creating the flow of analyzing processes giving the result. Further, Network Analyst has been used for defining test- and service areas. The data used is shown in the Table 1. Most of the data is vector data except the nature layer that are in grid files (raster).
Table 1: Applied data
Data
Feature
Source
Danish Square net - raster
Cells
University of Copenhagen
Road network - vejnet_dk
Lines
University of Copenhagen
CPH_C
Points
Digitized
Stations
Points
University of Copenhagen
Schools
Points
Geomatic aps – CVR register
Shopping
Points
Geomatic aps – CVR register
Homes for sale
Points
Boligtorvet
Nature: woods, ocean, lake,
Polygon
KMS - TOP10DK
beach with rescue equipment
Converted into raster
The road network data, from which the service areas are created, contains information of road type, speed limits, and one-way streets. This makes it possible to calculate the real travel distance by car to the city centre. The housing data shows the location of houses/apartments for sale with attached information about cash price, address, house/ground area, number of rooms, and construction year. This data is acquired from a national housing portal called “Boligtorvet” that among other things provides an overview of all house and apartments for sale in Denmark.
Because of limited time, we have made the assumption that the raw data and the created data inputs are accurate and valid for our model. If we have had the time, the data should have been tested and verified continuously throughout the process of making and running the model.
Page of 15
2
The alternative housing portal
Spatial Analysis 2007
Method The alternative housing portal is meant to be running as a homepage where users can find houses and apartments of interest online. We have therefore chosen primary to work with simple vector data and a delimited test area. Large amounts of raster data require more storage capacity than vector, which could limit the accessibility for users and reduce the speed for getting a result. Especially, more data layers or an expansion of the test area would take up more computer space and thereby increase the expenses for the model (Longley et al. 2005;181-186). Additionally, vector data is good for this type of model because input data is based on merged service areas, which contain a limited amount of polygons. The size of the data is then relatively small, which makes it fast to work with. As mentioned above network analyst is used to identify firstly; the test area, secondly; the service areas. With the help of linear referencing these areas can be pointed out within a specified distance from the chosen origin(s) (Longley et al. 2005;188). Access to nature is represented as layers of raster data. Raster is chosen to give a simpler grid structure e.g. with the usage of scales. Furthermore, it will not slow down the operation of the model because the data is not used as input in the analysis but only for visualizing (Longley et al. 2005;181).
The model for finding alternative areas to live outside Copenhagen is created in Model Builder. With this application it is possible to create, edit, and manage GIS models by specifying input data, the functions to be used, and the output data. It is also possible to have a changing parameter, which is essential in this model because the distance to certain areas is specified by the user (ESRI 2007). The tools used in the model are selection and clip, which are relatively easily operating functions. The selection function extracts features from an input layer and stores them in a new output. The selection in this model is based on a Structured Query Language (SQL) expression. SQL may be used directly via an interactive command line interface (Longley et al. 2005;225), working as a statement or expression that command the database to perform a function. It could for example be a selection of houses over (>) 100 sqm. In this model data manipulating language is mainly used. The clip function cuts out a piece of a feature using one or more of the features as overlay (see figure 1). In this model it is a very useful function because most the output dataset can be preformed as clips of the specified service areas. Moreover, the clip function is a simple procedure which makes it fast. This is important since these functions are meant to be active (should be defined by the users) while offering quick operating.
Page of 15
3
The alternative housing portal
Spatial Analysis 2007
Figure 1: Illustration of the clip function (ESRI 2007)
Analysis The first step was choosing a test area for running our model. We choose only to work with the area of a 70km travel distance from Copenhagen central station to limit the amount of data. This area is also within the most attractive distance to live outside Copenhagen when commuting. The test area was identified through Network Analyst and is shown in Figure 2. Figure 2: Test area
Page of 15
4
The alternative housing portal
Spatial Analysis 2007
The data on infrastructure was prepared for the model. In Network Analyst service areas were created by loading all the points of interest to the application. From these points distance intervals were set and drawn as polygons. The polygons from all the service areas were exported to a new shape-file, which is used in the model. Table 2 describes the type of points and defined intervals of the points of interest. Figure 3 shows an example of a service area in this case shopping. All the supermarkets in the test area are presented as points. Every point has 1,2,3,4,5,6,7,8,9,10 km service areas, shown as lines creating polygons, around it. If any of the service area overlaps, the polygons are merged together.
Table 2: Service areas created in Network Analyst
Point of interest
Type
Intervals
CPH_C
Central Station
5,10,20,30,40,50,60,70 km
Stations
IC-, Re-, S-, and Metro-train
1,2,3,4,5,6,7,8,9,10 km
Schools
Elementary
1,2,3,4,5,6,7,8,9,10 km
Shopping
Supermarket
1,2,3,4,5,6,7,8,9,10 km
Figure 3: Example of a service area (shopping)
Page of 15
5
The alternative housing portal
Spatial Analysis 2007
Data on houses and apartments for sale on Sealand was lent to us by Boligtorvet; this data contains the basic real estate information such as address, price, number of rooms etc. For loading this file into GIS we had to geo-code the addresses in X-point (program available at Geomatic aps) for assigning them with X and Y position. Once the geo-coding was executed the tabular file was loaded into ArcMap with the help of the add XY data tool in ArcMap. Finally, the file was exported to a shape-file.
Model 1 – Locate homes in areas of interest The next step was to create a model that could locate homes for sale in the area of a given interest. The service areas were loaded to Model Builder and the select and clip functions were applied. Annex 1 shows the complete model as a flowchart. The preferred maximum distance to the different services are then meant to be specified by the user. In the flowchart parameters that can be applied by the user are shown in a light blue colour. The service layers are continuously clipped together and the output will be houses/apartments within the areas of interest.
Figure 4: Example of selection in Model 1
Figure 4 shows an example of the selection in Model 1. The distance to areas of interest (km) can be specified in the fields. The expressions are simple SQL statements that select areas with a maximum distance to Copenhagen Central, train stations, schools and shopping. If the distance to one of the areas does not matter, the statement can be deleted and will not have any influence on the output. When the distances have been specified, the model will run the selection and the clips. The last clip is in the layer with all the houses and apartments for sale. The final output is the areas of interest, wherein the houses and apartments for sale are shown as points.
Page of 15
6
The alternative housing portal
Spatial Analysis 2007
Model 2 – Specify criteria for the home This model will help the user in closing in on the houses of interest by specifying certain factors about the home. The model is exclusively using the select function with the parameters listed below •
Price: max kr
•
Size: min sqm
•
Rooms: min number
•
Age of house: max age (calculated from construction/restoration year)
Annex 2 shows the flowchart of model 2. After executing the model only houses/apartments that apply the choices/wishes of the user in the areas of interest will be shown.
Figure 5: Example of selection in Model 2
Figure 5 shows an example of selection in Model 2. The parameters of importance for finding a house or apartment, in the area of interest, can be specified in model 2. When all the parameters are specified, the model selects and shows all the houses and apartments within the specified criteria in the area of interest.
Page of 15
7
The alternative housing portal
Spatial Analysis 2007
The nature factors The distance to nature can be displayed as background on the output. Four nature types can be showed – sea, lake, forest and beach with rescue equipment. These nature types will appear as raster layers showing the distance in the form of a colour code. Low distances are coloured red and orange, where a long distance to nature will appear blue on the map (a hot-cold scale). A legend according to the nature types will indicate the scale of the colouring. Figure 6 demonstrates the test area with the beach layer displayed. The layers can be clicked on and off after desire of the user. This makes it straightforward for users to divide between nature types in case they e.g. prioritise nearness of forest above other nature types.
Figure 6: Distance to different nature types
The total search system with both models is attached as a CD-Rom and Annex 3 is a guide in how to run the search.
Page of 15
8
The alternative housing portal
Spatial Analysis 2007
Results The purpose of the alternative housing portal is to gather all factors that could influence on the choice of place to live, on a web-based portal. ArcMap should be seen as an Internet browser where the homepage of the housing portal is open to all. The search options are illustrated by the models in ArcToolbox and the layers in the left part of the figure are the nature types, which users can click on and off to see the distance to certain nature themes. The map to the right visualizes the results of the search using the models. Figure 7 shows an example of a search on the “webpage”. The distance parameters input for this specific search was: •
CPH C <= 10 km
•
Schools <= 3 km
•
Stations <= 1 km
•
Shopping <= 1 km
Figure 7: Example of a search on the alternative housing portal (Model 1)
The figure shows the output of the search where the area of interest is the brown layer and the houses/apartments of interest are the points. There are 3695 houses and apartments within these specific search criteria. Some of these homes are shown in the attribute table (Figure 8).
Page of 15
9
The alternative housing portal
Spatial Analysis 2007
Figure 8: Attribute table of a specific search
Figure 9: Example of a specific search on the alternative housing portal (Model 2)
Page of 15
10
The alternative housing portal
Spatial Analysis 2007
For reducing the number of houses/apartments the user can specific criteria for these by using the second model. An example on a search is shown in Figure 9. The input parameters here were: •
Price <= 2 million kr
•
Size >= 70 sqm
•
Number of rooms >= 3
•
Age of house – expression was deleted because it did not matter
The output of this more specific search is 85 houses and apartments. So by running Model 2, the number decreases considerably and helps the users to find the right house/apartment in areas of interest.
Discussion The alternative housing portal offers a new approach in the search for a place to live. Here several factors are gathered in one database. Infrastructure, services in the area, the home and nearness of nature are all factors that influence on the choice of place to live. By joining all these variables in a single portal the search for a new house/apartment can be more efficient and focused. At the same time the search builds on an individual valuation of the different factors, which shapes the result after the users own personal preferences. In the design of the models and the layout output, we have considered the importance of a simple, well-arranged application that is easy and fast operating. This is significant if the portal should work as a homepage. On top of that it is uncomplicated to update the data for the home to current houses for sale.
There are ways to improve or expand the portal. The selected nature types do of course not represent all types of nature, but gives a good picture of the nearness of the most common nature types. Landscape topography and the slope of the site are also factors that would influence on the amenity value of an area. South-facing gardens or a great view over the surroundings could increase the amenity value considerably. Although, amenity value has great importance, a representation of it would demand great amounts of data and processing of this. A way to do it in a housing portal like this could be as a weighted average in raster cells between the different natural amenities. This would be very difficult to calculate because; firstly, there are many factors to be considered. Secondly, it is hard to make an objective valuation of subjective things such as forest or a view. Thirdly, when we tested the weighted average function on our data it seemed to run slow, so if the
Page of 15
11
The alternative housing portal
Spatial Analysis 2007
function should run online it had to be optimized quite a bit. However, the amenity value has a huge impact on the price of a house.
Beside the nature factors that can have a positive effect on the house prices, there are also features that can have a negative effect. These are environmental issues such as; •
soil and air pollution
•
noise from big roads, windmills or airports
•
industry
•
waste disposal site or waste water treatment plant
Alternatively, a weighted average of the amenity value should include these negative factors. This would complicate the calculation and data processing further, but could be of great importance for the buyers.
At other housing portals demographic factors for the neighbourhood area are shown when searching for houses. Here data concerning the age distribution or the average income or education levels and other demographic factors are available for the user to study. These factors could be displayed as raster layer similar to the nature factors. The frequency of crime and types of crime in an area are also factors that might be interesting to consider displaying. On this type of portal where travel distance and distance to services is of great importance the most obvious thing to incorporate is travel time, both with personal and public transportation. Users could then specify a more precise timeframe in which they would be willing to travel per day. At the homepage www.rejseplanen.dk users can calculate travel time for travelling with public transportation when the addresses from and to are given. If this system with specific addresses and public transport could be incorporated, it would be a big asset for the model. The model could then give more exact distances (address to address) which would have a wider focus than on the people with a car.
Page of 15
12
The alternative housing portal
Spatial Analysis 2007
Conclusion In this project we have created an alternative housing portal for people looking for a place to live within the range of Copenhagen. We have managed to collect various important factors influencing on the choice of home. The models have shown to be well functioning and fast in the process of helping people find houses/apartments in areas of their interest. Therefore, we can conclude that Model Builder with the right data can be a very useful tool for creating a model (portal) for people searching for a new home. Furthermore, we can conclude that the model is dependent on the amount of data available. The more data available, the more specific and aimed toward the individual user, the output can be.
Page of 15
13
The alternative housing portal
Spatial Analysis 2007
References ¾ Longley, P.A. et al (2005): Geographic Information Systems and Science. 2nd Ed., Wiley. ¾ ESRI (2007): ArcGIS Desktop Help http://webhelp.esri.com/arcgisdesktop/9.1/index.cfm?TopicName=welcome ¾ Politiken: Middelklassen flygter fra København (4.4.2007) http://politiken.dk/erhverv/article278763.ece ¾ Realkreditrådet: www.realkreditraadet.dk ¾ Statistikbanken: www.statistikbanken.dk
Data ¾ Boligtorvet ¾ Geomatic aps
Page of 15
14
Related Documents
The Alternative Housing Portal
December 2019 14
Portal
October 2019 54
Portal
November 2019 54
Portal
May 2020 39