GEMO90008 - Foundations of Spatial Information  - ASSIGNMENT 2- MODULE 3 -  Solved

ROAD NETWORK ANALYSIS IN NORTHERN VICTORIA 

1.     OBJECTIVE 

To obtain an understanding of network analysis by working through the road network dataset for Northern Victoria.

2.     INTRODUCTION 

Road networks are the key form of transport in Northern Victoria, a region characterized by extensive irrigation-based agriculture, small rural populations and large distances between towns. To help plan for future government services in Northern Victoria, a useful tool is the network analysis function now available in many GIS. In this exercise, you will use road network analysis to:

1)      Calculate optimal road routes;

2)      Find the closest facilities to nominated locations; 3) Calculate the facility service areas.

3.     DATA REQUIRED 

In this exercise, you will use road network data supplied by Spatial Information Infrastructure Victoria (the State government agency tasked with providing key land-related digital datasets for Victoria). You will require:

• zipped data from this directory – contains towns.shp, roads.shp, sites.shp and their corresponding files. The data can be found in LMS (NetworkAnalysis-data.zip).

4.     PROCEDURE 

4.1 Getting started  
Ensure that the Network Analyst extension is switched on. There are two steps to this: File > Licensing > Esir Extensions.

                                                                                                                                                                                                       1 

4.2 Understanding the content of your data  Let's examine the data themes more closely.  

Add the data (roads.shp, sites.shp and towns.shp) to your ArcGIS session using the ‘Add data’ button which you learnt about in the previous tutorial.

•      Display the town's theme and right-click on towns layer and go to Symbology. Display single symbols for each town in the form of a blue star of size 16.  

•      Right-click on towns layer and select ‘Label’ to show labels for the towns based on the locality field (label field can be found by right click and select label properties). Alternatively, you could click on the layer and use the Labelling ribbon.

•      Make the theme of the road active, display it, and use the ‘Symbology’ to display unique values for the different road classes occurring in the type field. (Hint: Unique Values Value field = Type and click Add All Values). Edit the symbol colours so that main roads are red and size 2 (thick lines), regional roads are dark brown, streets are red, local roads are mid-brown, farm roads are green, and tracks are black.  

 

 

NOTE: The Eastern edge of the dataset indicates the approximate position of the Murray River – the border between New South Wales and Victoria.  

4.3 Building a network dataset
Q1. The data structure in ArcGIS Pro: Can you tell, after this exercise, the relations among 1) geodatabase; 2) feature datasets; 3) feature classes?

 

•      Open Catalog pane – recall that Catalog takes charge of creating a dataset. First of all, you need to create a geodatabase to save the road network. You can only build a network dataset in a geodatabase (see appendix for details).  

•      Create a feature dataset (this is a collection of related feature classes you would like to work on, so one database can contain multiple datasets, and one dataset contain multiple feature classes) in the database by right-clicking the database > new feature dataset

•       Import the road shapefile (which is a stand-alone file yet in a database) to the dataset in the targeted geodatabase by right-clicking the dataset> import file (road).  

•      Then create the network dataset (this is literally a “copy” of the imported road feature class but include more advanced important topological and other relevant analysis information specified by you and generated by the software).  

 

 

 

 

4.3.1 Creating network dataset 

•      Search for Create Network Dataset in the Geoprocessing pane.

•      Select the feature dataset and tick road layer.

•      Set the elevation model to no elevation.  

•      Click ‘Run’.  

 

Note:  

•       When specifying the attributes for the new network dataset, length (meters) is the default cost attribute. If you want to specify other cost attributes (such as travel time in minutes) you can change these parameters in Catalog pane.

•      To change parameters, go to Catalog pane and find the network dataset in the feature dataset.

•      Right-click on network dataset and select properties.

•      Select travel attribute and choose cost tab. Here, you can specify a different cost function. In this assignment, we use the default (length) cost function. The optional part of the assignment guides to using time in network analysis.

•      Refer to this link for more details: https://pro.arcgis.com/en/proapp/help/analysis/networks/opening-the-network-dataset-properties-dialog-box.htm 

 

Finally, after checking the parameters, you can build the network dataset using the following instructions.

•      In Geoprocessing pane, search for Build Network.

•      Select the network dataset.

•      Click ‘Run’.

 

 

4.4 Optimal Routing  
WHAT THIS FUNCTION DOES:

Optimal routing finds the shortest path between two given points, known as origin and destination (O/D). The shortest path is referred with regard to either network travel time or length. An extension to optimal routing is the OD matrix function that calculates the shortest travel costs, no matter time or distance, between pairs of input O/D points.

STEPS:

•       Analysis ribbon > Network Analysis > Route. This procedure creates a complex layer which includes stops and routes.

•     Open the towns attribute table, then select #1 (Ultima), #4 (Culgoa) and #19 (Cohuna) – these are the ones which you will find the shortest route between. Hold down the Ctrl button on your keyboard to select multiple towns. Notice that the towns are listed as numbered stops.  

•     Load stops: Search for Add Locations in Geoprocessing pane. Set the route layer as the input network analysis layer. Set sub-layer as ‘Stops’ and set ‘towns’ for input locations. Set the field name to ‘FID’. Execute add locations using the Run button.

•     From Geoprocessing pane, find Solve function.

 

 

•     The cumulative distance is shown in the Total_Length column of the Routes attribute table.  

Q2. Find the optimal route between stops #1 (Ultima), #4 (Culgoa) and #19 (Cohuna). Create a map visualization of the results and attach the map to your report.

4.5 Selecting Closest Facilities  
WHAT THIS FUNCTION DOES:

Closest facilities function finds the closest one or a designated number of facilities to the given origin. It requires the input of one origin and a set of potential facility locations (which functionally are identical to destinations). The analysis then calculates the travel cost between the origin and each of the facilities and return only the requested number of facilities by ascending order of the cost. Therefore, you can get a list of nearest and 2nd nearest facility, etc.

STEPS:

•      Clear selected features.

•      From ‘site’ layer, select Site 1 (FID = 0, Locality = Site 1).

•      Analysis ribbon > Network Analysis > Closest Facility. This will create a complex layer which includes facilities, incidents, and routes.

 

•      Now, use Add Locations to load Site 1 as an incident.  

•      Use Add Locations to load all towns as facilities.

•      Click on Closest Facility in Contents pane. Go to Closest Facility ribbon and set the number of facilities to 2.

•      Search for Solve in Geoprocessing pane. Set the input to Closest Facility layer and execute the function. You could also use Run in the Closest Facility ribbon.

Q3. What are the names of the two towns closest to Site1, and what are their road distances?

4.6 Calculating Service Areas  
WHAT THIS FUNCTION DOES:

Service areas function also requires the input of a given origin. Besides, it also requires a given travel cost budget (in time or distance), say, 10 mins by car, or 1km. The analysis then returns all the road segments that are within the travel budget from the given origin. The mechanism is to calculate the travel cost from the origin to each vertex on each road segment of the road network until it hits the travel budget, for which this analysis is computationally expensive.

STEPS:

•      Clear selected features.

•      Select Swan Hill from ‘towns’ (i.e. site#0 of towns.shp).

•      Analysis ribbon > Network Analysis > Service Area. This will create a complex layer which includes facilities, and polygons.

•      Click on Service Area layer in Contents pane. This will activate the Service Area ribbon.  

•      Service Area ribbon > Import Facilities. Set load input to ‘towns’. Click ‘Run’ to add locations.

•      Service Area ribbon > set Cutoffs to 25000 m (25km). Click Run to execute service area analysis.

 

 

Q4. Describe the result of your service area network problem. Include a screen grab (i.e.

“print screen” button, paste it, and crop it).  

Q5. Without calculating the service areas, would the service area within 25 km of the township of Chinkapook be larger than that of Barham? Justify your answer.  

4.7 A few more exercises if you like (optional)
So far, we have used distance as the cost factor in our network analysis computations, but a more effective cost unit is travel time. As such, various speeds have been assigned to the different roads types: that is, regional roads (100 km/h), main roads (80 km/h), local roads and streets (60 km/h), farm roads (15 km/h) and tracks (10 km/h). Then the travel time for each road segment was computed by dividing each road segment by 1000 (to give the distance in km), dividing the km distance by the allowable speed for that segment and multiplying the result by 60 to give the travel time in minutes. This value is given in the travel time column in the roads database.

Now let's solve some very real problem faced when allocating services. We need to specify the time as the cost function for network dataset.  

•       First, create a new network dataset. After creating the network dataset, remove it from Contents pane (right-click > remove).  

•      Go to Catalog pane and find the network dataset inside the geodatabase.

 

•      Right-click on the network dataset and go to properties.

•  

•      Travel attributes > Costs > New.  

•      Add time as the cost function. Delete length from the list.

 

https://pro.arcgis.com/en/pro-app/help/analysis/networks/cost-attributes.htm 

 

1.     CREATING A GEODATABASE 

To manage your own spatial database, you can create a file geodatabase. If you do this, your data will be stored in a Microsoft Access database.

•      Open Catalog pane. Connect to your working directory by clicking on the Connect to  

Folder (right-click on Folders in Catalog pane). This creates a link to your working directory which shortens the access to your folder for future use.

•      Right-click on your working directory then select New > File Geodatabase to create a new database that will handle all the data for your project.

2.     CREATING A FEATURE DATASET 

A Geodatabase can contain different Feature Datasets, each dataset represents the data for one project. For example, if you create a Geodatabase for Australia, then inside this database you can create Feature Datasets for Melbourne, Sydney, and each dataset contains the data for a specific state.

Right-click on the Geodatabase you just created and click New > Feature Dataset. First, define a name for the dataset and click next, then click the Import button to set the coordinate system for this feature class from an existing feature class. See Fig. 1.

 

 

Figure 1: New Feature Dataset. 

3.     CREATING A FEATURE CLASS 

Right-click on the Feature Dataset you just created and click New>Feature Class, the New Feature Class dialogue box will appear then write the name for the new feature class. In Type Field you can select Point, Line, Polygon or others, then click next, see Fig. 2.

  

Figure 2: Name of a new feature class. 

 

 Finally, click Finish to create the new feature class. See Fig. 3.

  Figure 3: Properties of a new feature class. 
 To receive 2 marks the answers have to be complete and correct. There are no fractions.