LUPTAI Project Structure

A LUPTAI model consists of a number of files and directories, however all the essential input and output data lives within a single spatialite file for each modelled scenario that is being analysed.

The LUPTAI QGIS structure inherits a degree of similarity from the prior ArcGIS plugin.

The typical project structure looks similar to the following:

Projects

├── 2021_SEQ_model

│ ├── Model.qgz

│ ├── Scenarios

│ │ ├── base_model.sqlite

│ │ └── raw_gtfs_pt_model.sqlite

│ └── ModelData

│ ├── base_model

│ └── raw_gtfs_pt_model

│

├── luptai_sample_project

│ ├── Model.qgz

│ ├── Scenarios

│ │ ├── base_scenario.sqlite

│ │ └── Story_bridge_removal.sqlite

│ └── ModelData

│ ├── base_scenario

│ └── Story_bridge_removal

The Scenarios folder and the spatialite (sqlite) files within store all the data comprising a scenario, along with the results of any model runs completed with that scenario. The ModelData folder contains the raw unparsed model input and output in an xml interchange format. It is not interacted with in general use.

Any project or scenario created through LUPTAI QGIS will automatically propagate the correct structure. Note that as a scenario primarily consists of a single file, this can be copied between projects freely (unlike the ArcGIS version which contained a project tracker). You can also rename a scenario, however the results tables contained within the spatialite file will not be updated correspondingly (which may or may not be important depending on your use case).

LUPTAI Scenario Structure

Now that we have seen what a luptai project looks like on disk, we examine the luptai scenario format. This may be useful if you are looking to import your own custom data into LUPTAI. Throughout this discussion it will be beneficial to be able to inspect the sample scenarios to examine the data format in more detail. This can be done with an external tool, such as DB Browser (sqlite), or directly with QGIS by adding a spatialite layer and including all the tables within a scenario file (including those without geometry).

Here we see a list of all the geometry columns and the tables with no geometry columns. We now cover the format of the tables intended to be constructed by the user, to aid in the process of importing custom data. This includes the tables network_links, network_nodes, activities_points and activities_attributes. Note that whilst any Coordinate Reference System (CRS) is supported for the import layers, they will be converted to EPSG:4283 in the scenario spatialite. Layers with invalid CRS will have the CRS EPSG:4326 manually provided to them, which may or may not produce sufficient results.

Network Links Table

The following descibes the network_links table structure within the spatialite file. The columns marked essential contain information used by LUPTAI and must be provided. The remaining columns are nullable, containing non-essential metadata inherited from the QBase and LUPTAI ArcGIS data formats.

Column	Data Type	Description	Example Values	Used in Comp.
fromnodeid	INTEGER	node in network_nodes	17708	x
tonodeid	INTEGER	node in network_nodes	17709	x
linkid	INTEGER	unique per link	219119	x
meters	FLOAT	link shape length in metres	1240.042	x
enabled	INTEGER	boolean flag to ignore from model	1	x
luptaitype	INTEGER	see below	5	x
roadtype	INTEGER	see below	5	x
onewayind	INTEGER	see below	0	x
shape	BLOB	linestring geometry	<wkb blob>
objectid	INTEGER	auto incremented id	18400
street	VARCHAR(50)	street name	HUCKER ROAD
fromleft	FLOAT	Historical [1]	0
toleft	FLOAT		0
fromright	FLOAT		204
toright	FLOAT		204
alias_name	VARCHAR(50)		HUCKERS ROAD
pid	FLOAT		3E+08
qual_pnt	VARCHAR(20)		CDR4M1220100910
road_id	FLOAT		335581
locality	VARCHAR(50)		SPRING CREEK
state	INTEGER		3
source	VARCHAR(2)		2

RoadType Field

The road type field provides a classification of the type of link. The following comments and encoding table is extracted from the original LUPTAI technical documentation. (See Section 5.6.8.2 for more details)

Non-walkable links should be removed from the network (manually, this is not done as part of the import). This will include motorways, on and off ramps etc. This can be done by using the ROADTYPE field from the SDRN. Some guidelines are in the following table.

Value	Classification	Exclude from LUPTAI Walk Network
1	Freeways/Motorways	Yes exclude – no pedestrian access. Note there are a few class 1 roads which need to be kept in to maintain connectivity of the walk network
2	Highways	Some network segments may need to be removed. Requires some local knowledge to determine if they should be removed.
3	Secondary Roads	Do not exclude.
4	Local Connector Roads	Do not exclude.
5	Street/Local – only provides property access	Do not exclude.
6	Notified Private or Restricted Roads	Do not exclude. These roads while private or restricted to vehicles, may provide pedestrian access especially where they occur within “gated” communities.
7	4WD and Tracks	Do not exclude – these tracks are likely to be walked by pedestrians; however, it is believed that they will occur in locations that would not impact duly on the results of a LUPTAI analysis.
8	Malls/Passenger Ferry	Do not exclude except remove Passenger Ferry.
9	Construction lines	Yes exclude – these lines are used to construct some road segment, e.g. roundabouts which are included under other classification codes.
10	Unconstructed and/or Dedicated	Yes exclude – these roads either do not exist or do not have pedestrian access.
11	Vehicle Ferry	Yes exclude – to appropriately include vehicle ferries would require a time schedules for services and additional capabilities within the LUPTAI model to account for these services. Even so, it is considered that most of these services are located in areas that would not influence pedestrian access within the LUPTAI analysis.
12	Bikeways/Walkways	Do not exclude but using the LUPTAITYPE field identify which are cycle and which are walk paths While bikeways may not be shared with pedestrians, it is assumed in the majority of cases they will be shared. As there is no attribute that identifies shared and unshared bikeways they are not excluded.
13	Busways	Exclude except for a few short links which are classed as busway but have footpaths

LUPTAI Type Field

The luptai type field encodes additional information for walk and cycle links. The following comments and encoding table is extracted from the original LUPTAI technical documentation. (See Section 5.6.8.5 for more details)

For roads the LUPTAITYPE is the same as the SDRN RoadType however for cycle, walk or shared walk/cycle links these are specified by the following codes.

Value	Description	Comments
14	Footpaths	Footpaths, cyclists can only travel slowly on these links
15	Shared walk/ cycle paths	These represent larger footpaths or shared paths in parks etc. where cyclists can travel at a reasonable speed but must still be wary of pedestrians
16	Bikeways	Cyclists can travel at maximum speed on these links
17	Cyclable but non walkable links	There are some road links which can be cycled on but have no walk access. These should be flagged with this code

OneWayInd Field - Network Connectivity

The OneWayInd Field encodes legal traversal directions for the network links. These are only used for the cycle mode (as car is not implemented in LUPTAI QGIS).

The following comments and encoding table is extracted from the original LUPTAI technical documentation. (See Section 5.6.8.1 for more details)

The network is non-directional for walk mode but is directional for cycling. It uses the OneWayInd field from the SDRN which has the following values. It is assumed that cyclists follow the same restrictions as cars on the road network.

Value	Classification
0	Both ways
1	One way with the vector
2	One way against the vector
3	No way

Network Nodes Table

The network nodes table contains the nodes of active transport network. The required values identify the position of the node.

Column	Data Type	Description	Example Values	Used in Computation
nodeid	INTEGER	Unique nodeid	18482	x
easting	FLOAT		382101.2809	x
northing	FLOAT		6922582.148	x
objectid	INTEGER	Auto incremented id	18411
longitude	FLOAT	Historical [1]	151.8029726
latitude	FLOAT		-27.81620836
utm_zone	INTEGER		56
shape	BLOB	Point geometry	<wkb blob>

Activities Points Table

The activity points table encodes information regarding the activity points. The node id field should not be populated, it should be set to 0. It will be updated as part of the import process to reflect the base activity transit network in the scenario. Note that the sentinel values of -1 are used to denote unsnapped nodes. Node id of -2 denotes open space activity type points which were snapped correctly, but have been excluded from the model as they are in high proximity to other open space points. The activity types currently supported in LUPTAI QGIS are [‘CHILDCARE’, ‘COMMUNITY’, ‘HOSPITAL’, ‘GP’, ‘PRIMARY SCHOOL’, ‘SECONDARY SCHOOL’, ‘TERTIARY’, ‘OPEN SPACE’, ‘SUPERMARKET’, ‘EMPLOYMENT’], noting that ‘EMPLOYMENT’ is generally supplied as an attributed activity.

Column	Data Type	Description	Example Values	Used in Comp.
nodeid	INTEGER	node in network_nodes	226781	x
enabled	INTEGER	boolean flag to ignore from model	1	x
act_type	VARCHAR(50)	see below	PRIMARY SCHOOL	x
id	INTEGER	unique nodeid	490
feature	VARCHAR(50)	auto incremented id	Primary School
name	VARCHAR(200)	Historical [1]	Nerang State School
label	VARCHAR(400)		Qld_Digital Cadastral Database
source	VARCHAR(50)		QSC2020 - Landmark Data
shape	BLOB	point geometry	<wkb geometry>

Activities Attributes Table

In LUPTAI QGIS, attributed activities are only used for employment and the data format is specified as follows. Note that the given employment columns are only sample values, any non conflicting column name can be specified. These are manually associated with the respective employment types through the model run dialogue.

Column	Data Type	Description	Example Values	Used in Comp.
emp_blue2016 [2]	FLOAT	Blue Collar emp	14.02	x
emp_white2016 [2]	FLOAT	White Collar emp	14.13	x
emp_total2016 [2]	FLOAT	Total emp	28.12	x
(arb. float cols.) [2]	FLOAT	see below		x
enabled	INT		1	x
act_type	VARCHAR(50)	see act type above	EMPLOYMENT	x
nodeid	INT	network_nodes id	17008	x
zonecode	BIGINT	zone system id (i.e. SA1)	30908126121

[2] (1,2,3,4)

The table supports attributed activities with arbitrary column names however only the three employment type attributes are exposed by the QGIS plugin. The python notebook api can use the other column names with appropriate setup.

[1] (1,2,3)

Historic fields for compatibility with LUPTAI ArcGIS