Difference between revisions of "Requirements to Set up a Model"

From IWRAP
Jump to navigation Jump to search
(New page: == With AIS Data == Sea chart A sea chart containing depth curves for 0m, 5m and 10m (or close to this). The chart must be in a commonly used picture format like jpg, bmp or tif. It mus...)
 
 
(4 intermediate revisions by the same user not shown)
Line 1: Line 1:
The following describes what is needed before a model can be set up.


== With AIS Data ==
== Without AIS Data Available==


Sea chart
{| style="color: black;" width="100%" class="wikitable"
A sea chart containing depth curves for 0m, 5m and 10m (or close to this). The chart must be in a commonly used picture format like jpg, bmp or tif. It must contain some latitudes and longitudes so that readings can be made.
!colspan="3"|Without AIS data available
|-
|Sea chart||colspan="2"|The sea chart must be in a commonly used picture format like jpg, bmp or tiff. This can be imported into Google EarthTM for defining the grounds.


Density plot If this is available it should be in a commonly used picture format like jpg, bmp or tiff. It must contain some latitudes and longitudes so that readings can be made. Coordinates of the corners of the picture is the best.
Shipping lanes The most important shipping lanes/legs are drawn in a sea chart. It must be possible to read the coordinates of the legs.
A density plot is not required, just nice to have.
|-
|Ship data||colspan="2"|For each leg the number of ships in each direction must be known. this should be in the form of a statistical summary of Ship types and length classes
 
Based on the ship ength, IWRAP extracts the default breadth, draught and speed from an internal statistical database. However the default draught and speed can also be overridden by the user.
|-
|Traffic distributions||The traffic distribution of each leg in each direction must be given.
 
In many cases a normal distribution will do the job. In that case the mean distance and the standard deviation from the leg must be given.  


AIS data At least 3 month of AIS data for the area. GRISK calculates frequencies for 1 year, so 3 month of AIS data will be multiplied by 4.
|[[Image:20090407 Normal DistributionTVD.gif|200px]]


In many cases a sampling time of 6 minutes will do the job. But if the lanes are very curved and the ships are fast the original AIS data should be used.
|-
|Junction data||Where legs intersect or split we need to know which fraction sails to which leg.


The AIS data must be in one or more text files (CSV files) with at least the following fields:
For the picture below the following info is needed
MMSI, IMO, TIME, “NAME”, “DESTINATION”, TYPE_OF_SHIP_AND_CARGO, DRAUGHT, SIZE_A, SIZE_B, SIZE_C, SIZE_D, LAT, LON, COG, SOG


Time should be in a format close to DD-MM-YYYY hh:mm:ss
Proportion of ships on leg 8 going from 8 to 7
Latitude and longitude must be a decimal number like 56.8723 not 56º 52’ 20.3’’
Text fields like Name and Destination should be encapsulated by “
The preferred decimal comma is ‘.’ At least make sure the columns are not separated by the same symbol as the decimal comma.


Each CSV-file should not contain more than 1-3 month of data and should be less than 1Gb.
Proportion of ships on leg 8 going from 8 to 33


Statistics for validation For validation of the results statistic of the number of collisions and groundings that have occurred in the area within the last 10-15 years or so can be used.
Proportion of ships on leg 7 going from 7 to 8


Proportion of ships on leg 7 going from 7 to 33


Proportion of ships on leg 33 going from 33 to 8


== Witout AIS Data ==
Proportion of ships on leg 33 going from 33 to 7


The fractions can be based on individual ship types or the total number of ships


Sea chart
|[[Image:20090407 Leg Junction.gif|300px]]
The sea chart must be in a commonly used picture format like jpg, bmp or tiff. This can be imported into Google EarthTM for defining the grounds.


Shipping lanes The most important shipping lanes/legs are drawn in a sea chart. It must be possible to read the coordinates of the legs.
|-
|Statistics for validation||colspan="2"|For validation of the results statistic of the number of collisions and groundings that have occurred in the area within the last 10-15 years or so can be used.
 
|}


Ship data For each leg the number of ships in each direction must be known. The type and length [m] of the ships must be given. If information is available the following information should be given.


== With AIS Data and Automatic Extraction Tools Available ==


If the ship types are not known in detail the following can be used.
(Commercial version - This is not the case for users using the free IALA version of IWRAP).


    or just   
{| style="color: black;" width="100%" class="wikitable"
!colspan="2"|With AIS data and Automatic Extraction Tools Available
|-
|Sea chart||A sea chart containing depth curves for 0m, 5m and 10m (or close to this). The chart must be in a commonly used picture format like jpg, bmp or tif. It must contain some latitudes and longitudes so that readings can be made.
|-
|Density plot||If this is available it should be in a commonly used picture format like jpg, bmp or tiff. It must contain some latitudes and longitudes so that readings can be made. Coordinates of the corners of the picture is the best.


For the frequency analysis the individual ship type is not that important. From the ship type and length GRISK calculates the breadth, draught and speed. However draught and speed can also be given by the user.
A density plot is not required, just nice to have.
Traffic distributions The traffic distribution of each leg in each direction must be given. In many cases a normal distribution will do the job. In that case the mean distance and the standard deviation from the leg must be given.  
|-
|AIS data||At least 3 month of AIS data for the area. IWRAP calculates frequencies for 1 year, so 3 month of AIS data will be multiplied by 4.


In many cases a sampling time of 6 minutes will do the job. But if the lanes are very curved and the ships are fast the original AIS data should be used.


Junction data Where legs intersect or split we need to know which fraction sails to which leg. For the picture below the following info is need
The AIS data must be in one or more text files (CSV files) with at least the following fields:
MMSI, IMO, TIME, “NAME”, “DESTINATION”, TYPE_OF_SHIP_AND_CARGO, DRAUGHT, SIZE_A, SIZE_B, SIZE_C, SIZE_D, LAT, LON, COG, SOG


Proportion of ships on leg 8 going from 8 to 7
Time should be in a format close to DD-MM-YYYY hh:mm:ss
Proportion of ships on leg 8 going from 8 to 33
Proportion of ships on leg 7 going from 7 to 8
Proportion of ships on leg 7 going from 7 to 33
Proportion of ships on leg 33 going from 33 to 8
Proportion of ships on leg 33 going from 33 to 7


The fractions can be based on individual ship types or the total number of ships


Latitude and longitude must be a decimal number like 56.8723 not 56º 52’ 20.3’’
-----


Text fields like Name and Destination should be encapsulated by ""


The preferred decimal comma is '.' At least make sure the columns are not separated by the same symbol as the decimal comma.


Each CSV-file should not contain more than 1-3 month of data and should be less than 1Gb.
|-
|Statistics for validation||For validation of the results statistic of the number of collisions and groundings that have occurred in the area within the last 10-15 years or so can be used.
|}





Latest revision as of 10:19, 13 May 2012

The following describes what is needed before a model can be set up.

Without AIS Data Available

Without AIS data available
Sea chart The sea chart must be in a commonly used picture format like jpg, bmp or tiff. This can be imported into Google EarthTM for defining the grounds.

Shipping lanes The most important shipping lanes/legs are drawn in a sea chart. It must be possible to read the coordinates of the legs.

Ship data For each leg the number of ships in each direction must be known. this should be in the form of a statistical summary of Ship types and length classes

Based on the ship ength, IWRAP extracts the default breadth, draught and speed from an internal statistical database. However the default draught and speed can also be overridden by the user.

Traffic distributions The traffic distribution of each leg in each direction must be given.

In many cases a normal distribution will do the job. In that case the mean distance and the standard deviation from the leg must be given.

20090407 Normal DistributionTVD.gif
Junction data Where legs intersect or split we need to know which fraction sails to which leg.

For the picture below the following info is needed

Proportion of ships on leg 8 going from 8 to 7

Proportion of ships on leg 8 going from 8 to 33

Proportion of ships on leg 7 going from 7 to 8

Proportion of ships on leg 7 going from 7 to 33

Proportion of ships on leg 33 going from 33 to 8

Proportion of ships on leg 33 going from 33 to 7

The fractions can be based on individual ship types or the total number of ships

20090407 Leg Junction.gif
Statistics for validation For validation of the results statistic of the number of collisions and groundings that have occurred in the area within the last 10-15 years or so can be used.


With AIS Data and Automatic Extraction Tools Available

(Commercial version - This is not the case for users using the free IALA version of IWRAP).

With AIS data and Automatic Extraction Tools Available
Sea chart A sea chart containing depth curves for 0m, 5m and 10m (or close to this). The chart must be in a commonly used picture format like jpg, bmp or tif. It must contain some latitudes and longitudes so that readings can be made.
Density plot If this is available it should be in a commonly used picture format like jpg, bmp or tiff. It must contain some latitudes and longitudes so that readings can be made. Coordinates of the corners of the picture is the best.

A density plot is not required, just nice to have.

AIS data At least 3 month of AIS data for the area. IWRAP calculates frequencies for 1 year, so 3 month of AIS data will be multiplied by 4.

In many cases a sampling time of 6 minutes will do the job. But if the lanes are very curved and the ships are fast the original AIS data should be used.

The AIS data must be in one or more text files (CSV files) with at least the following fields: MMSI, IMO, TIME, “NAME”, “DESTINATION”, TYPE_OF_SHIP_AND_CARGO, DRAUGHT, SIZE_A, SIZE_B, SIZE_C, SIZE_D, LAT, LON, COG, SOG

Time should be in a format close to DD-MM-YYYY hh:mm:ss


Latitude and longitude must be a decimal number like 56.8723 not 56º 52’ 20.3’’

Text fields like Name and Destination should be encapsulated by ""

The preferred decimal comma is '.' At least make sure the columns are not separated by the same symbol as the decimal comma.

Each CSV-file should not contain more than 1-3 month of data and should be less than 1Gb.

Statistics for validation For validation of the results statistic of the number of collisions and groundings that have occurred in the area within the last 10-15 years or so can be used.