The Multi-Modal Transport RDM correctly supports all transport modes via abstraction of concepts so that, for example, specific modes such as vessels, vehicles, trains, aircraft are all abstracted to become "TransportMeans". This is a useful construct because many related entities such as "Consignments" can leverage multiple modes in the journey from consignee to consignor. However, the abstract entities contain a mix of mode independent properties (like "name" and "loadedCargoMeasure") and mode specific properties (like "helipadIndicator" and "draftLevelMeasure" which are vessel specific). This leads to a risk of confusion and impossible conflicts (such as a transport means that has both a keel depth and an axle count for example).
Proposed Solution
A better modelling approach that does not change the muti-modal untility) is to add specific sub-types such as vessel, vehicle, aircraft, etc and to move mode specific properties to the sub-type. This will also help with transport industry adoption of UN/CEFACT standards because the industry organisations tend to be very mode specific.
Vessel bias in MMT?
Having moved mode specific properties, it appears that all specific properties are about vessels. Even conferenceCode is mode specific (see wikipedia entry) Information about other modes such as GVM & axle count for vehicles, seems to be missing. The separation of mode specific properties makes this bias much more obvious and serves as a useful trigger to collaborate with truck, rail, and air freight carriers to expand their mode specific models.
Problem
The Multi-Modal Transport RDM correctly supports all transport modes via abstraction of concepts so that, for example, specific modes such as vessels, vehicles, trains, aircraft are all abstracted to become "TransportMeans". This is a useful construct because many related entities such as "Consignments" can leverage multiple modes in the journey from consignee to consignor. However, the abstract entities contain a mix of mode independent properties (like "name" and "loadedCargoMeasure") and mode specific properties (like "helipadIndicator" and "draftLevelMeasure" which are vessel specific). This leads to a risk of confusion and impossible conflicts (such as a transport means that has both a keel depth and an axle count for example).
Proposed Solution
A better modelling approach that does not change the muti-modal untility) is to add specific sub-types such as vessel, vehicle, aircraft, etc and to move mode specific properties to the sub-type. This will also help with transport industry adoption of UN/CEFACT standards because the industry organisations tend to be very mode specific.
Vessel bias in MMT?
Having moved mode specific properties, it appears that all specific properties are about vessels. Even conferenceCode is mode specific (see wikipedia entry) Information about other modes such as GVM & axle count for vehicles, seems to be missing. The separation of mode specific properties makes this bias much more obvious and serves as a useful trigger to collaborate with truck, rail, and air freight carriers to expand their mode specific models.