The Unique Consignment Reference - a Nirvana powered by DID/VC?

Hello colleagues,

I'd like to apologise in advance for dropping this new question / use case
upon you so soon after the one about VC trust chains for cross-border
documents - but hopefully you'll find this use case interesting enough to
justify a little of your time.

A "consignment" is probably the most fundamentally important thing in
cross-border trade.  It represents the movement of goods from consignor to
consignee.  It is created by a transport service provider (eg fedex) on
request of a transport service buyer (which can be the consignor or
consignee depending on incoterms). In today's paper world it is described
by a paper document called a "bill of lading" (sea freight) or an "airway
bill" (airfreight).  There are road freight versions too. Consignments also
exist in a hierarchy.  For example an "ocean bill" represents the transport
contract between a forwarder (eg DBSchenker) and a carrier (eg Maersk)
whilst a "house bill" represents the transport contract between a trader
(eg ACME) and a forwarder (eg DB Schenker). A "House Bill" represents the
common scenario where an intermediary buys a container from a carrier and
then re-sells space on that container to multiple traders.  Master Airway
bills / Ocean Bills already have a unique naming convention - for example
the lufthansa cargo AWB at https://lufthansa-cargo.com/toolbox has ID
020-22469812 where "202" is the IATA issued code for Lufthansa cargo.

In the paper world, the bill is an authoritative legal document that is
referenced by many other documents.  In fact the content on a bill is often
duplicated and extended on other documents. For example, the certificate of
origin mentioned in the previous email thread about trust chains has maybe
50 data elements, all but 3 of which duplicate consignment information
already on the bill (ie goods description, value, tariff code, consignee,
consignor, etc). if instead the certifciate of origin could just reference
a discoverable and non-repudiable consignment document then the certificate
really only need say "as an authorised body (eg chamber of commerce), I
certify that the goods referenced by consignment XYZ123 comply with orgin
criteria ABC under Free Trade Agreement XYZ".  Basically 3 new data
elements instead of 50.

But the certificate of origin is only one of dozens of documents that
reference a consignment.

   - regulatory declarations (to both the exporting and importing
   regulator) reference and duplicate consignment information.
   - financial instruments like letters of credit reference and duplicate
   consignment information.
   - and so on.

in many cases, each stakeholder creates their own reference for the
consignment. Australian Border Force for example calls is a "Customs
reference number" or CRN.  Traders create their own shipping reference
numbers.  The proliferation of identifiers for the same thing and the
challenges of confident data linking have led the WCO (World Customs
Organistion) to push for a UCR "Unique consignment reference" that all
stakeholders in the supply chain can reference.  See
http://www.wcoomd.org/en/topics/facilitation/instrument-and-tools/tools/ucr.aspx.
There is a fatal flaw with the WCO model in that they envision the issuer
as a centralised cartel member like GS1 or DUNS. That's partially because
they cant envision a document discovery model that doesn't include some
kind of uber central registry. I think the decentralised model offered by
DID/VC offers a more scalable, privacy preserving, higher integrity
approach that can still achieve the same goal. If this can be made to work
then it is genuinely transformative for the international supply chain and
will shave a fair chunk off the "cost of trade" - the value of which is
measured in $ trillions / year globally.

With all this in mind, here's some naive questions for you all:

   1. Can DIDs be used to identify things (eg a consignment) as well as
   entities (eg people or organisations)?
   2. Would the actual data about the consignment be contained in a VC or
   in a DID related document ?  It's worth noting here that the bill is a
   commercial in confidence (ie non-public) document.
   3. In order to blend with the non-DID/VC world it's reasonable to expect
   that various EDI documents might reference a consignment via it's DID- eg
   did:{some method}:{some uuid}.  Or would it be better to reference the
   consignment as a VC via a URI/QR code that is the key to discovery? This
   depends on the answer to 1 & 2.  ie is the bill of lading a VC issued by a
   party identified by a DID or is the bill of lading itself a DID?
   4. If the UCR is a DID did:{some method}:{some uuid} - then which of the
   bewildering array of existing methods would be most appropriate to use
   where the goal is a to use the DID as a discovery key to access a (possibly
   encrypted) document.  ie "I have the DID now I want the data in the
   document this DID references"

I expect (hope) that this will spark a bit of a discussion and that more
questions and answers will ensue.

-- 
Steve Capell