Verification of requirements

Verification of requirents

When specifications of requirements are expressed in Gellish and also design data about imaginary objects or measured data about observed real objects are expressed in Gellish then it becomes possible that software verifies whether design data or data about fabricated real objects satisfy those requirements. In other words, when requirements are expressed according to the Gellish Modeling Method and delivered information that intents to satisfy those requirements is delivered according to the Gellish Modeling Method as well, then that enables computer-aided verification of the completeness and other quality aspects of the delivered information.

For example, it may be a requirement that when a objects of particular kinds are delivered that then particular kinds of documents shall be delivered as well and that such objects shall have particular kinds of parts for which spare parts shall be delivered. The Gellish Modeling Method enables that such requirements are automatically verified, provided that the information about the delivered documents, parts and spare parts is also delivered according to the Gellish Modeling Method.

1. The verification process

The Gellish Modeling Method includes a process to verify the requirement that objects of specified kinds shall have particular kinds of aspects, kinds of parts, etc. The core of that process is the following.

Assume that for a number of kinds of objects there are requirements specified. For example, assume that a facility owner has expressed the requirement that all vessels shall have a length between tangent lines. This is expressed as follows (omitting the UID’s):

Name of left hand object Name of relation type Name of right hand object
vessel shall have as aspect a length between tangent lines

In Gellish requirements are specified using kinds of relations that are denoted by a phrase that start with ‘shall have …’ or ‘shall be …’. The Gellish dictionary (ontology) specifies for each requirement that is expressed using such a kind of relation which kind of relation can satisfy the requirement. For example, the dictionary specifies (in the ‘extended language definition’ section) that the kind of relation denoted by the phrase ‘shall have as aspect a’ can be fulfilled by assertions that use a kind of relation that is denoted by the phrase ‘has as aspect’. This done as follows:

Name of left hand object Name of kind of relation Name of right hand object
shall have as aspect a can be fulfilled by a has as aspect

The above implies that for every object that is classified as a vessel (or one of its subtypes) the computer should search whether such objects have a relation that is classified as a ‘has as aspect’ relation and if found whether the aspect is classified as a ‘length between tangent lines’ and is quantified by a value on a scale. If that is the case, then the requirement is satisfied.

For example, assume that a piece of equipment, called V-1206, is delivered to the above facility owner and that V-1206 is classified as a vessel. Then the database system of the receiving party can automatically ‘know’ from the above described specification of requirements that the delivered data is only complete when V-1205 has a length between tangent lines with a value. Furthermore, assume that the data about the equipment is delivered in the form of Gellish expressions as follows:

UID of left hand object Name of left hand object UID of idea UID of kind of relation Name of kind of relation UID of right hand object Name of right hand object UID of UoM UoM
1 V-1206 101 1225 is classified as a 520121 horizontal vessel
1 V-1206 102 1727 has as aspect 2 length of V-1206
2 length of V-1206 103 1225 is classified as a 550362 length between tangent lines
2 length of V-1206 104 5025 has on scale a value equal to 922365 3100 570423 mm
3 V-1207 105 1225 is classified as a 520242 vertical vessel

This enables that the software can verify the data automatically and can report that for V-1206 the requirement is satisfied and for V-12-7 the requirement is not satisfied. This enables to verify automatically the progress of the delivery of the data. It also enables to determine an information quality Key Performance indicator for completeness of the data.
The requirement as well as the qualification can also be expressed by variants of the relation type, such as:

  • 5,489 has on scale a value greater than or equal to
  • 5,490 has on scale value less than or equal to
  • 5,468 has on scale a value within numeric range
  • 5,020 is qualified as

The specification of the requirement that the aspect values may only be selected from a list of standard qualitative aspects is also possible.

A requirement for having a part of a particular kind is specified using a ‘shall have as part a’ relation, which can be satisfied by a ‘has as part’ relation. For example, the requirement that a vessel shall have at least 2 nozzles can be specified by expressing a minimum and maximum cardinality, as follows:

vessel shall have as part a 2,n nozzle

This is satisfied when that data about each vessel specifies that it has as part at least two nozzles.

The verification of other kinds of requirements follow a similar process.

Continue with Automated Translation