Investigation into the Theoretical Properties of, and the Relationship Between, AI Planning Domain Models.

Over the past decade, there have been substantial advances in knowledge representation (KR)
techniques for AI planning. Typically, planners search a space of solution to find a suitable
and most accurate sequence of actions to achieve a specific task from a set of initial and goal
states. However, the progress in this field still cannot cope with the ever increasing of the
complexities of modern systems, which makes knowledge representation an expensive and
error prone process. Planning is considered as one of artificial intelligence fields where
knowledge representation (KR) is extremely critical. However, a little work has been aimed
at “measuring” domain models; the aim of this research is to develop a set of criteria and
metrics to assess the accruing and complexity of a particular classical planning problem
domain model. To reach that point the system has to have enough knowledge and knowledge
has to be well represented for the problem at hand. In this report, we have outlined the
prototype the system and design planning domain model metric tools.
1.2 Introduction