OWL based on Description Logic technology is released by W3C. According to OWL, “Classes, Attributes, Properties, and Individuals” are used for representing the knowledge structure, at the same time, “axioms” and “functions” are used for supporting semantic reasoning processes (Sengupta and Hitzler, 2014). Based on OWL, we can define the elements and their structural relationships of the real domain or cross-domain systems explicitly, and then the cross-domain modeling and interoperating are available in the information spaces. SWRL is a rule description language evolves from RuleML (Lezcano et al., 2011). Based on OWL, SWRL integrated various rule description mode, such as Horn-like, Unary/Binary Datalog, and RuleML, to improve OWL’s rule description and semantic reasoning capacities. Then, based on the nuclear emergency system ontologies based on OWL, the cross-domain nuclear emergency response logics can be transformed into SWRL rules. And by intelligent logic reasoning, the available integrated response solutions will be obtained rapidly. Besides, SWRL rule is one kind of production rule, there is no much nested or iterative process. And with the relative low time computational complexity, SWRL rules are useful for improving the efficiency of integrated decision in the nuclear emergency response processes. Therefore, theories above provide some new ideas. Firstly, by feature mapping and modeling, the key elements, relationships, actions, constraints of the disaster scenarios and the domain systems in original physical space can be extracted and abstracted into the target integration space. At the same time, by uniform expressing, the inherent heterogeneity of the integration processes in grammar level can be avoided. Secondly, in the process of mapping, it is possible to redefine and reorganize the relationships between the elements from different domain systems, and the systems can be reconstructed into an ordered cross-domain system in the target space. Thirdly, based on OWL and SWRL, the structure models and reasoning rules of cross-domain nuclear emergency systems can be represented formally in the target spaces. Finally, based on the above two points, an interoperable, reachable and ordered knowledge network is formed in the target space, and then the cross-domain integrated response solutions can be discovered intelligently by knowledge reasoning. Thus, based on the ideas above, this paper proposes the cross-domain integrating and reasoning spaces (CDIRS) to support the integrated decision making process of the offsite nuclear emergency response. The paper is organized as follows. Section 2 describes the CDIRS framework we proposed for the cross-domain integrating and reasoning process. Section 3 presents the mapping and modeling principle to construct the CDIRS spaces. Section 4 describes the interoperability problems of the CDIRS in semantic and business logic levels, and it presents the corresponding solutions and interoperability validation methods. Section 5 presents the ordering rules constructing and reasoning principle for integrated decision making in the CDIRS spaces. Furthermore, Section 6 gives an example of offsite nuclear emergency response to verify the validity of the methods presented in this paper. Finally, the paper is concluded in Section 7.
