To support creation and exploitation of expert systems and knowledge bases by non-programmers for the industrial safety expertise. We propose an approach and a tool for designing rule-based expert systems (ES) and knowledge bases (KB) using the modified (adopted) model-driven development (MDD) approach; in particular, we employ a model-driven architecture (MDA). The purpose of this paper is to describe the adaptation (modification) and application of the MDA/MDD approach for designing rule-based KBs and ESs. The proposed modification uses the main principles of MDA/MDD (e.g., model types and creation stages), and it also considers the features of developing intelligent systems, in particular, rule-based ESs and KBs through specialization and redefinition of certain models and stages. In particular, we suggest the following: – using ontologies and conceptual models to represent a computation-independent model; – using the original author’s notation – a rule visual modelling language (RVML) to improve the visibility of representations of cause-effect relations for designing platform-independent models; – using a CLIPS (C language integration production system) as the platform model; – defining the rules of model transformation in accordance with the principles of the MDA/MDD in the context of designing KBs and ESs. The proposed approach is designed for non-programmers: experts and system analytics who are able to develop only two information models: a CIM (ontology) and PIMs (models of a rule-based KB and ES). In this case, it is possible to automate the PIMs’ creation with automated analyses of conceptual models. According to the MDA/MDD approach, other models are either integrated into the software that implements the approach or they are created automatically. The benefits of the proposed approach in comparison with the standard method of ES development are as follows: – a significant reduction of time for the implementation stage and the elimination of programming errors through automatic code generation; – a reduction in time for the identification, conceptualization, and formalization stages due to the use of an ontology and cognitive graphics. The proposed approach is implemented in the form of a research prototype of software (Personal Knowledge Base Designer) that is intended for the rapid development of prototypes of rule-based KBs and ESs. The main advantages of the personal knowledge base designer are listed below: the built-in editor of models; integration with IBM Rational Rose (in terms of imports of UML-models); generation of CLIPS code and specifications for the interpreter; using models at runtime. The proposed approach and the tool are used in Irkutsk Research and Design Institute of Chemical and Petroleum Engineering (IrkutskNIIhimmash). The main effect of its using is achieved both in research to identify laws of change for the technical conditions of the petrochemical objects (pipes, apparatus and etc.) and in the industrial safety expertise. The reported study was partially supported by RFBR (research projects No. 16-37-00041, 15-07-05641).