Aeronautical and Automotive Engineering


Reliability modelling projects

Project 1 - Automated Petri Net Model Generation

With the current developments in engineering technologies the complexity of these engineering systems or processes is ever increasing and proportionally so potentially do the risks and hazards. The requirement of the detection of the most critical components and design errors at an early stage of design becomes more challenging and the need for an automated reliability analysis with the view to save time, money and effort increases. Hence, to overcome limitations of existing reliability models which can be cumbersome, error-prone and time consuming, the automated generation of reliability modelling methods for complex industrial systems or processes is investigated. My current work is based on a new methodology of constructing Petri Nets from a system or process description.

Hence, the contribution of my work is the development of a methodology that targets the automatic generation of Petri Net models by taking as input the system or process description, including both structural and behavioural information, in a form widely used in industry, overcoming the laborious challenge to develop the system or process description manually. The proposed methodology also overcomes the current semi-automatic attempts focuses on the Petri Net generation by enhancing the methodology’s strength and applicability.

The automated Petri Net generation retrieves the structural and behavioural information required for the model construction from the system or process description using a UML Activity Diagram (AD). The UML AD of a system or process is exported in XMI format and the XMI file is transformed into XML format using XSLT templates. A MySQL algorithm has been developed, that uses the XML model, to automatically construct the incidence matrix of the overall Petri Net, which is the mathematical representation of the model. The automated graphical representation and simulation of Petri Nets are also included in my research. Future work involves the application of the automated PN generation methodology in complex processes to verify its applicability.