Using decisional DNA to enhance industrial design and manufacturing

Research Doctorate - Doctor of Philosophy (PhD)This research addresses issues associated with using ever-increasing amounts of industrial manufacturing information and knowledge more effectively, and taking advantage of knowledge generated through experience. The thesis proposes a novel approach where the Set of Experience Knowledge Structure (SOEKS) and Decisional DNA (DDNA) techniques are used to provide engineering artefacts, processes and systems with an experience based representation. In the presented approach comprehensive Knowledge Representation (KR) of industrial manufacturing processes, data, and information is captured at three levels: machine, shop floor, and factory level. Consequently, a new concept of Virtual Industrial Manufacturing (VIM) comprising of Virtual Engineering Object (VEO), Virtual Engineering Process (VEP) and Virtual Engineering Factory (VEF) is introduced. One key difference between VIM and traditional approaches is that the knowledge imbedded in VIM is based on the concept of collecting formal decision events (experiences), which are extracted by capturing the ongoing daily interactions between the manufacturer and engineering objects, processes and the whole plant. A Virtual Engineering Object is a novel Knowledge Representation of an engineering entity that draws on the past experience and formal decisions associated with this given entity. In other words, a VEO gathers, stores and reuses its own experience in a way similar to a human expert. The concept of a Virtual Engineering Process is an experience-based KR of engineering processes and is an extension of the VEO. The VEP model includes the overall process information along with the experience of engineering objects (VEOs) required for manufacturing an engineering component. The Virtual Engineering Factory is the representation of experience at the plant level having links to associated VEOs and VEPs. Finally, a Manufacturing DNA which is a dynamic knowledge structure, flexible and able to change according to the dynamics of manufacturing environment, is created by integrating experience of VEOs, VEPs and VEF. The VEO-VEP-VEF system supplies compressed information and data derived from their complex interrelationships and communicates this in a personalised manner as the basis for their intervention in the process. This gives rise to a new form of cooperation among processes, machines and parts of machines, supporting both short-term flexibility and medium-term transformability and improving the production resilience. The fast expansion of the Information and Communication Technology (ICT) in recent time has evolved into Industry 4.0 - a new industrial revolution. Industry 4.0 assimilates several concepts among which the Cyber-Physical System is probably the most important one. This research also demonstrates that a VEO is a specialization of Cyber-Physical System (CPS) in terms of its extension in real time knowledge gathering, semantic analysis, visualisation and reuse. The main contribution of the VEO-VEP-VEF notion is not only to formulate a procedure to classify types of manufacturing knowledge, but also to create an appropriate methodology to utilise experience using Decisional DNA. The VEO-VEP-VEF model qualifies to be used as a decision-making tool that can enhance different manufacturing systems with its predicting capabilities. The proposed system readily copes with self-organising production and control strategies, which is a significant example of linking product lifecycle management, industrial automation and semantic technologies. The decision making process at every level in industrial design and manufacturing will benefit from this approach, as it includes capturing, storage and reuse of experience and knowledge in an efficient smart manner