Role based modelling in support of configurable manufacturing system design

Business environments, in which any modern Manufacturing Enterprise (ME) operates, have grown significantly in complexity and are changing faster than ever before. It follows that designing a flexible manufacturing system to achieve a set of strategic objectives involves making a series of complex decisions over time. Therefore manufacturing industry needs improved knowledge about likely impacts of making different types of change in MEs and improved modelling approaches that are capable of providing a systematic way of modelling change impacts in complex business processes; prior to risky and costly change implementation projects. An ability to simulate the execution of process instances is also needed to control, animate and monitor simulated flows of multiple products through business processes; and thereby to assess impacts of dynamic distributions and assignments of multiple resource types during any given time period. Further more this kind of modelling capability needs to be integrated into a single modelling framework so as to improve its flexibility and change coordination. Such a modelling capability and framework should help MEs to achieve successfully business process re-engineering, continuous performance development and enterprise re-design. This thesis reports on the development of new modelling constructs and their innovative application when used together with multiple existing modelling approaches. This enables human and technical resource systems to be described, specified and modelled coherently and explicitly. In turn this has been shown to improve the design of flexible, configurable and re-usable manufacturing resource systems, capable of supporting decision making in agile manufacturing systems. A newly conceived and developed Role-Based Modelling Methodology (R-BMM) was proposed during this research study. Also the R-BMM was implemented and tested by using it together with three existing modelling approaches namely (1) extended Enterprise Modelling, (2) dynamic Causal Loop Diagramming and (3) Discrete Event Simulation Modelling (via software PlantSimulation ®). Thereby these three distinct modelling techniques were deployed in a new and coherent way. The new R-BMM approach to modelling manufacturing systems was designed to facilitate: (1) Graphical Representation (2) Explicit Specification and (3) Implementation Description of Resource systems. Essentially the approach enables a match between suitable human and technical resource systems and well defined models of processes and workflows. Enterprise Modelling is used to explicitly define functional and flexibility competencies that need to be possessed by suitable role holders. Causal Loop Diagramming is used to reason about dependencies between different role attributes. The approach was targeted at the design and application of simulation models that enable relative performance comparisons (such as work throughput, lead-time and process costs) to be made and to show how performance is affected by different role decompositions and resourcing policies. The different modelling techniques are deployed via a stepwise application of the R-BMM approach. Two main case studies were carried out to facilitate methodology testing and methodology development. The chosen case company possessed manufacturing characteristics required to facilitate testing and development; in terms of significant complexity and change with respect to its products and their needed processing structures and resource systems. The first case study was mainly designed to illustrate an application, and benefits arising from application, of the new modelling approach. This provided both qualitative and quantitative results analysis and evaluation. Then with a view to reflecting on modelling methodology testing and to address a wider scope manufacturing problem, the second case study was designed and applied at a different level of abstraction, to further test and verify the suitability and re-usability of the methodology. Through conceiving the new R-BMM approach, to create, analyse and assess the utility of sets of models, this research has proposed and tested enhancements to current means of realising reconfigurable and flexible production systems.