Simulating information flow to assist building design management
2010-11-05T16:15:48Z (GMT) by
The design of modern buildings has become an increasingly complex activity. This is because of greater demands by Clients in terms of performance, quality, economy and time. These demands coupled with the complex iterative nature of design have resulted in increasing challenges in building design and in the management of the design process. The design process is information driven. Initial research by the writer showed that the main difficulties encountered during the management of the design process are information related. Information transfer and communication issues have been identified as key factors in the successful management of the process. It was concluded that current planning techniques are ill-suited for planning, monitoring and controlling building design because they neither accommodate the iterative nature of design nor permit the choice of alternatives. This research sought to develop better tools to aid design managers in improving the management of the process. Although all phases of the design process were examined, the main focus of this research was the Conceptual and Schematic design stages. To investigate these stages a generic data flow model was developed using the structured analysis diagramming technique of Data Flow Diagrams. The model was based on data from preliminary case studies and was validated by interviews with construction industry professionals. Industry feedback showed that improved management of the design process should not only include better techniques for planning and scheduling but also allow design managers to investigate the iterations between design tasks and predict the effects of different scenarios. Matrix partitioning techniques were used to identify ioops of iterative design tasks in the data flow model. A Discrete Event Simulation Model was developed to predict the effects of different scenarios. This model was based on data from the Data Flow Model and the identified iterative design loops. In addition, dynamic factors input by the user such as the durations and resources of the design tasks allowed the examination of the effects of different scenarios of information related criteria. These criteria were identified from industry survey and interviews. The simulation model was rigorously tested and validated through subsequent case studies and review by industry practitioners. The thesis concludes that the use of Data Flow Modelling in conjunction with Matrix Analysis and Discrete Event Simulation techniques provides a powerful tool for assessing the impact of change within the design process and could form the basis for managing and planning multi-disciplinary design work.