Modular System Design
One of the key ways to manage complexity of the modern engineering system is to decompose the system into several distinct modules, enabling shorter module development time, ease of assembly and disassembly (for maintenance), and ease of future upgrade. However, the challenge is to identify optimal modular architecture of the system that can maximize the benefit of the modular architecture, while properly allocating system complexity to each module. Such design can be generated using a matrix based system model (Design Structure Matrix, DSM) and heuristic-based optimization algorithm. It was demonstrated through a case study, where a set of Pareto-optimal set of modular architecture designs for high speed train bogie.
The train bogie system is the chassis part of the high speed train, responsible for supporting the train body. The DSM model of the bogie system was constructed using engineering drawings and CAD models. The DSM model represents the entire bogie system, with current modular decompositions marked by red squares.Using the DSM model, the bogie system is modularized. Metrics for measuring system modularity and system complexity is used for modularization of the sytem. Results show that the modularity of the bogie system is improved, while variation of complexity for bogie modules were minized. Metrics for system modularity and complexity is shown, together with improved modular configuration of the bogie system.
Using the DSM model and optimization algorithm, many different optimal bogie system modular architecture configurations were identified. Figure below shows the plots for several modular architecture configurations (represented by data points), showing the degree of modularity and deviation of average module's complexity value.
Decision makers have range of modular bogie architecture configurations to select from. Firms who have strong integration capability can select 19 module architecture configuration option, while firms who have competent module developing suppliers can select 11 module architecture configuration option. The modular architecture optimization framework allows generation of many Pareto-optimial modular configurations, from which decision makers can select based on their firm's strength.
The train bogie system is the chassis part of the high speed train, responsible for supporting the train body. The DSM model of the bogie system was constructed using engineering drawings and CAD models. The DSM model represents the entire bogie system, with current modular decompositions marked by red squares.Using the DSM model, the bogie system is modularized. Metrics for measuring system modularity and system complexity is used for modularization of the sytem. Results show that the modularity of the bogie system is improved, while variation of complexity for bogie modules were minized. Metrics for system modularity and complexity is shown, together with improved modular configuration of the bogie system.
Using the DSM model and optimization algorithm, many different optimal bogie system modular architecture configurations were identified. Figure below shows the plots for several modular architecture configurations (represented by data points), showing the degree of modularity and deviation of average module's complexity value.
Decision makers have range of modular bogie architecture configurations to select from. Firms who have strong integration capability can select 19 module architecture configuration option, while firms who have competent module developing suppliers can select 11 module architecture configuration option. The modular architecture optimization framework allows generation of many Pareto-optimial modular configurations, from which decision makers can select based on their firm's strength.
Related Papers
