Complex Network–Based Change Propagation Path Optimization in Mechanical Product Development

Change propagation is a process in which a change to one part or element of a developed mechanical product tends to induce additional changes to other elements of the product, and the changes can propagate further iteratively. Different change propagation paths yield different change fulfillment costs (development expenditures, lead times, quality losses, etc.). In this study, we seek an optimal change propagation path based on complex theories, and by using this approach the cost of the change can be minimized. In particular, we introduce a modified Dijkstra algorithm coupled with a complex network described based on two variables: change probability and change impact. The network depicting the components of a mechanical product and their dependencies is directional while considering both node weights and edge weights. The roles of the components such as change absorbers, carriers, and multipliers are identified. Moreover, the loop change propagation paths and “AND” logical relations between sibling components are investigated. The practical application of the proposed models and methods is assessed using an industrial case example of an elevator system, and satisfactory results are obtained. The proposed method can be employed for analyzing different change propagation paths and their optimization in terms of overall costs.