Grant: Discovery Grants Program - Individual

The manufacturing industry represents the backbone of the economy, providing products and employment to millions in Canada. The manufacturing sector faces increased pressure to deliver goods in a short time. However, unexpected events on the shop floor (e.g., machine breakdown, quality issues) and in the supply chains (e.g., COVID-19) have shown the vulnerability of manufacturing operations. To address customer requirements, advanced manufacturing systems (AMS) (e.g., flexible and reconfigurable manufacturing systems) have been proposed to provide flexibility and responsiveness. Although these systems possess adaptation capabilities, they lack the ability to interact with the manufacturing environment. Nowadays, with the possibility of gathering data in real-time, AMS have the opportunity for data collection, analysis, prediction, learning, and executing decisions. As a result, a new era of smart manufacturing systems will be possible with the introduction of data-driven solution approaches. These new systems will be able to control the dynamic events in the manufacturing shop floors and supply chains. Hence, the design and implementation of smart decision-making tools are imperative to provide a rapid response to tackle system uncertainties. While responsiveness is critical, the consideration of environmental sustainability is equally relevant, as the manufacturing industry produces high CO2 emissions. Thus, smart solutions could aid in reducing such emissions. Manufacturing functions are affected by dynamic events causing scheduling, layout configuration, materials, and process planning problems. Traditionally, they are studied independently and in a static manner. However, the interaction between the different functions and the dynamic environment calls for their simultaneous analysis to achieve an optimized system. The current advances in industry 4.0 (e.g., big data analytics, digital twins) can be used to tackle those problems holistically. For instance, a virtual representation of the manufacturing system could be used to experiment and observe the impacts of possible strategies. Moreover, with advanced data analytic tools, real-time data analysis and optimization can be performed in the virtual world. Consequently, the strategies implemented (e.g., rescheduling, layout change, process plan adaptation) in the physical system are already assessed for feasibility and optimality. Therefore, the proposed research program will support the manufacturing industry and communities by providing smart decision-making tools to efficiently navigate the dynamic manufacturing environment. The tools will be designed with the integration of robust methodologies (e.g., mathematical programming, simulation, data analytics) to help AMS self-adjust and self-optimize in the quest for autonomous manufacturing. This research will have significant outcomes, such as saving costs and protecting Canada's environment (reducing energy consumption and emissions).