The path to the simulation of digital manufacturing systems
Manufacturing systems are achieving unprecedented levels of complexity, which arises from the need to manage a significant set of conflicting objectives. At the supply chain level, these systems must be capable of performing relevant and fast scale-ups of the production volume and having flexible processes. Factories must run efficiently with large product diversity in their production lines, while still guaranteeing the output levels. To be sustainable, production lines need to reach the maximum utilization of the available resources, while at the same time need to keep under control the delivery times.
The manufacturing industry strives then for tools that are capable to quantify, demonstrate, and find the sweet spot of running the processes, production lines, plants, and supply chains. The immediate objective is to smooth out the running systems as much as possible, even when the systems are running under severe conditions. Although this goal appears to be contradictory, emerging manufacturing paradigms, as is the Industry 4.0, support the development and application of tools focused on improving output and quality and reducing downtimes. While this path is generally well seen, there is still work to do to estimate the extension of such benefits in the existing systems. To accomplish this, data-driven methods, such as simulation, take advantage of data to learn how to make the best decisions. The video below shows how simulation can be used to model a complete manufacturing site.
Simulation is a powerful method that can provide significant levels of detail and flexibility in modelling systems. Although building and validating simulation models tend to be complex and often time-consuming tasks, simulation provide the means to understand large-scale problems and gain insights into key tradeoffs for decision-making. Several advantages can be pointed out to simulation. Simulation can model the variability and its effects in a quite straightforward way. If the system being modelled is subject to significant levels of variability, simulation is often the only technique for accurately modelling its performance. Simulation can handle easily different modelling scales. A model can include both relaxed or more restrictive assumptions. Simulation is also appealing and most of the times supplies animated displays of the system, improving the understanding and confidence of the model.
Simulation provides the means to fully realize the value of digital manufacturing systems. Here, a key observation is the merging of the physical processes with the digital world. In this environment, most of the problems to solve involve optimizing systems that are running in their design limits. This denotes that simulation models must generate solutions for systems that have already very challenging performance indicators. Therefore, to demonstrate real improvements and to search for disruptive solutions, simulation must enlarge the traditional boundaries of the systems analysis.
New approaches for simulation will leverage best practices to engineer factories and production processes at the earlier stages of their development. This can be accomplished by using different models to investigate a broad spectrum of complex issues related to the design of the manufacturing system. The development of digital prototypes of whole supply chains, factories or processes will simulate real-world conditions and study their response. And, finally, the digital twin of the working system will reflect the current condition of the actual system and changes during operation.
Samuel Moniz (samuel.moniz@dem.uc.pt)
University of Coimbra, Industrial Engineering and Management
Cristóvão Silva
University of Coimbra, Industrial Engineering and Management
References
1. Narciso, C., et al. (2019). A Simulation Approach for Spare Parts Supply Chain Management. Proceedings of the International Conference on Industrial Engineering and Operations Management.
2. Marques, C. M., et al. (2018). “Strategic decision-making in the pharmaceutical industry: A unified decision-making framework.” Computers & Chemical Engineering 119: 171–189.
3. Vieira, Miguel, et al. “Simulation-Optimization Approach for the Decision-Support on the Planning and Scheduling of Automated Assembly Lines.” 2018 13th APCA International Conference on Control and Soft Computing (CONTROLO). IEEE, 2018.
