Using Digital Twins to Increase Operational Output by 15%: Part Two
Four Challenges to Consider When Pursuing a Simulation Initiative
Discrete event simulation (DES) has undoubtedly stood the test of time as a powerful tool for testing material and production flow within manufacturing operations – increasing annual production and eliminating unnecessary inventory by millions of dollars.
With its longstanding presence in the industry, one might wonder why we are still discussing DES. However, our experience working with clients has revealed that there are four pertinent challenges that manufacturers must consider before embarking on their journey. In this article, we delve into these challenges, shedding light on their implications in the present day. Furthermore, we provide our perspectives on how and if these challenges are being addressed, while highlighting strategies that manufacturers can adopt to maximize the value generated through the utilization of simulation technologies.
Challenge 1: Operational Simulation Has Inherent Drawbacks
Operational simulation, although highly effective in providing real-time feedback, comes with inherent drawbacks that pose challenges for engineering teams.
One major hurdle lies in the complexity that arises when there are numerous design choices to consider. In such cases, executing simulations becomes arduous, requiring engineers to resort to conventional methods to narrow down the options for factory layouts and designs.
This involves relying on experience, historical designs, and external input from third parties to generate a manageable set of designs that can then be simulated and validated.
As a potential solution, there is an ongoing exploration of reinforcement learning capabilities to automate the design generation process. The idea is to combine these capabilities with digital twins, enabling the iterative and autonomous refinement of factory layouts. However, it's crucial to acknowledge that such approaches are in their early stages of development and implementation. Similar solutions, referred to as Expert Systems, have shown promise in industry applications, but significant advancements and refinement are required to realize their potential fully. Therefore, for the foreseeable future, human expertise and involvement will remain critical parts of the design and simulation processes.
Challenge 2: Complex Engineering Systems Require Diverse Expertise to Model
Although simulation can help in optimizing manufacturing systems, it requires a multidisciplinary approach that goes beyond relying on a single modeling expert. Successful factory simulation projects necessitate the involvement of a diverse set of skills and expertise, including data engineering, simulation engineering, domain-specific knowledge, and statistical analysis. Each discipline brings unique perspectives and insights, contributing to a comprehensive understanding of the manufacturing processes and their dynamics.
However, assembling and coordinating such a multidisciplinary team of experts can be a challenge. Acquiring and retaining individuals with diverse skill sets within a single organization can be resource-intensive and time-consuming.
As a result, many manufacturers turn to collaboration with experienced third parties to kick-start their simulation initiatives. Collaborating with external partners offers access to a broader range of skill sets, ensuring that all crucial aspects of the simulation process are addressed. This approach provides companies with a fast start in generating value as well as time and support in building their internal capabilities.
Challenge 3: Maintenance of Simulation Models Can Be Tedious
There are many out-of-the-box simulation tools in the market that are well-known within the community, including Simio, AutoMod, FlexSim and Rockwell Automation’s Arena. These tools are incredibly powerful, and the market for these tools is still growing between 6-7% annually, according to ARC Advisory Group.
However, if the simulation tool is not adequately integrated with other systems (e.g., CAE, PLM, MES), the practical method for maintaining the simulation models is to rely on manual updates by the engineer.
As engineering teams continue to make improvements and modifications to the real-world production lines, the accuracy of the simulation model gradually declines. Maintaining and updating these models is often perceived as tedious and frequently left incomplete.
To address this challenge, special attention should be given to the integration of the simulation tool with upstream and downstream systems. Seamless integration enables automated updates to the simulation model, ensuring its accuracy and usefulness as real-world changes occur. By addressing the integration and maintenance challenges associated with simulation tools, manufacturers can ensure that their simulation models remain accurate, up-to-date, and scalable, ultimately maximizing the value derived from simulation technologies.
Challenge 4: The Fidelity of Traditional Simulation Is Limited
Traditional operational simulation tools offer the advantage of speed and the ability to simulate long periods of production, providing valuable insights into the long-term effects of factory designs. However, these tools have limitations in terms of fidelity compared to other technologies.
DES tools, while fast and useful for testing material flow, rely on certain assumptions that can be difficult to validate.
For example, DES tools cannot verify if conveyor speed will cause bottles to topple over on a turn without considering physics-based properties like mass, center of gravity, and acceleration.
To address this challenge, integrating physics-based simulation technology with conventional DES tools becomes essential. By combining these approaches, manufacturers can achieve a more comprehensive and accurate assessment of their manufacturing processes. Physics-based simulation provides a deeper understanding of the physical interactions between equipment and materials, allowing engineers to evaluate and optimize system behavior with a higher level of fidelity.
The integration of physics-based simulation technology complements traditional DES tools, enhancing the overall simulation capabilities. This integration enables manufacturers to overcome the limitations of traditional simulation and gain a more realistic understanding of their systems, leading to improved decision-making and optimization of manufacturing processes.
Stay tuned for Part Three where we will explain how to complement traditional simulation with physics-based simulation.
Authors
