Simulation capabilities are critical for operations, as well as design
Product changeovers that join the physical with the virtual
Industry 4.0 is a term that describes data-driven, artificial intelligence (AI)-powered, networked “smart factories” as the harbingers of the fourth industrial revolution. One key aspect of data-driven operations is using simulation as a tool to optimize production, reduce downtime risks, and deliver the highest efficiency possible in industrial settings.
Simulation as part of the design process is standard across many industries when planning new plants, production lines or individual machines. Simulation ensures that designs will operate as intended across mechanical, electrical and automation functions. After commissioning, these simulations are often not used in operations. Though simulation and modeling offer substantial cost savings for design and startup, reusing their predictive results in operations also brings real value.
A comprehensive, closed-loop, calibrated digital twin of a unit, line or whole plant delivers ongoing, daily predictive insights that simulate a whole plant or process. Reaching beyond design, simulation can augment production planning & process optimization. What is a calibrated digital twin, and why isn’t it yet standard as simulation is in the design process?
It’s the twins
For digital-twin accuracy we need to close the loop with the physical equipment by feeding data from the physical world into the virtual model. The expected result is a calibrated digital twin. This calibrated simulation is essential to trust the predictions and recommendations it provides, such as on associate assignment, machine utilization or material-handling bottlenecks.
A calibrated digital twin updated with real-time data is always current and capable of delivering critical analysis on units, production lines and whole facilities. Accurate, real-time data is the key that unlocks the ability to use simulation as a predictive tool in operations. Reliable predictions can’t be made with outdated data, leading many manufacturers to only use simulations as a design tool before operations are up and running.
Optimizing daily process planning and predicting maintenance needs are core aspects of ensuring maximum uptime and productivity. With real-time data fed into evolving simulations that encompass the entirety of the process or facility, operations managers can evaluate the needs and have forward visibility into equipment and process performance. Simulation of the next shift or day with real-time data helps reduce downtime risk and ensure processes meet production goals.
Similar to the physics simulations and modeling done as part of design, simulation for optimizing and predicting manufacturing process outcomes accurately is a data-driven task. The ability of simulations to accurately predict machine operation requires the simulation to be calibrated with accurate, up-to-date data.
The data is sent directly from machines, processed in edge-computing devices, stored dynamically on-site or in the cloud, and then fed into simulations in real-time to deliver the most accurate predictions. This output helps decision-makers identify safety issues, optimize maintenance intervals and procedures and reduce risk.
Simulation can optimize the manufacture of multiple products on a single line. Using an existing line to produce new versions of a product leads to higher efficiency and productivity, without the need for new machines. The downside is the possible downtime or faults possible when lines are switched from one product to the next.
One good example are polymer spinning machines. When switching from a polypropylene-cored product to a polyester-cored product, there are specific purge times and uptake times needed before the machine can switch products. If not properly purged, the machine will fail due to a mixing of raw materials. If the time and processes to switch raw materials are not optimized, then manufacturers lose valuable production time. Dynamic simulation using real-time data can help reduce the risks inherent to switching products at need. As more manufacturers implement predictive simulation, it will become difficult to compete without it.
Best Practices for scaling
For manufacturers implementing digitalization and simulation solutions, consider the following:
- Identify where downtime risks are greatest to focus on implementing simulation first.
- Change management, process management and training are areas where use of simulation provides substantial benefits.
- Real-time data is crucial. Without up-to-date data, accurate predictions are impossible.
- Understand how production data is delivered and stored. That data needs to be fed into simulations in real-time to calibrate a digital twin and power simulations.
- Choose a simulation solution that has a user-interface allowing real-time predictions by operators and managers, and not just analysts. This allows simulation to be used at every level of design, process, production, and optimization — every day — not just when an expert is on hand.
- Ensure the solution chosen delivers flexibility and open customization to provide insight and optimizations for unique applications.
Manufacturers need to develop simulation standards for their manufacturing processes. These standards would describe production in four distinct digital-twin levels:
- Unit digital twin: Optimizes machines, robots and associates while prioritizing their safety and comfort.
- Process digital twin: Builds on the work done at the unit level but adds logistics flows to minimize waste.
- Facility digital twin: Extends the efficiency model to building heating/cooling, and implements machine stop/start algorithms in coordination with operations.
- Manufacturing network digital twin: Connects facilities providing context, transparency, and insight.
As Industry 4.0 technologies become standard, manufacturers must keep abreast of new technologies to stay competitive. There are many approaches to simulation, and many solutions that offer some level of modeling. Niche, disconnected simulations left to gather dust after the design phase are being supplanted by live, customizable, dynamic simulations that use real-time data to offer insights into production needs every day. Predictive simulations form the basis for manufacturers to reduce risk and deliver competitive advantage into the future.