The digital twin is a powerful Industry 4.0 capability that is transforming manufacturing operations. The digital twin is a digital representation of a physical system, which allows for data monitoring and analysis. It is a simulation tool that creates a digital model and enables the user to anticipate problems, prevent downtime, uncover new possibilities, and plan for the future, without impacting the physical production environment, the physical twin. For example, digital twin technology is utilized in automotive production environments to improve the performance of production lines. The digital twin allows planners to test different scenarios and optimize line performance to maximize throughput, efficiency, and resource utilization.
The digital twin requires the same inputs that the physical twin requires, in order to simulate the effect of those inputs and one of the most important inputs is the production schedule. Unfortunately, in production environments, where a large variety of products are produced in large volumes, the complexity of the operation makes scheduling very challenging and as a result it is not possible to generate high quality schedules through simulation means alone.
The Limitations of Simulation
Note that a simulation tool is an analysis tool: given a set of inputs, it can predict the outcome on the plant floor. It is not a synthesis tool: it is not capable of generating a schedule. A brute force approach that simulates the outcomes from a large number of candidate schedules and selects the best one is simply not feasible in practice. However, if we start with a low quality schedule as an input to the digital twin, we severely limit the ability of the digital twin technology to proactively identify and fix inefficiencies in the production environment.
In contrast, a modern planning & scheduling system, such as Optessa MLS, is a synthesis tool, and excels in generating high quality schedules rapidly. Optessa MLS represents Planning & Scheduling 4.0, P&S 4.0, the next generation of planning and scheduling tools required to realize the full benefits of Industry 4.0. It makes sense therefore to use an Industry 4.0 planning and scheduling tool capable of generating a high quality, holistic production schedule that will drive not only the physical twin, but also the digital twin. To realistically simulate the operation of the physical twin, the digital twin must also be integrated with the planning & scheduling system.
P&S 4.0 Unlocks the Power of the Digital Twin
When the digital twin is integrated with an effective P&S 4.0 tool, powerful capabilities are unlocked. P&S 4.0 tools are capable of generating plans that consider all requirements and constraints of the system, across the entire planning horizon. This is critical because ignoring some constraints and requirements will lead to plans which are far from optimal. This technology stack is now able to run “what-if” scenarios/ simulations in the virtual model, before implementing the best possible option on the physical twin.
Additionally, the digital twin can be used in conjunction with a scheduling tool to draw inferences on important long term investments such as capacity expansion (CapEx). In some cases, better planning and scheduling can avoid, or at the very least, delay the need for capital expenditure.
Many manufacturing facilities run simulations to determine the need for CapEx. A digital twin integrated with a P&S 4.0 tool ensures that the digital twin has the high quality schedule input required to run an accurate simulation and allow the business to make more informed decisions on CapEx.
Case Study – Unlocking Capacity to Avoid Capital Expenditure:
A large auto Original Equiment Manufacturer (OEM) runs a 2-model assembly line. The production facility is flexible enough to handle small cars and large Sports Utility Vehicles (SUVs) on the same line. However, to balance the line, and keep the throughput high, a highly optimized schedule is necessary. The same schedule also drives the upstream nodes such as the paint shop and weld shop.
In order to react to an increase in demand for a new model, the automaker needed to add a third model on the same assembly line. While the final assembly line, and the weld shops were determined to be flexible enough to handle the added complexity, the paint shop was identified as a bottle neck in maintaining throughput. The current paint shop was handling up to 17 color changes per day. Each color change required production to stop for a significant period of time while the pain lines were flushed to remove the old pain and the new pain was primed in the system.
Adding the third model would result in more color changes in the paint shop, resulting in a lower throughput. It was estimated that based on the model-mix, there would be roughly 25 color changes per day after the addition of the third model. So the automaker was contemplating the expansion of the paint shop to maintain throughput.
P&S 4.0 Allows Auto Manufacturer to Run What-If Simulations
The automaker used a simulation integrated with Optessa MLS to run “what-if” scenarios to create different variations of assembly line schedules using the three-model data. All possible constraints and restrictions were imposed on the schedule. Through testing on various data sets, it was determined that the average number of color changes could be restricted to 19 per day, even after the addition of the third model. This was a sharp drop from the estimated 25 changes per day. Based on this data, and an analysis of the paint shop capability, it was determined that 19 color changes per day could be handled, without having any large adverse effects on the final assembly and weld shop throughput.
The “what-if” scenarios also allowed the automaker to determine the number of color changes per day at which the plant throughput would be affected. By projecting possible changes in the model-mix based on sales trends, the automaker could forecast that in case of more complex model-mixes in production orders, even an optimized schedule would result in 23 changes per day. At this point, the throughput of the Paint shop would fall below the required throughput, which meant there would be a need for additional capacity.
This gave the automaker two important results:
- There was no need for an immediate addition to capacity.
- By managing model-mixes between different plants that built the newly added model, the need for added capacity could be avoided, as far as the sales data supported such an optimal distribution between different plants building the same model.
Are you ready to take your manufacturing operation to the next level of efficiency?
Start by implementing smart planning and scheduling software from Optessa. Contact us for a live demo today.