How production planners and plant managers use simulation results to make better decisions
In many companies, simulation is primarily used during the planning phase. New facilities are evaluated, layout options are compared, or material flows are analysed. However, once a decision has been made, the simulation model is often no longer used.
Yet a model can still provide valuable insights even after the project phase. In production controlling in particular, questions regularly arise where simulation can help. Production planners and plant managers must make decisions under conditions of uncertainty, such as fluctuating demand, new products or limited capacity.
Simulation makes it possible to analyse production systems under realistic conditions and examine the effects of changes in advance.
Define relevant key performance indicators before the simulation
Simulation generates a large amount of data. This includes, for example, throughput, utilisation of individual machines, stock levels or waiting times.
To ensure these results are truly useful in production controlling, the key performance indicators to be focused on should be defined before the simulation begins. Only then can different scenarios be meaningfully compared.
Typical key performance indicators in a production environment include, for example:
- Throughput per shift
- Utilisation of critical machines
- Working capital
- Lead times
- On-time delivery
Once these key performance indicators are clearly defined, simulation results can be evaluated in a targeted manner.
Consider fluctuations rather than just average values
In controlling, key performance indicators are often presented as average values. These frequently convey a stable picture of production.
In practice, however, production systems are characterised by fluctuations. Process times vary, disruptions occur, or orders change at short notice. It is precisely this variability that often determines whether a system runs stably or regularly reaches its limits.
Simulation makes these dynamics visible. In addition to average values, peak loads, queues or bottlenecks can also be analysed, for example. This provides a more realistic picture of actual production processes.
Practical example: When average values give a false impression of stability
A practical example illustrates this difference: For a new product line, a controller calculates the expected utilisation of a plant. The calculation yields a utilisation rate of around 85 per cent. At first glance, the system therefore appears stable.
However, when the system is simulated, a different picture emerges. Fluctuations in process times regularly cause queues to form before the bottleneck. Several times a week, significant bottlenecks occur in the material flow.
The reason lies not in the average value itself, but in the fluctuations within the system. Thanks to the simulation, this problem was identified at an early stage. The buffer sizes were adjusted before the line went into operation.
Systematically comparing scenarios
Production planning always involves uncertainties. Demand may rise, machines may break down, or new variants may be introduced.
Simulation enables structured scenario analyses without affecting actual operations. Typical questions include, for example:
- How does rising demand affect capacity?
- What happens if a critical machine fails?
- What consequences does a new product have for existing lines?
Such analyses help to identify risks at an early stage and prepare decisions more effectively.
Combining simulation with real production data
A current trend in industry is the integration of simulation with real production data. Data from MES systems or machines can be used to update models and perform analyses based on real operational data.
Such models are often referred to as digital twins. They enable production systems not only to be planned, but also analysed and optimised whilst in operation.
Using simulation for operational decisions
In day-to-day production, decisions must regularly be made that affect capacities and material flows.
Simulation can help to analyse the consequences of such decisions in advance. Typical use cases include:
- Adjustment of shift models
- Planning of maintenance windows
- Changing material flows
- Adjusting buffer sizes
- Integrating new products
This allows measures to be evaluated before they are implemented in real-world operations.
Using simulation models in the long term
Many simulation models are created for a single project and are not used further afterwards. Yet they can be a valuable analytical tool in the long term.
If a model is maintained and updated, it can be used, for example, for:
- Variant analyses for process changes
- Preparing investment decisions
- Training for production planners
- Continuous process improvement
This transforms a project model into a tool for production controlling that can be used on a long-term basis.
Simulation is frequently used as a tool for planning new production systems. However, its potential does not end with the implementation of a project.
Used correctly, simulation can also make an important contribution to production controlling. It helps to highlight fluctuations, analyse scenarios and prepare decisions on a more informed basis.
Companies that use their simulation models over the long term gain additional transparency regarding their production systems. Instead of working exclusively with average values, they can better understand the dynamics of their processes and control them in a more targeted manner.
