The Simulation Series, Part II:
Manufacturing Simulation Types

In Part I of this series, we discussed the importance and benefits of using simulation technology to optimize efficiency and drive innovation in the manufacturing industry.

Now in Part II, we’ll explore the various types of simulations used in manufacturing and how to decide which ones to implement for different industrial challenges.

First: What is Simulation Analysis?

Technically, simulation analysis is a methodology in which a simulation model of physical systems is built. The simulation model is built with consideration of the physical system’s governing factors and constraints; thus, the simulation model can virtually imitate the physical system's behaviour.

Analogically, a simulation model works like a crystal ball – it takes internal and/or external factors as inputs and gives out outputs accordingly. Simulation analysis, then, is like forward time travel – it allows us to see what could happen in the future in defined scenarios. In other words, simulation analysis helps in determining how target variables will be affected based on changes in input variables over time.

Different types of simulation models will define the simulation type. One must understand the varied natures of simulation models in order to harness each one’s full potential.

Types of Simulation Models

Simulation models can be categorised based on various defining factors such as changes in state variables over time, predictability, nature of changes in state variables and level of aggregation. The key categories of simulation models are:

• Dynamic or Static
• Deterministic or Stochastic
• Continuous or Discrete
• Individual or Aggregate

Let's breakdown each one:

Dynamic or Static

Simulation models can be categorised either as dynamic or static based on changes of state variables over time. In a dynamic model, the state variable changes over time whereas a static model is a snapshot at a single point of time. System dynamics, discrete event, and agent-based models are examples of dynamic simulation types whereas Monte Carlo simulation is an example of a static model.

Deterministic or Stochastic

If the behaviour of a simulation model is entirely predictable, then it is called a deterministic model. If the behaviour can’t be entirely predictable, then the model is stochastic. Deterministic models produce the same exact results for a particular set of inputs whereas stochastic models present data and predict outcomes that account for certain levels of unpredictability or randomness.

Continuous or Discrete

The nature of change among a model’s state variables can be described as either continuous or abrupt. In a continuous model, the state variables change in a continuous way whereas in a discrete model the state variables change only at a countable number of points in time. These points in time are the ones at which the event occurs to change the state. System dynamic models are mainly continuous in nature whereas discrete event and agent-based models are mainly discrete in nature.

Individual or Aggregate

A model's level of aggregation defines whether it is an individual-based model or an aggregate model. In an individual-based model, the characteristics of each individual are tracked through time whereas in an aggregate model, the characteristics of population are averaged together, and the model attempts to simulate changes using these averaged characteristics for the whole population. System dynamic models are aggregate models whereas discrete event and agent-based models are mainly individual-based models.

In conclusion, the categorisation of simulation types serves as a helpful guide in deciding when to use which simulation analysis.

Up next: In Part III and Part IV of this series, we’ll examine Discrete Event Simulation and Robotic Simulation, the two most frequently used simulations.

To find out how Actalent can help your business manage its manufacturing system using simulation, contact us.

_References: