Stop ‘guesstimating’ and make informed decisions with our supply chain simulation solutions
A simulation model aims to replicate the way a system or process works. We usually build supply chain simulation models to gain a better understanding of how the supply chain works and how its performance or outcome can be affected by random events, uncertain external factors or changes in policies, configuration or flows.
Simulation is a powerful technique, ideal in situations where we want to analyze the relationship between key variables and their impact on the system or supply chain we want to analyze. A supply chain simulation will model the probability of different scenarios and outcomes, allowing us to predict the behavior and performance of complex operations as well as quantifying risk factors.
Unlike their optimization ‘cousins’, supply chain simulation models do not optimize nor recommend optimal strategies or decisions. What they do is predict the outcome resulting from these strategies and decisions. In other words, they help us understand the “What Is”, and then predict the “What If”.
With our partners at SDI we have developed and deployed numerous supply chain simulation models and applications covering different topics, such as:
- Plant to warehouse replenishment operations
- The production scheduling of high speed processing lines
- Forecasting bulk storage capacity requirements in an expanded chemical manufacturing facility
- Modeling plant and warehouse receiving and shipping operations
- Simulating supply chains
Our tool of choice is Flexsim, one of the premier simulation software platforms in the market.
By definition, a simulation model must be built to reflect the specific characteristics of the operation it is aiming to study. This fits right in with our customized approach to modeling and application development.
So… Should I Simulate? Or Optimize?
Simulation and optimization are both very powerful analytical techniques, but completely different in their approach and objectives.
The goal of an optimization model is to solve scenarios – i.e. identify the best solution given a set of dependencies, rules and restrictions. A simulation model, conversely, does not identify or recommend solutions. Its goal is to model the temporal behaviour of an operation and predict its performance.
Lets look at a specific example. Lets assume we have a fleet of cargo vessels servicing dozens of ports, and we want to ensure we do this in a cost-effective way.
We would first develop an optimization model to optimize the routes and the allocation of individual vessels to them. This model would have to incorporate all relevant business rules (such as service targets), operational restrictions (such as the capacity and speed of vessels) and costs for its output to be meaningful.
Once we have identified what seems to be an optimal solution, we could build a simulation model to predict its performance. In this model we would want to incorporate random or external events (such as delays at port due to bad weather conditions) that we know impact performance in the real world.
Recommends what to do
Predicts what will happen