optimization and simulation

Blurring Boundaries between Optimization and Simulation Tools

Reading time: 2 mins

Meet the Experts

Key Takeaways

⇨ Optimization and simulation have been used in supply chain analytics for decades.

⇨ These two methodologies sometimes complement each other perfectly in many scenarios.

⇨ Many tools are therefore offering both these methodologies in the same solution.

Blurring Boundaries between Optimization and Simulation Tools

We have used optimization simulation tools in supply chain analytics for a long time.  We will cover these technologies in our December 2022 research report, Supply Chain Analytics and Data Management. We have categorized them into prescriptive analytics methodologies that help you understand the best way to do something. This is not a concrete characterization, however. Suppose your simulation model of your current state, for example, helps you understand which part or element of your process is the process bottleneck. In that case, it can very well be a diagnostic analytics solution. As defined in this article, while categorizing methodologies is useful for understanding analytics capabilities maturity, the key is understanding what we can do leveraging each approach.

We have leveraged optimization solutions in supply chains for decades in many areas, like manufacturing scheduling optimization, inventory optimization, and supply chain network design. These tools primarily leverage linear programming to minimize or maximize a business aim, like minimizing cost or maximizing profit. For example, in inventory optimization, you may ask the model to minimize your total inventory cost subject to certain constraints, like service levels. Some network models maximize profit, allowing the model not to fulfill certain demand nodes. You are trying to answer the question- What is the best way?

Simulation is about finding the answer to “What if” questions. For decades, we have also used simulation models in the supply chain, particularly in warehousing and manufacturing. Consider a manufacturing example again. You have a manufacturing setup and are leveraging an optimization model to understand the optimal number of products you need to manufacture to maximize your profit, given the setup and constraints. But you want to understand the impact of changing certain variables on your manufacturing throughout. Think about finding answers to questions like:

  • How does new equipment impact my manufacturing process?
  • How many additional lines do we need to add to meet production goals?
  • What element of the process is the bottleneck, and what will help eliminate the bottleneck?

As you may already think, it can process some overlapping problems through these advanced analytics approaches. And this is being addressed increasingly by tools available in the market. While traditionally optimization and simulation tools are separate offerings, many vendors now offer products with both functionalities. A big advantage is that you can use one model for both these analytics approaches, significantly saving time and preserving analysis integrity.

More Resources

See All Related Content