We asked five of our own Simul8ers for their number one tip when it comes to creating successful simulations. Here they are…

Identify the right question

Focusing on a specific outcome is how you will get the most out of your simulation. It all starts from identifying the question you want to answer. This is key because it keeps you and your client focused on the objective and it guides every other aspect of how you build the simulation. You start with the question, set out how you measure it, then you identify the levers that influence it and what you can change. This gives you the scope to start building your simulation and a point to return to at every stage of the project to make sure that you don’t get sidetracked.

Keep it simple

Instead of trying to map out an entire organization or department, think of the smaller, individual gains that are going to make an immediate difference and focus on those. Strip a process right back to its elements to set your scope as narrow as possible. Then, only add in extra detail if it will help answer the question you identified.

For example, if you’re trying to model an ER department to understand the number of beds needed, then you need to know that beds are kept on hold whilst a patient goes to x-ray, but you don’t need to model the whole x-ray department. The simpler you can make your simulation, the simpler it will be to analyze and see what’s impacting the process. Do this and you will find it much easier to reach 100% accuracy with your simulations.

Don’t be restricted by data

A common misconception is to think step one of a simulation project is ‘data collection’.  There are two important parts to this. First of all, the use of data should be woven throughout a project and should never be a blocker to getting started if you think you don’t have enough data readily collectable. And this leads to the second point – that there are so many ways to build an accurate simulation, and they don’t depend on having vast amounts of data! 

In fact, simulations are not reliant on recorded stats and numbers at all. It’s perfectly possible to build a model with only very simple measurements collected from those involved in the process. It’s too easy to get sucked into a massive data collection and cleanse exercise before you even start. This won’t only delay your project, you could end up with analysis paralysis from looking at too much data! It all comes back to your question – collect just enough data that’s needed to answer it.

Make it iterative

The best way to make full use of simulation is to make it iterative. This means using the “build, measure, learn” loop. So, you build your sim to answer your question, run your sim, look at your results, learn something about your process, and then you set the next question. You’ll either need to adapt the sim to answer that next question, or potentially build another sim if you need to look at it differently. 

Sometimes building a series of simple simulations is the quickest way to understand your process. Once you’ve understood your process and are moving onto experimenting with set scenarios, this is the time to start thinking about ways to make your simulation adaptable with a user interface. Think of these early sims as a scoping exercise, or the minimum viable product, and then move onto creating a more versatile decision tool.

Make it visual

Simulations are an important way to gain better visibility of what is actually going on in a process, and then it’s easier to spot the solution. The beauty is that they lift data and information from spreadsheets or inside people’s heads and turn them into a tangible, relatable picture that brings a problem and its solution to life. The more visual the better. Not only does a visual simulation help to find the solution quicker, we find that being able to present business processes in this way creates an instant hit with the various different stakeholders involved. Senior management that are removed from the day-to-day but want to quickly understand a proposal for change appreciate it, as do those involved at ground level that will feel the impact of improvements. 

And for a final thought, remember that simulations are there for trial and error, so the best approach in any case is to be brave and play around with them as much as possible!