Get Started

Monte Carlo sampling for a discrete-event model with SimPy

Now let's build the same Monte-Carlo sampling analysis using a SimPy version of the model. The SimPy model has a generator named manufacturing_process, which simulates different processes, and a function called run_monte_carlo that runs the model multiple times, storing the information in a NumPy array named time_record.

The code that plots the results is similar to the one used in the previous exercise, but it has been moved to a function named plot_results() that is shown below.

def plot_results():

    df_disc = pd.DataFrame({cNam[0]: process_line_space, cNam[1]: time_record})
    fig = sns.lineplot(data=df_disc, x=cNam[0], y=cNam[1], marker="o")
    fig.set(xlim=(0, len(processes) + 1))
    plt.plot()

The Monte-Carlo sampling loop will produce a series of possible process trajectories, as shown in the figure. Monte Carlo trajectories for different process scenario.

This is a part of the course

“Discrete Event Simulation in Python”

View Course

Exercise instructions

  • Clock-in and yield process_duration.
  • Save the current time in time_record.
  • Run a for-loop for n_trajectories samples with dummy variable t.
  • Create the SimPy environment, add processes and run the model.

Hands-on interactive exercise

Have a go at this exercise by completing this sample code.

def manufacturing_process(env):
    global time_record
    for p in range(len(processes)):
        proc_p = processes[p]
        process_duration = random.gauss(proc_p["Average_Duration"], proc_p["Standard_Deviation"])

        # Clock-in and yield the process_duration
        yield ____

        # Save the current time in time_record
        time_record[p + 1] = ____

def run_monte_carlo(n_trajectories):

    # Run a for-loop for n_trajectories samples with dummy variable t
    ____

        # Create the SimPy environment, add processes and run the model
        env = ____
        env.____(manufacturing_process(env))
        env.____()
        
        plot_results()
    plt.show()

run_monte_carlo(n_trajectories = 100)
Edit and Run Code

This exercise is part of the course

Discrete Event Simulation in Python

AdvancedSkill Level
4.3+
6 reviews

Discover the power of discrete-event simulation in optimizing your business processes. Learn to develop digital twins using Python's SimPy package.

Start Course for Free

You’ll learn optimization methods to maximize the impact of your discrete-event models. You’ll learn how to perform simulation ensembles using Monte Carlo approaches and discover how to identify clusters in your model results to help you understand its behavior and identify critical processes and tipping points. You’ll also use objective functions to set targets for your model optimization efforts. To end this course, you’ll explore how to make your model scalable so that it can grow stable and in a controlled manner.

Exercise 1: Simulation ensembles: Monte-Carlo samplingExercise 2: Monte Carlo sampling for discrete-event modelsExercise 3: Monte Carlo sampling for a discrete-event model with SimPy
Exercise 4: Clustering and cluster modelsExercise 5: Logistics eCommerce model: Analyzing resultsExercise 6: Logistics eCommerce model: k-means analysisExercise 7: Objective functions and system optimizationExercise 8: Manufacturing Optimization: Search & StopExercise 9: Manufacturing Optimization: Score & RankExercise 10: Model modularity to optimize continuous developmentExercise 11: Logistics eCommerce model: Model modularityExercise 12: Garment Production: Multi-processes and modularityExercise 13: Congratulations!

What is DataCamp?

Learn the data skills you need online at your own pace—from non-coding essentials to data science and machine learning.

Start Learning for Free