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