from ciw_model import (Experiment,
multiple_replications)Using the model
The code for the simulation model is contained in the ciw_model.py module. The module contains wrapper functions to allow easy creation of new experiments and execution of multiple replications.
A user only needs to import the Experiment class and the multiple_replications function.
Imports
Default run of the model
To perform a default of the model, create an instance of Experiment and pass it to the multiple_replications function. By default the model is run for 5 replications. The function results a pandas.DataFrame containing replications and performance measures.
default_experiment = Experiment()
results = multiple_replications(default_experiment)
results| 01_mean_waiting_time | 02_operator_util | 03_mean_nurse_waiting_time | 04_nurse_util | |
|---|---|---|---|---|
| rep | ||||
| 1 | 2.784746 | 91.374255 | 47.927266 | 96.753369 |
| 2 | 1.367329 | 91.853442 | 31.935974 | 97.776217 |
| 3 | 4.009524 | 95.925268 | 79.743715 | 98.214845 |
| 4 | 1.519486 | 89.297980 | 46.554721 | 96.708677 |
| 5 | 3.183769 | 95.467289 | 55.718766 | 97.688507 |
To summarise results call the .describe() method of the results dataframe.
The code below also rounds to 1 decimal place and transposes the dataframe
results.describe().round(1).T| count | mean | std | min | 25% | 50% | 75% | max | |
|---|---|---|---|---|---|---|---|---|
| 01_mean_waiting_time | 5.0 | 2.6 | 1.1 | 1.4 | 1.5 | 2.8 | 3.2 | 4.0 |
| 02_operator_util | 5.0 | 92.8 | 2.8 | 89.3 | 91.4 | 91.9 | 95.5 | 95.9 |
| 03_mean_nurse_waiting_time | 5.0 | 52.4 | 17.5 | 31.9 | 46.6 | 47.9 | 55.7 | 79.7 |
| 04_nurse_util | 5.0 | 97.4 | 0.7 | 96.7 | 96.8 | 97.7 | 97.8 | 98.2 |
Creating new scenarios
The code below changes the number call operators, advanced nurse practitioners and chance of a call back from a nurse.
what_if_experiment = Experiment(n_operators=14, n_nurses=10, chance_callback=0.45)
what_if_results = multiple_replications(what_if_experiment)
what_if_results.describe().round(1).T| count | mean | std | min | 25% | 50% | 75% | max | |
|---|---|---|---|---|---|---|---|---|
| 01_mean_waiting_time | 5.0 | 1.2 | 0.4 | 0.5 | 1.1 | 1.2 | 1.5 | 1.7 |
| 02_operator_util | 5.0 | 86.5 | 2.6 | 82.4 | 85.8 | 86.8 | 88.3 | 89.2 |
| 03_mean_nurse_waiting_time | 5.0 | 55.1 | 19.0 | 22.4 | 58.1 | 58.7 | 66.5 | 70.0 |
| 04_nurse_util | 5.0 | 97.3 | 0.5 | 96.8 | 96.9 | 97.6 | 97.7 | 97.7 |
Running more replications
By default the multiple_replications function runs 5 replications. To change the default se the n_reps parameter. The example below runs 50 replications of the model.
default_experiment = Experiment()
results = multiple_replications(default_experiment, n_reps=50)
results.describe().round(1).T| count | mean | std | min | 25% | 50% | 75% | max | |
|---|---|---|---|---|---|---|---|---|
| 01_mean_waiting_time | 50.0 | 3.5 | 1.9 | 1.5 | 2.3 | 2.9 | 4.0 | 11.0 |
| 02_operator_util | 50.0 | 92.9 | 2.3 | 86.8 | 91.9 | 92.8 | 94.7 | 97.0 |
| 03_mean_nurse_waiting_time | 50.0 | 48.5 | 19.4 | 14.5 | 36.3 | 44.7 | 61.9 | 95.0 |
| 04_nurse_util | 50.0 | 97.1 | 1.0 | 94.1 | 96.8 | 97.2 | 97.7 | 98.3 |