utils.event_log_from_ciw_recs

utils.event_log_from_ciw_recs(ciw_recs_obj, node_name_list)

Given the ciw recs object, return a dataframe in the format expected by the vidigi functions - reshape_for_animation OR - animate_activity_log

Parameters

Name	Type	Description	Default
ciw_recs_obj		The output of the .get_all_records() method run on the ciw simulation object. This is a list of named tuples. See https://ciw.readthedocs.io/en/latest/Tutorial/GettingStarted/part_3.html and https://ciw.readthedocs.io/en/latest/Reference/results.html for more details.	required
node_name_list		User-defined list of strings where each string relates to the resource or activity that will take place at that ciw node	required

Returns

Name	Type	Description
	pd.DataFrame

Notes

Given the ciw recs object, if we know the nodes and what they relate to, we can build up a picture the arrival date for the first tuple for a given user ID is the arrival

Then, for each node: - the arrival date for a given node is when they start queueing - the service start date is when they stop queueing - the service start date is when they begin using the resource - the service end date is when the resource use ends - the server ID is the equivalent of a simpy resource use ID

A more complex multi-node example can be found in https://github.com/Bergam0t/ciw-example-animation in the files - ciw_model.py - vidigi_experiments.py

Examples

Example taken from https://ciw.readthedocs.io/en/latest/Tutorial/GettingStarted/part_3.html

Let us interpret the servers as workers at aa bank, who can see one customer at a time

import ciw

N = ciw.create_network( arrival_distributions=[ciw.dists.Exponential(rate=0.2)], service_distributions=[ciw.dists.Exponential(rate=0.1)], number_of_servers=[3] )

ciw.seed(1)

Q = ciw.Simulation(N)

Q.simulate_until_max_time(1440)

recs = Q.get_all_records()

event_log_from_ciw_recs(ciw_recs_obj=recs, node_name_list=[“bank_server”])