number_of_selfloops: Metric: Number of selfloops in trace
In gertjanssenswillen/edeaR: Exploratory and Descriptive Event-Based Data Analysis

Description Usage Arguments Details Methods (by class) References

Provides information statistics on the number of selfloops.

number_of_selfloops(eventlog, type, level, append, ...)

## S3 method for class 'eventlog'
number_of_selfloops(eventlog, type = c("all",
  "repeat", "redo"), level = c("log", "case", "activity", "resource",
  "resource-activity"), append = FALSE, append_column = NULL,
  sort = TRUE, ...)

## S3 method for class 'grouped_eventlog'
number_of_selfloops(eventlog, type = c("all",
  "repeat", "redo"), level = c("log", "case", "activity", "resource",
  "resource-activity"), append = FALSE, append_column = NULL,
  sort = TRUE, ...)

`eventlog`	The dataset to be used. Should be a (grouped) eventlog object. `eventlog`.
`type`	The type of repetitions, either all, repeat or redo.
`level`	Level of granularity for the analysis: log, case, activity, resource or resource-activity. For more information, see `vignette("metrics", "edeaR")`
`append`	Logical, indicating whether to append results to original event log. Ignored when level is log or trace.
`...`	Deprecated arguments
`append_column`	Which of the output columns to append to log, if append = T. Default column depends on chosen level.
`sort`	Sort output on count. Defaults to TRUE. Only for levels with frequency count output.

Activity instances of the same activity type that are executed more than once immediately after each other by the same resource are in a self-loop (length-1-loop). If an activity instance of the same activity type is executed 3 times after each other by the same resource, this is defined as a size 2 self-loop.

Two types of self-loops are defined, which are repeat self-loops and redo self-loops. Repeat self-loops are activity executions of the same activity type that are executed immediately following each other by the same resource. Redo self-loops are activity executions of the same activity type that are executed immediately following each other by a different resource. Repeat and redo repetitions are explained further on.

These metrics are presented on five different levels of analysis, which are the complete event log, cases, activities, resources and resource-activity combinations.

On the level of the complete event log, the summary statistics of the number of self-loops within a trace can give a first insight in the amount of waste in an event log. As stated earlier, each combination of two occurrences of the same activity executed by the same resource will be counted as one repeat self-loop of this activity.
This metric on the level of cases provides an overview of the absolute and relative number of repeat and redo self-loops in each case. To calculate the relative number, each (repeat or redo) self-loop is counted as 1 occurrence, and the other activity instances are also counted as 1.
On the level of the distinct activities in the event log, the absolute and relative number of self-loops per activity can be an indication for the company which activities are causing the most waste in the process.
Similar to the metric on the level of the activities, the number of self-loops on the level of the resources executing the activities can give a company insights in which employee needs to repeat his or her work most often within a case, or for which employee the work he or she did should be redone by another employee within the same case. This metric shows the absolute and relative number of both repeat and redo self-loops for each resource in the event log.
Finally, the metric can be applied to the level of the specifc resource-activity combinations, in order to get an insight in which activities are the most crucial for which resources. This metric shows the absolute and relative number of both repeat and redo self-loops for each of the resource-activity combinations that occur in the event log. Two different relative numbers are provided here, one from the resource perspective and one from the activity perspective. At the resource perspective, the denominator is the total number of executions by the resource under consideration. At the activity perspective, the denominator is the total number of occurrences of the activity under consideration.

eventlog: Compute number of selfloops for eventlog
grouped_eventlog: Compute number of selfloops for grouped eventlog

Swennen, M. (2018). Using Event Log Knowledge to Support Operational Exellence Techniques (Doctoral dissertation). Hasselt University.

gertjanssenswillen/edeaR documentation built on July 22, 2019, 7:08 p.m.