Process Mining With R
MiLin
3/17/2022
- install
- create Event Log Object
- Basic Event Log Functionalities
- Create a summary of an eventlog-object.
- Obtain the mapping of an eventlog (set of identifiers) or obtain single identifiers.
- Calculate the number of distinct activities, activity instan- ces, cases, events, resources and traces.
- Subset of slice of the event log from row n until row m
- Sample n cases from the event log.
- Add new variables to the event log
- Group an event log on one or more event or case attribu- tes.
- Reading and Writing XES-files
- data set in BupaR
- Exploratory and Descriptive Event Data Analysi
- event data subsetting
- process Visulization
- Process Dashboard
install
install.packages("bupaR")
install.packages("edeaR")
install.packages("eventdataR")
install.packages("processmapR")
install.packages("processmonitR")
install.packages("xesreadR")
install.packages("petrinetR")
then library
library(bupaR)##
## Attaching package: 'bupaR'## The following object is masked from 'package:stats':
##
## filter## The following object is masked from 'package:utils':
##
## timestamp
create Event Log Object
An eventlog-object can be created using the eventlog function. This function needs as arguments a data.frame and the column names of the appropriate fields descri- bing the case identifier, activity identifier, timestamp, li- fecycle transition, resource and an activity instance iden- tifier.
An event log with minimal requirements (timestamp, ca- se and activity identifier) can be created with the sim- ple_eventlog function.
for example
library(lubridate)##
## Attaching package: 'lubridate'## The following objects are masked from 'package:base':
##
## date, intersect, setdiff, uniondata <- data.frame(case = rep("A",5),
activity_id = c("A","B","C","D","E"),
activity_instance_id = 1:5,
lifecycle_id = rep("complete",5),
timestamp = now()+ddays(5),
resource = rep("resource 1", 5))
Meventlog <- eventlog(data,case_id = "case",
activity_id = "activity_id",
activity_instance_id = "activity_instance_id",
lifecycle_id = "lifecycle_id",
timestamp = "timestamp",
resource_id = "resource")Basic Event Log Functionalities
Create a summary of an eventlog-object.
summary(Meventlog)## Number of events: 5
## Number of cases: 1
## Number of traces: 1
## Number of distinct activities: 5
## Average trace length: 5
##
## Start eventlog: 2022-03-22 04:26:53
## End eventlog: 2022-03-22 04:26:53## case activity_id activity_instance_id lifecycle_id
## Length:5 A:1 Length:5 complete:5
## Class :character B:1 Class :character
## Mode :character C:1 Mode :character
## D:1
## E:1
##
## timestamp resource .order
## Min. :2022-03-22 04:26:53 resource 1:5 Min. :1
## 1st Qu.:2022-03-22 04:26:53 1st Qu.:2
## Median :2022-03-22 04:26:53 Median :3
## Mean :2022-03-22 04:26:53 Mean :3
## 3rd Qu.:2022-03-22 04:26:53 3rd Qu.:4
## Max. :2022-03-22 04:26:53 Max. :5bupaR::activities(Meventlog)## # A tibble: 5 × 3
## activity_id absolute_frequency relative_frequency
## <fct> <int> <dbl>
## 1 A 1 0.2
## 2 B 1 0.2
## 3 C 1 0.2
## 4 D 1 0.2
## 5 E 1 0.2bupaR::cases(Meventlog)## Warning: `tbl_df()` was deprecated in dplyr 1.0.0.
## Please use `tibble::as_tibble()` instead.## # A tibble: 1 × 10
## case trace_length number_of_activities start_timestamp complete_timestamp
## <chr> <int> <int> <dttm> <dttm>
## 1 A 5 5 2022-03-22 04:26:53 2022-03-22 04:26:53
## # … with 5 more variables: trace <chr>, trace_id <dbl>, duration_in_days <dbl>,
## # first_activity <fct>, last_activity <fct>bupaR::resources(Meventlog)## # A tibble: 1 × 3
## resource absolute_frequency relative_frequency
## <fct> <int> <dbl>
## 1 resource 1 5 1bupaR::traces(Meventlog)## # A tibble: 1 × 3
## trace absolute_frequency relative_frequency
## <chr> <int> <dbl>
## 1 A,B,C,D,E 1 1Obtain the mapping of an eventlog (set of identifiers) or obtain single identifiers.
bupaR::mapping(Meventlog) ## Case identifier: case
## Activity identifier: activity_id
## Resource identifier: resource
## Activity instance identifier: activity_instance_id
## Timestamp: timestamp
## Lifecycle transition: lifecycle_idbupaR::activity_id(Meventlog) ## [1] "activity_id"bupaR::activity_instance_id(Meventlog) ## [1] "activity_instance_id"bupaR::case_id(Meventlog)## [1] "case"bupaR::lifecycle_id(Meventlog)## [1] "lifecycle_id"bupaR::resource_id(Meventlog)## [1] "resource"bupaR::timestamp(Meventlog)## [1] "timestamp"Calculate the number of distinct activities, activity instan- ces, cases, events, resources and traces.
n_activities(Meventlog)## [1] 5bupaR::n_activity_instances(Meventlog)## [1] 5bupaR::n_cases(Meventlog) ## [1] 1bupaR::n_events(Meventlog) ## [1] 5bupaR::n_resources(Meventlog)## [1] 1bupaR::n_traces(Meventlog)## [1] 1Subset of slice of the event log from row n until row m
bupaR::slice(eventlog, n:m)Sample n cases from the event log.
bupaR::sample_n(eventlog, n)Add new variables to the event log
bupaR::mutate(eventlog, ...)Group an event log on one or more event or case attribu- tes.
bupaR::group_by(eventlog, ...)Reading and Writing XES-files
xesreadR::read_xes(xesfile)
xesreadR::read_xes_cases(xesfile)
xesreadR::write_xes(eventlog,
xesfile,
case_attributes)data set in BupaR
eventdataR::BPIC_14_incident_log eventdataR::BPIC_14_incident_case_attributes eventdataR::BPIC_15_1
eventdataR::sepsis
eventdataR::patientsExploratory and Descriptive Event Data Analysi
edeaR provides a varied set of metrics to analyse event logs.
edeaR::activity_frequency
edeaR::activity_presence
edeaR::end_activities
edeaR::idle_time
edeaR::number_of_repetitions
edeaR::number_of_selfloops
edeaR::number_of_traces
edeaR::processing_time
edeaR::resource_frequency
edeaR::resource_involvement
edeaR::resource_specialisation
edeaR::size_of_repetitions
edeaR::size_of_selfloops
edeaR::start_activities
edeaR::throughput_time
edeaR::trace_coverage
edeaR::trace_length
for example
library(edeaR)
patients %>% activity_frequency(level = "activity")## # A tibble: 7 × 3
## handling absolute relative
## <fct> <int> <dbl>
## 1 Registration 500 0.184
## 2 Triage and Assessment 500 0.184
## 3 Discuss Results 495 0.182
## 4 Check-out 492 0.181
## 5 X-Ray 261 0.0959
## 6 Blood test 237 0.0871
## 7 MRI SCAN 236 0.0867we can do visulization
patients %>% activity_frequency(level = "activity") %>% plot()
patients %>% trace_length() ## min q1 median mean q3 max st_dev iqr
## 2.0000000 5.0000000 5.0000000 5.4420000 6.0000000 6.0000000 0.5790567 1.0000000event data subsetting
EVENT FILTERS
edeaR::ifilter_activity
edeaR::ifilter_activity_Frequency
edeaR::filter_attributes
edeaR::ifilter_resource
edeaR::ifilter_resource_frequency
edeaR::ifilter_time_period
edeaR::ifilter_trimCASE FILTERS
edeaR::ifilter_activity_presence
edeaR::ifilter_case
edeaR::ifilter_endpoints
edeaR::ifilter_precedence
edeaR::ifilter_processing_time
edeaR::ifilter_throughput_time
edeaR::ifilter_time_period
edeaR::ifilter_trace_frequency
edeaR::ifilter_trace_lengthfor example
patients %>% filter_activity_frequency(percentile = 0.8) ## Warning in deprecated_perc(percentage, ...): Argument percentile_cut_off is
## deprecated. Use percentage instead.## Log of 4496 events consisting of:
## 6 traces
## 500 cases
## 2248 instances of 5 activities
## 5 resources
## Events occurred from 2017-01-02 11:41:53 until 2018-05-05 07:16:02
##
## Variables were mapped as follows:
## Case identifier: patient
## Activity identifier: handling
## Resource identifier: employee
## Activity instance identifier: handling_id
## Timestamp: time
## Lifecycle transition: registration_type
##
## # A tibble: 4,496 × 7
## handling patient employee handling_id registration_type time
## <fct> <chr> <fct> <chr> <fct> <dttm>
## 1 Registration 1 r1 1 start 2017-01-02 11:41:53
## 2 Registration 2 r1 2 start 2017-01-02 11:41:53
## 3 Registration 3 r1 3 start 2017-01-04 01:34:05
## 4 Registration 4 r1 4 start 2017-01-04 01:34:04
## 5 Registration 5 r1 5 start 2017-01-04 16:07:47
## 6 Registration 6 r1 6 start 2017-01-04 16:07:47
## 7 Registration 7 r1 7 start 2017-01-05 04:56:11
## 8 Registration 8 r1 8 start 2017-01-05 04:56:11
## 9 Registration 9 r1 9 start 2017-01-06 05:58:54
## 10 Registration 10 r1 10 start 2017-01-06 05:58:54
## # … with 4,486 more rows, and 1 more variable: .order <int>patients %>% filter_activity_presence(
c("Registration","Check-out"), method = "all") ## Log of 5388 events consisting of:
## 2 traces
## 492 cases
## 2694 instances of 7 activities
## 7 resources
## Events occurred from 2017-01-02 11:41:53 until 2018-05-03 03:34:55
##
## Variables were mapped as follows:
## Case identifier: patient
## Activity identifier: handling
## Resource identifier: employee
## Activity instance identifier: handling_id
## Timestamp: time
## Lifecycle transition: registration_type
##
## # A tibble: 5,388 × 7
## handling patient employee handling_id registration_type time
## <fct> <chr> <fct> <chr> <fct> <dttm>
## 1 Registration 1 r1 1 start 2017-01-02 11:41:53
## 2 Registration 2 r1 2 start 2017-01-02 11:41:53
## 3 Registration 3 r1 3 start 2017-01-04 01:34:05
## 4 Registration 4 r1 4 start 2017-01-04 01:34:04
## 5 Registration 5 r1 5 start 2017-01-04 16:07:47
## 6 Registration 6 r1 6 start 2017-01-04 16:07:47
## 7 Registration 7 r1 7 start 2017-01-05 04:56:11
## 8 Registration 8 r1 8 start 2017-01-05 04:56:11
## 9 Registration 9 r1 9 start 2017-01-06 05:58:54
## 10 Registration 10 r1 10 start 2017-01-06 05:58:54
## # … with 5,378 more rows, and 1 more variable: .order <int>process Visulization
The processmapR package provides functions to visualize processes, both from a control-flow perspective and from a resource perspective. The process_map function allows the user to analyse control-flow from a frequency and a performance per- spectives. The precendence_matrix provides a more compact overview of the process flows. Furthermore, the package provides functions to explore (in)frequent traces and a dotted chart (including an interactive version). Also resource maps and matrices can be made.
library(processmapR)
process_map(patients)patients %>% processmapR::precedence_matrix() %>% plot()
patients %>% trace_explorer()## Warning in trace_explorer(.): No coverage or number of traces set. Defaulting to
## 0.2 for frequent traces.## Warning: `rename_()` was deprecated in dplyr 0.7.0.
## Please use `rename()` instead.
patients %>% idotted_chart()patients %>% resource_map()patients %>% processmapR::resource_matrix() ## # A tibble: 7 × 3
## antecedent consequent n
## <fct> <fct> <int>
## 1 r1 r2 500
## 2 r2 r3 237
## 3 r2 r5 261
## 4 r3 r4 236
## 5 r4 r6 236
## 6 r5 r6 259
## 7 r6 r7 492Process Dashboard
The processmonitR package provides predefined dashboards to interactively monitor processes from different perspectives. Cur- rently, four different dashboards are provided: 1) an activity dashboard, focused on activities, 2) a resource dashboard, focusing on resources, 3) a rework dashboard, focusing on rework and waste, such as self-loops and repetitions, and a 4) performance dashboard, focusing on the time perspective, i.e. throughput time, processing time and idle time. Each dashboard combines several of the metrics and visualization from other bupaR packages into easy to use and navigate dashboards. The dashboards, implemented in Shiny, can be used as standalone dashboards, or incorporated into larger, tailor- made process monitoring dashboards.
library(processmonitR)
patients %>% activity_dashboard()
resource_dashboard()
rework_dashboard()
performance_dashboard()