Filtering cases Gert Janssenswillen Creator of bupaR DataCamp - - PowerPoint PPT Presentation

▶
filtering cases
SMART_READER_LITE
LIVE PREVIEW

Filtering cases Gert Janssenswillen Creator of bupaR DataCamp - - PowerPoint PPT Presentation

Jul 21, 2023 338 likes •806 views

DataCamp Business Process Analytics in R BUSINESS PROCESS ANALYTICS IN R Filtering cases Gert Janssenswillen Creator of bupaR DataCamp Business Process Analytics in R DataCamp Business Process Analytics in R DataCamp Business Process

slide-1
SLIDE 1

DataCamp Business Process Analytics in R

Filtering cases

BUSINESS PROCESS ANALYTICS IN R

Gert Janssenswillen

Creator of bupaR

slide-2
SLIDE 2

DataCamp Business Process Analytics in R

slide-3
SLIDE 3

DataCamp Business Process Analytics in R

slide-4
SLIDE 4

DataCamp Business Process Analytics in R

slide-5
SLIDE 5

DataCamp Business Process Analytics in R

slide-6
SLIDE 6

DataCamp Business Process Analytics in R

slide-7
SLIDE 7

DataCamp Business Process Analytics in R

slide-8
SLIDE 8

DataCamp Business Process Analytics in R

slide-9
SLIDE 9

DataCamp Business Process Analytics in R

slide-10
SLIDE 10

DataCamp Business Process Analytics in R

Categories of Case Filters

Performance Control-flow characteristics Time period

slide-11
SLIDE 11

DataCamp Business Process Analytics in R

Performance filters

slide-12
SLIDE 12

DataCamp Business Process Analytics in R

Performance filters

slide-13
SLIDE 13

DataCamp Business Process Analytics in R

Filter by absolute interval

filter_throughput_time(log, interval = c(5,10))

slide-14
SLIDE 14

DataCamp Business Process Analytics in R

Filter by absolute interval

filter_throughput_time(log, interval = c(5,10))

slide-15
SLIDE 15

DataCamp Business Process Analytics in R

Filter by Relative Percentage

filter_throughput_time(log, percentage = 0.5)

slide-16
SLIDE 16

DataCamp Business Process Analytics in R

Adjusting filter configurations

Negate the filter Cases shorter than 5 days, or longer than 10 days The 50% longest cases Use half-open intervals Select cases with throughput time longer than 5 days.

filter_throughput_time(log, interval = c(5,10), units = "days", reverse =TRUE) filter_throughput_time(log, percentage = 0.5, reverse = TRUE) filter_throughput_time(log, interval = c(5,NA), units = "days")

slide-17
SLIDE 17

DataCamp Business Process Analytics in R

Control-flow filters

Activity presence/absence Precendence requirements Start and/or End points Frequency of the trace

slide-18
SLIDE 18

DataCamp Business Process Analytics in R

Time filters

Select cases that started in a specific time window ended in a specific time window are contained in a specific time window intersect, i.e. had at least on activity in a specific time window

slide-19
SLIDE 19

DataCamp Business Process Analytics in R

Let's practice!

BUSINESS PROCESS ANALYTICS IN R

slide-20
SLIDE 20

DataCamp Business Process Analytics in R

Filtering events

BUSINESS PROCESS ANALYTICS IN R

Gert Janssenswillen

Creator of bupaR

slide-21
SLIDE 21

DataCamp Business Process Analytics in R

Categories of event filters

Trim filters Frequency filters Label filters General Attribute filters

slide-22
SLIDE 22

DataCamp Business Process Analytics in R

Trim to time period

filter_time_period(log, interval = ymd(c("20180110","20180122")), filter_method = "trim")

slide-23
SLIDE 23

DataCamp Business Process Analytics in R

Trim to start and end points

filter_trim(start_activities = "blues")

slide-24
SLIDE 24

DataCamp Business Process Analytics in R

Trim to start and end points

filter_trim(start_activities = "blues", end_activities = "greens")

slide-25
SLIDE 25

DataCamp Business Process Analytics in R

Trim to start and end points

slide-26
SLIDE 26

DataCamp Business Process Analytics in R

Filter by frequencies

Activity frequency Resource frequency

filter_activity_frequency(log, interval = c(50,100)) filter_activity_frequency(log, percentage = 0.8) filter_resource_frequency(log, interval = c(60,900)) filter_resource_frequency(log, percentage = 0.6)

slide-27
SLIDE 27

DataCamp Business Process Analytics in R

Filter by labels

filter_activity(log, activities = c("reds","oranges","purples")))

slide-28
SLIDE 28

DataCamp Business Process Analytics in R

Filter by conditions

Any condition using data attributes can be used Multiple conditions can be combined using &, |, !, etc.

filter(log, cost > 1000, priority == "High", ...)

slide-29
SLIDE 29

DataCamp Business Process Analytics in R

Let's practice!

BUSINESS PROCESS ANALYTICS IN R

slide-30
SLIDE 30

DataCamp Business Process Analytics in R

Aggregating events

BUSINESS PROCESS ANALYTICS IN R

Gert Janssenswillen

Creator of bupaR

slide-31
SLIDE 31

DataCamp Business Process Analytics in R

Aggregation types

Is-A Part-of

slide-32
SLIDE 32

DataCamp Business Process Analytics in R

Is-a aggregation

slide-33
SLIDE 33

DataCamp Business Process Analytics in R

Is-a aggregation

act_unite(log, "New name" = c("Old Variant 1","Old Variant 2","Old Variant 3"), ...)

slide-34
SLIDE 34

DataCamp Business Process Analytics in R

Part-of aggregation

slide-35
SLIDE 35

DataCamp Business Process Analytics in R

Part-of aggregation

act_collapse(log, "Sub process" = c("Part 1","Part 2","Part 3"), ...)

slide-36
SLIDE 36

DataCamp Business Process Analytics in R

Impact on Activity Types and Instances

Is-a Decreased number of activity types Equal number of activity instances Part-of Decreased number of activity types Decreased number of activity instances

slide-37
SLIDE 37

DataCamp Business Process Analytics in R

Let's practice!

BUSINESS PROCESS ANALYTICS IN R

slide-38
SLIDE 38

DataCamp Business Process Analytics in R

Enriching events

BUSINESS PROCESS ANALYTICS IN R

Gert Janssenswillen

Creator of bupaR

slide-39
SLIDE 39

DataCamp Business Process Analytics in R

Mutate new variables

slide-40
SLIDE 40

DataCamp Business Process Analytics in R

Mutate new variables

log %>% group_by_case() %>% mutate(total_cost = sum(cost, na.rm = T)

slide-41
SLIDE 41

DataCamp Business Process Analytics in R

Mutate new variables

log %>% group_by_case %>% mutate(total_cost = sum(cost, na.rm = T) %>% mutate(impact = case_when(cost <= 1000 ~ "Low", cost <= 5000 ~ "Medium", T ~ "High"))

slide-42
SLIDE 42

DataCamp Business Process Analytics in R

Mutate new variables

log %>% group_by_case %>% mutate(refund_made = any(str_detect(activity, "Pay Claim")))

slide-43
SLIDE 43

DataCamp Business Process Analytics in R

Adding process metrics

Adding information about a case to the original data Its througput time Its length Its amount of rework ... Adding information about activities Its frequency Its specialization by resources ...

slide-44
SLIDE 44

DataCamp Business Process Analytics in R

Adding process metrics

log %>% throughput_time(level = "case", units = "days", append = TRUE) log %>% throughput_time(level = "case", units = "days", append = TRUE) %>% mutate(on_time = processing_time_case <= 7)

slide-45
SLIDE 45

DataCamp Business Process Analytics in R

Let's practice!

BUSINESS PROCESS ANALYTICS IN R