Examples and use cases
Applies to: CELONIS 4.7
Description
Examples and use cases which show how through combining various operators, complex process related questions can be answered.
This section shows examples and use cases that can be solved with PQL. The solutions are presented in interactive analyses. Some of the problems below are also described in more detail on distinct documentation pages in this section. Examples of the following problems are available:
Activity sequences: Do calculations on activity sequences containing more than two activities.
PQL functions: ACTIVITY_LEAD, ACTIVITY_LAG
Case complaints: For each case, check if there was a customer complaint recorded in a separate
Complaints
table.PQL functions: IN
Clustering: Cluster data based on different groupings.
PQL functions: KMEANS
Coformance queries: Use the results of the conformance checker in your PQL queries.
PQL functions: CONFORMANCE
Fallback status: Find activities which trigger a certain status activity. The status activity is always triggered by the preceding activity as well as all activities of the case that happened at the same time as the preceding activity.
PQL functions: PU_COUNT, DOMAIN_TABLE, SOURCE/TARGET, ACTIVITY_LEAD
Filling ranges: Generate a range of DATE or INT values to also show missing values in the dimension of your chart.
PQL functions: RANGE_APPEND, RUNNING_TOTAL
Machine utilization in Production: Find out how long each machine takes
Multiple invoices per case: Calculate the total number of days spent between creating and clearing multiple invoices of one purchase order.
PQL functions: PU_FIRST, PU_LAST, DOMAIN_TABLE
Order item descriptions: For each item, count how often it has been ordered. Unfortunately, the order item descriptions were entered in a free text field and contain typos. This example shows how we still can count the number of orders for each item by grouping same items together, despite the slightly different spelling.
PQL functions: CLUSTER_STRINGS
Ping pong cases: Identify direct and indirect ping-pong-cases with PQL.
PQL functions: ACTIVITY_LEAD, ACTIVITY_LAG
Revenue per country: Find the top three countries based on their total revenue.
PQL functions: INDEX_ORDER, PU_SUM
Rework per activity: For each activity, calculate the fraction of the cases where this activity is a rework activity based on different rules.
PQL functions: INDEX_ACTIVITY_TYPE, INDEX_ACTIVITY_LOOP, GLOBAL
Segregation of duties: Enforce that within one case certain activities are not executed by the same department.
PQL functions: SOURCE / TARGET, REMAP_VALUES
Throughput times: Summary of ways to calculate throughput times in PQL.
PQL functions: REMAP_TIMESTAMPS, CALC_THROUGHPUT, SOURCE and TARGET, PU_MIN and PU_MAX, DAYS_BETWEEN
Triggering activities: Find out which activities trigger which kind of errors.
PQL functions: ACTIVITY_LAG
Working Capital Optimization: Calculate the ratio of cases that were paid early for working capital optimization.
PQL functions: CALC_THROUGHPUT, COALESCE, PROCESS EQUALS, AVG