-Qlik Sense R platform based MEHRWERK ProcessMining is the first tool to offer self-service process mining including the data governance necessary for enterprise-wide use. Fast implementation through various data extraction options and integrability with machine learning functions make MPM a customizable tool for every use case. Basic process discovery and conformance checking algorithms in combination with a powerful business intelligence platform allow in-depth analysis as flexible as the investigation requirements themselves.
As process mining is a highly explorative analysis technique, process mining software should provide interactive and visual analysis to support insights in complex processes. MEHRWERK ProcessMining (MPM), deployed on the Qlik Sense R platform, is designed to offer comprehensive analytics to the process analyst. MPM combines market leading self-service business intelligence (self-service BI) with the insights achieved by process mining algorithms and visualizations for process discovery and conformance checking and, thus, becomes a powerful tool for professional process analysis. Through Qlik Sense R ’s APIs it is possible to enhance the solution according to the given use case’s requirements, for example, by integrating data mining scripts from R or Python. To introduce MPM properly to the interested reader we will firstly discuss the innovation our software offers to the process mining universe by shortly explaining the platform and the implemented functions. Then a description of the tool’s maturity is presented to outline its usability and its adaptability to future developments. Future developments are shortly mentioned to elaborate our vision and in the concluding part of the paper is a link to a 25 minute introduction video.
mds
The Qlik Sense R platform provides user-friendly
analysis based on governed data discovery. With the Qlik
Associative Engine it is possible to discover relations in even vast
amounts of data by technically performing a many-to-many
full outer join in the back-end [
As business analytics is dependent on up-to-date and
high-quality data, Qlik Sense R offers the import of various
data formats as well as connectors to a large number of
enterprise information systems. Official connectors can be
found at [
MPM provides two options of feeding event logs into the process mining algorithm. The first is to run the code on an existing event log that is read into Qlik Sense R and transformed into a minimal log with information about case ID, activity type and activity timestamps such as start or end of the event. The second option is to extract data directly from sources, as databases, XML, xlsx or csv with the MPM RuleEngine and create the event log. Doing this, there is no need to provide a process mining-ready event log beforehand and the import effort is significantly reduced. Advantageous in this data-directly from source scenario is the opportunity to easily alter the scope. If time periods or activities of interest change, a few clicks lead to a new event log that is automatically analyzed by our process discovery algorithm. Also, due to the MPM RuleEngine the event log generation is implicitly and comprehensively documented. Through these distinct initiation possibilities, MPM has a short implementation time since it adapts to the use case and the existing data architecture.
To offer the best insights in process data, MPM connects the process log to context data by Qlik-typical data modeling. Omitted KPIs are, if needed, fastly calculated and appended to the data model, often without the need to go back to the data extraction step. Thus, the enrichment of event logs, process logs or data models for e.g. root cause analysis purposes is rapidly accomplished given some knowledge in Qlik Sense R formular syntax.
Through MPM, the complex task of integrating process mining with modern business intelligence is achieved. By defining reusable template applications the analyst is enabled to perform exactly the same analysis on top of different event logs – some refer to this as history mode. For example, the comparison of different time periods before and after a certain process intervention is simply achieved. Since performance and quality analysis, compliance checking as well as process monitoring are important areas of process mining applications, the opportunities provided by the integration with BI that are offered by MPM are outstanding. 2.3.1. Process Discovery. Process Discovery in MPM is performed by fuzzy mining the event log and extracting the resulting directly-follows graph. Hence, the deterministic MPM Process Discovery Algorithm is rather simple but efficient. The MPM ProcessAnalyzer shows the real processes as-is and enhances them by standard metrics like process step duration, lead or idle times and further use casedependent performance indicators such as automation rates. Filter and selection options help reducing the spaghettidiagram to relevant process variants. Through Qlik’s Associative Engine the whole application content is then recalculated for the selected data. The MPM QueryBuilder provides the useful capability to search for activity patterns of concern thus, allowing to fastly check critical process variants and their context information.
ality is conformance checking. With the MPM
ProcessModeler the user is able to define a happy path via drag
and drop which is displayed in the process visualization
to investigate process deviations. If a should-be process
model is defined, its list of distinct process variants can be
used to calculate the process conformance between reality
and model. The MPM ConformanceChecking Algorithm
encompasses alignment based logic and, therefore,
evaluates activities and moves that are either synchronous or
in model or log. With these information a fitness metric
for the most fitting happy path and each real-life process
variant is calculated. Hence, the analyst can rapidly identify
highly deviating process variants due to deviating activities
or changed activity ordering and compares process variants
that should, but, do not correspond to one common happy
path.
2.3.3. Process Analysis Perspective. Given the analytic
flexibility of Qlik Sense R , analysts can use MPM to evaluate
processes from different perspectives. At first, the control
flow perspective can be taken to get an overview on the
general process by generating the process visualization and
calculating the process variants. The ordering of activities
and the deviation from a happy path get clear in the
visualization and can be further assessed by a variety of diagrams,
KPIs and the alignment based conformance checking. With
the process instance as central investigation object, the case
perspective is the easiest perspective to take with MPM.
Along with resource data, the use case dependent context,
e.g. supplier, material, machine or weather, can be included
in the analysis. To cover the organizational or rather resource
perspective the analyst can employ a social network graph,
an extension provided by an open source extension on Qlik
Branch R Garden [
Figure 1 demonstrates the seamless interaction of the Qlik selection (green, white, light and dark grey), the MPM ProcessAnalyzer, and the MPM ConformanceChecking. The left process visualization exhibits the most common process variant in company MEHRWERK UK’s purchasing group M00. This variant is set as happy path for the right visualization to compare it with the process variants for purchasing group M03. Red moves in the right graph show where M03 deviates from the process of M00 which is marked greenly.
As Qlik Sense R is the platform for MPM, the maturity of the analytic capabilities is eminently high with more than 50,000 customers of different sizes and industries trusting in Qlik Sense R ’s functionality. Advantageous for any development actions is the large community of Qlik users which also provides open source code for extension development. The MPM functionality itself has been tested in various reallife projects receiving strong positive feedback by customers with environments varying from small up to large and e.g. multiple SAP R systems and heterogenous data sources. The deterministic MPM ProcessDiscovery Algorithm creates a dependency graph as process visualization showing the spaghetti-like process variants. The filter and selection functions of Qlik Sense R help to reduce the process variants to the relevant scope. The MPM ConformanceChecking Algorithm handles should-be process models by splitting them into the distinct process variants which are then compared to the real-life process variants. Planned but not yet implemented is conformance checking that also includes the should-be process step durations or process step idle times in the fitness calculation. In none of the projects where MPM was implemented problems with respect to scalability occurred. Nevertheless, to meet the future demands by Big Data, MPM is prepared to confront the upcoming challenges with Qlik Sense R ’s strategies.
To overcome performance problems due to big data, Qlik
follows different approaches like on demand app
generation and large scale architectures for horizontal or vertical
scalability. On-demand apps allow the user to load and
analyze big data sources, where aggregated, representative
visualizations are created for the whole data set to identify
interesting data subsets on which detailed analysis is carried
out afterwards by interactively generating an on-demand
app with the full Qlik in-memory capabilities [
One of the first projects realised with MPM was to rapidly identify compliance issues as well as to ensure and prevail corporate governance in a purchase-to-pay process of a large German energy supplier. The audit, governance and industrial control system area of this enterprise initially comprised 1,828,864 cases and 3,036,204 events composed by 27 activities. The data were obtained from SAP R ERP and Excel sources. With MPM the company achieved full asis purchasing process transparency across multiple company codes and purchasing organisations combined with automated calculation and evaluation of risk indicators.
Another application was an order-to-cash project at a German raw material supplier where MPM was used to identify optimization potential for days of sales outstanding (DSO), order lead time, order processing cost and the general process. 477,345 process instances with more than 4.4 million events leading to over 45,000 process variants have been analyzed.
Another project by a German automotive logistics supplier was the optimization of their logistic process. The small logistics data set came from a SAP R ERP including 6,159 cases, 49,711 events which were built up of 15 activities and formed 85 process variants. MPM generated insights in optimization potential, for example, by identifying the refueling process as bottleneck. Major achievements were the discovery of “forgotten” cars on parking sites and a new benchmark system due to understanding which transport system - train or truck - was the better option in certain situations.
One project that was not based on SAP R input data was the analysis of a production process which’ data were recorded in two different information systems – a computeraided quality system and a manufacturing execution system. During production, the process instance changed from batch process to single production process. Therefore, the requirement was to achieve an overview on the overall process by investigating the subprocesses in one single app. Hence, two event logs were extracted by the MPM RuleEngine from the source systems, the first with 6,483 batches as process instance, 11 activities and 51,334 events and the second with 49,396 single production numbers as process instances, 17 activties and 322,201 events. The two event logs were processed and led to two process logs that were associated to one another via the batch information. The processes were visualized seperatly but in the same app so the analysts could clearly see the overall process and track the single production number down to batch events. The aim of employing MPM was to generate a starting point for continuous process improvement. Slow process variants as well as production bottlenecks were discovered and selected for improvement activities.
A first version of decision tree forecasting and root cause analysis capabilities is in testing at the moment. Planned for the future is an intuitive root cause analysis modeling functionality and stronger prediction capabilities with deep learning algorithms for decision support and automated activity triggering. The enhancement of MPM’s process visualization with hierarchical functionalities that allow for the aggregation of activities to a super-node is in development. Aggregating and deaggregating the view on as-is processes will thereby be supported.
Having recent developments of Qlik Sense R regarding
chat bots and alerting functions [
MEHRWERK ProcessMining based on the Qlik Sense R platform is first class self-service process mining that provides the data governance necessary for enterprise-wide use. MPM offers fast implementation through flexible data extraction options for a wide range of use cases with or without a previously prepared event log and is easily adaptable to specific analysis requirements due to integratability with machine learning functions (e.g. DataRobot/Python/R).
We see application possibilities of MPM in technically every context where IT systems track events. Experiences have been collected in P2P and O2C processes, classic logistic and production processes, banking portfolio management and the monitoring of user-interaction with machines. MPM can be generally helpful for improving processes due to algorithmic process discovery and analysis. Furthermore, it works as enabler for root cause analysis. Improving auditing and compliance by algorithmic process comparison is an application field as well as supporting process automation or digital transformation by discovering opportunities and linking strategies to operations. An interesting use case would be the monitoring and controlling of S/4 Hana migrations. A 25 minutes introduction video can be found at: https://mpm-processmining.com/demo-video-mpm-2019-en/, password: MPMDemo2019