Bachelor Thesis - Using actionable insights provided by process analytics to workflow automation with a focus on Customer Relationship Management process
Process mining techniques are used to extract actionable insights from many different information systems, based on which one can execute management actions that improve the performance of underlying business processes. This BSc thesis project aims at bridging the process mining diagnostics to the automation of management actions. Particularly, the student is required to implement the integration of the PM4PY process mining library, which supports the process mining techniques and Integromat, which automates the execution of management actions. The integration involves the following tasks:
- Building an online connection to a customer relationship management (CRM) system (e.g., Salesforce and Dynamics365) using Integromat. The subtasks may include
- Realizing a workflow in the Integromat platform to automate the ETL of event data from supporting database systems.
- Implementation of a set of web services in the Python programming language that accept the events incoming from Integromat.
- Analyzing the streaming event data from the CRM with process mining techniques (supported by PM4PY), with the usage of some of the following techniques:
- offline process mining: producing diagnostics of the underlying process (e.g., process discovery, conformance checking, performance analysis),
- online process mining: detecting deviations of running process instances (e.g., online conformance checking, anomaly detection, and prediction),
- automating management actions over CRM process using the Integromat’s workflow automation platform, e.g.,
- record deviations in the daily report and send a Slack message to the manager and,
- scheduling automatically on Google calendar if “physical appointment” activity is missing with a customer.
Note that only part of the tasks described above will be the BSc thesis’s scope, and it can be adapted to your competency and interest.
The desired outputs of the project include:
- The implementation of the process discovery/conformance checking/anomaly detection/prediction/action-based process mining techniques based on PM4PY.
- Workflows in the Integromat platform (e.g., triggers of the chosen CRM system and connections to a database system)
- Web interface to support the tasks described above (e.g., showing the process model of the last 24 hours/detecting deviations of process instances).
Along with the implementation, the project’s output is a report that describes the workflow of the integration, the proposed process mining techniques, the experimental setting, and the related work in the field. Also, comparisons with commercial software are possible.
The student should be familiar with the Python programming language and have a basic knowledge of process mining. In particular, hands-on experience with process mining (practical courses, pro-seminar, BPI course) is preferred.
- PM4Py process mining library https://pm4py.fit.fraunhofer.de/
- Integromat system integration platform https://www.integromat.com/en/
- Coursera course “Process Mining: Data Science in Action” https://www.coursera.org/learn/process-mining
- Introduction to Customer Relationship Management systems https://de.wikipedia.org/wiki/Customer-Relationship-Management