A multimedia dataset for object-centric business process mining in IT asset management.

Area of Science:

• Computer Science
• Information Systems
• Business Process Management

Background:

• IT Asset Management (ITAM) processes are critical for IT operations.
• Existing datasets often lack the richness of real-world, unstructured data.
• Process mining techniques require comprehensive data for effective development and evaluation.

Purpose of the Study:

• Introduce a novel multimedia business process dataset.
• Provide a resource for advancing process mining on unstructured data.
• Facilitate research in ITAM process analysis.

Main Methods:

• Systematic data collection in a controlled laboratory environment.
• Integration of multi-modal data: video, motion tracking, environmental sensors, ITAM system traces, and user interaction logs.
• Data organized in an object-centric event log format (OCEL) with manual annotation of activities and entities.

Main Results:

• A comprehensive dataset comprising 121 process instances across six basic ITAM processes.
• Raw and processed data available, including 37 video recordings from dual perspectives.
• Detailed event data enabling traceability across diverse multimedia sources.

Conclusions:

• The dataset serves as a valuable benchmark for process mining research.
• Enables the development of techniques for analyzing unstructured data in business processes.
• Supports the evaluation of process mining algorithms in realistic ITAM scenarios.