Apromore Process Mining vs bupaR

Apromore Process Mining analyzes business process data so organizations can understand, monitor, and simulate their processes. It discovers process maps and BPMN models from event logs, with overlays for activity frequency and duration. Teams can filter visually, check conformance, compare processes, and examine model differences. Simulation supports current and hypothetical scenarios on BPMN models; task mining uses desktop application data to reveal routines and potential improvements or automation opportunities. Machine-learning models can predict outcomes such as SLA violations, remaining time, next activity, and case continuation. The Integration Center includes connectors for Salesforce, SAP, ServiceNow, Oracle NetSuite, and Jira, plus JDBC/ODBC access and a REST API. A no-code interface supports scheduled ETL pipelines, transformation mappings, and incremental event-log uploads. Apromore describes SaaS access for Enterprise and cloud, on-premises, or hybrid deployment for its SAP offering. Enterprise is an annual subscription with unlimited logs and process models and technical support; pricing is on request. A 30-day trial and a free version for research or education are also described.

bupaR is a free, open-source suite of R packages for working with business process data. It supports importing and preparing event data, calculating descriptive statistics, visualising processes, and checking conformance. Analysts can examine control flow, performance, and organisational aspects, while including custom process attributes in their workflow. It handles event logs, where each row represents an event, and activity logs, where each row represents an activity instance. Process maps help summarise processes and can be used in Shiny apps or exported to PDF, PNG, PS, or EPS. The bupaverse core contains bupaR, edeaR, eventdataR, processcheckR, and processmapR. The xesreadR package reads XES files from a local path or URL and writes event logs to XES. Installation instructions cover CRAN and GitHub development versions. The package requires R 3.5.0 or later and is licensed under MIT plus a license file. For prediction with processpredictR, Python, TensorFlow, and Keras are required. Validation of larger event logs can take time, though the documentation allows it to be disabled when the data meets requirements.