Description

This curriculum spans the full lifecycle of a multi-workshop process mining initiative, from scoping and data integration through discovery, deviation analysis, and redesign, mirroring the iterative diagnostic and implementation cycles seen in enterprise process improvement programs.

Module 1: Establishing Process Mining Objectives and Scope

Selecting which business processes to prioritize for mining based on operational pain points, regulatory exposure, and potential ROI from redesign.
Defining the boundaries of process instances (e.g., order-to-cash start and end events) to ensure consistent trace extraction from source systems.
Negotiating access to transactional data across departments with competing priorities and data ownership concerns.
Determining whether to include exception handling paths or focus only on standard workflows in initial analysis.
Aligning stakeholder expectations on what constitutes a "successful" process mining outcome—compliance, cycle time reduction, or cost savings.
Deciding whether to conduct a pilot on a single process variant or scale across multiple units from the outset.

Module 2: Data Extraction and Event Log Construction

Mapping disparate data sources (ERP, CRM, BPM) to the IEEE XES or CSV event log format while preserving case, activity, and timestamp semantics.
Resolving inconsistent or missing timestamps by applying business rules or imputation methods without distorting process flow.
Handling composite cases where a single business transaction spans multiple system-generated case IDs.
Filtering out test, sandbox, or administrative transactions that pollute real process behavior analysis.
Dealing with event attribute truncation or data type mismatches when extracting from legacy databases.
Establishing refresh mechanisms for event logs to support longitudinal analysis without overloading source systems.

Module 3: Process Discovery and Model Generation

Choosing between discovery algorithms (e.g., Inductive Miner, Heuristic Miner) based on log completeness and noise levels.
Adjusting frequency and dependency thresholds in Heuristic Miner to balance model simplicity and behavioral accuracy.
Interpreting spaghetti-like process maps by applying filtering strategies (e.g., frequency-based, performance-based) to isolate dominant paths.
Deciding when to split a discovered model by organizational unit, region, or customer segment to uncover variation causes.
Validating discovered models against known process documentation or SME interviews to identify undocumented workarounds.
Documenting model assumptions such as concurrency handling and loop detection for auditability.

Module 4: Conformance Checking and Deviation Analysis

Selecting a reference model for comparison—whether based on official policy, best-practice templates, or idealized workflows.
Configuring alignment-based conformance checking to identify deviations while managing computational load on large logs.
Distinguishing between harmful deviations (e.g., control violations) and beneficial ones (e.g., efficiency shortcuts).
Quantifying the operational impact of non-conformance by linking deviations to downstream outcomes like rework or delays.
Handling partial traces in conformance checking when event logs lack complete start-to-end coverage.
Reporting deviation findings in a way that avoids blaming individuals while highlighting systemic root causes.

Module 5: Performance and Bottleneck Analysis

Calculating end-to-end processing times while accounting for waiting, handover, and rework loops across activities.
Attributing delays to specific organizational roles or system interfaces using timestamp and resource field analysis.
Identifying resource contention by analyzing queuing patterns before high-utilization activities.
Adjusting for calendar-aware processing times (e.g., excluding weekends, holidays, or shift hours) in cycle time metrics.
Correlating performance outliers with external factors such as seasonality, system outages, or staffing changes.
Setting performance benchmarks based on historical data percentiles rather than arbitrary targets.

Module 6: Redesign Recommendation Development

Ranking process improvement opportunities by impact (e.g., time saved) and feasibility (e.g., system constraints).
Proposing role consolidation or task automation only after validating task repetition and decision logic from event data.
Designing parallel paths in redesigned workflows where sequence analysis shows unnecessary serial dependencies.
Specifying data retention rules for new process variants to ensure future mining remains viable.
Integrating control points in redesigned processes to enable future compliance monitoring via mining.
Documenting assumptions behind each redesign proposal to support impact simulation and stakeholder review.

Module 7: Change Implementation and Monitoring

Coordinating with IT teams to instrument new or modified processes with sufficient logging for future mining.
Establishing baseline metrics pre-implementation to measure the effect of redesign interventions.
Configuring ongoing process mining dashboards to detect regression or new deviations post-deployment.
Handling versioning of process models when multiple variants coexist during transition periods.
Updating access controls and data extraction jobs as process ownership or system landscapes change.
Conducting periodic mining cycles to identify emergent inefficiencies or adaptation drift over time.