Unlocking Sustainability Compliance: Characterizing the EU Taxonomy for Business Process Management

Abstract: To promote sustainable business practices, and to achieve climate neutrality by 2050, the EU has developed the taxonomy of sustainable activities, which describes when exactly business practices can be considered sustainable. While the taxonomy has only been recently established, progressively more companies will have to report how much of their revenue was created via sustainably executed business processes. To help companies prepare to assess whether their business processes comply with the constraints outlined in the taxonomy, we investigate in how far these criteria can be used for conformance checking, that is, assessing in a data-driven manner, whether business process executions adhere to regulatory constraints. For this, we develop a few-shot learning pipeline to characterize the constraints of the taxonomy with the help of an LLM as to the process dimensions they relate to. We find that many constraints of the taxonomy are useable for conformance checking, particularly in the sectors of energy, manufacturing, and transport. This will aid companies in preparing to monitor regulatory compliance with the taxonomy automatically, by characterizing what kind of information they need to extract, and by providing a better understanding of sectors where such an assessment is feasible and where it is not.

• The paper introduces a novel LLM-based pipeline that classifies 1636 EU taxonomy constraints to enable automated sustainability compliance in BPM.
• It bridges sustainability regulations with process execution data, guiding companies on monitoring key operational metrics.
• Validation shows 95% plausible characterizations, highlighting its potential for practical regulatory conformance checking.

Unlocking Sustainability Compliance: Characterizing the EU Taxonomy for Business Process Management

The paper "Unlocking Sustainability Compliance: Characterizing the EU Taxonomy for Business Process Management" presents an investigation into how the EU taxonomy for sustainable activities can be operationalized for conformance checking within the domain of Business Process Management (BPM). This study is grounded in the context of the EU's climate neutrality goals, whereby an increasing number of companies must assess and disclose how much of their revenue stems from sustainable business processes.

Introduction and Background

The EU taxonomy is a classification system intended to create clear indicators to determine when business activities contribute toward sustainability goals. This system defines technical screening criteria for several environmental objectives, including climate change mitigation, climate change adaptation, sustainable water use, transition to a circular economy, pollution prevention, and biodiversity protection. Companies are required to ensure their business processes align with these criteria to be considered sustainable or "taxonomy-aligned."

Given the urgency to mitigate climate change and promote sustainable practices, this study explores the role of conformance checking techniques in monitoring regulatory compliance. Conformance checking compares recorded business process executions (represented as event logs) against a formal model to identify deviations. This formal model embodies constraints derived from the taxonomy, focusing on process dimensions like control-flow, time, resources, and data.

Approach

The research addresses two core questions:

1. How can the EU taxonomy be operationalized with regard to business processes?
2. Which constraints of the EU taxonomy can be used for automatically assessing whether a business process fulfills its respective sustainability criteria with conformance checking techniques?

The authors employ a few-shot learning pipeline, leveraging a LLM, to characterize the constraints defined in the taxonomy. The pipeline consists of three stages: preprocessing the taxonomy texts, prompting the LLM to classify constraints, and parsing the responses to extract detailed constraint information.

Key Findings

The study identifies and categorizes 1636 constraints within the taxonomy, determining their applicability to various process dimensions:

• Activity Existence Constraints: 624 constraints (e.g., permits required for specific activities).
• Between-Activities Control-Flow Constraints: 8 constraints.
• Temporal Constraints: 44 constraints (e.g., periodic reviews required).
• Resource Constraints: 98 constraints (e.g., specific tools or roles required for activities).
• Data Constraints: 539 constraints (e.g., emission thresholds for activities).
• Process-Irrelevant Constraints: 323 constraints.

These constraints span across industries such as energy, manufacturing, transport, and water management. Notably, industries like finance, education, entertainment, and health/social work displayed fewer operationalizable constraints for conformance checking.

Implications

The research provides significant practical and theoretical contributions:

• Practical Implications: Companies can utilize the characterized constraints to prepare for automatic compliance monitoring, identifying relevant data to be captured during process execution. This aids in the creation of prescriptive models for compliance assessment.
• Theoretical Implications: The work extends the methodology of conformance checking by integrating sustainability regulations, enhancing our understanding of how regulatory frameworks like the EU taxonomy can be operationalized within BPM.

Validation and Robustness

The authors validated their approach by cross-checking a portion of the taxonomy against manual assessments, finding that 95% of the characterizations were at least somewhat plausible. This suggests that the LLM-based pipeline provides a trustworthy starting point for further manual refinement.

Future Directions

The study opens avenues for future research, emphasizing the need for:

• Enhanced tools and guidelines to facilitate the practical implementation of the characterized constraints.
• Investigations into other regulatory frameworks and their potential for conformance checking.
• Development of comprehensive frameworks for minimum safeguard assessments, aligning broader organizational compliance with the EU taxonomy.

Conclusion

This research articulates a method to bridge the gap between high-level sustainability regulations and the practical requirements of business process compliance. By characterizing the EU taxonomy's constraints, the study paves the way for more effective and automated sustainability compliance, contributing to the overarching goal of climate neutrality. The interaction between BPM and regulatory adherence now includes a focused lens on sustainable practices, underpinning the broader objective of promoting environmentally responsible business conduct.