Improving Regulatory Compliance with SHACL

In public administration, rules and regulations are vital. It is important to ensure that organizations and individuals follow these rules. However, modeling these requirements can be complex. Traditional methods, such as using the Web Ontology Language (OWL), have certain limitations. A different approach is needed to handle the specific nature of Regulatory Requirements.

#Challenges with OWL

OWL is designed to represent knowledge in a structured way. It allows for automated reasoning to help check if the information is correct and consistent. However, OWL has a major drawback: it operates under an open world assumption. This means that it assumes anything not currently known can still be true. In the context of regulations, this approach is not suitable. Regulations require that knowledge is complete. If something is not known to be true, it is assumed to be false.

Using OWL to model regulatory requirements can be difficult. The complexity of regulations often includes many alternatives and specific conditions. OWL does not easily allow for modeling these complexities in a way that is easy for non-experts to manage. This creates a challenge for organizations that do not have specialists in ontology management.

#Introduction to SHACL

The Shapes Constraint Language (SHACL) provides an alternative. It allows users to model regulatory requirements in a manner that is easier to read and maintain. SHACL operates under a closed world assumption. This means that it can check if specific information is missing based on defined rules. This makes SHACL an appealing choice for organizations needing to ensure compliance with regulations.

By using SHACL, complex requirements can be modeled as constraints. This lets domain experts, who might not have deep technical knowledge, manage the requirements without needing advanced training in ontology languages.

#The Role of the Norwegian Maritime Authority

The Norwegian Maritime Authority (NMA) oversees ships and vessels in Norwegian waters. They have strict regulations to ensure maritime safety and compliance. The NMA required a new supervision system that checks if vessels and individuals meet the necessary requirements. Their previous system was outdated and inefficient.

NMA identified that modeling their requirements using traditional methods, such as OWL, was not adequate. They needed a system that could efficiently handle complex regulations with many alternatives. This led them to adopt SHACL for their modeling efforts.

#Building the New System

The NMA gathered 150 documents that contained various regulations. The goal was to model these regulations in RDF, a format that is easy to read and understand. The first attempts to create these models manually proved to be time-consuming and inefficient. It became clear that a more automated approach was necessary.

Using Natural Language Processing (NLP) techniques, the NMA was able to extract relevant concepts and relationships from the regulatory documents. This helped streamline the modeling process and reduced the time needed to create these models significantly.

NMA’s new approach allowed them to automatically generate semantic knowledge graphs that visually represent the regulatory requirements. By introducing these graphs, the cost and time required for knowledge modeling were greatly reduced.

#SHACL in Action

SHACL allows for the modeling of requirements with AND and OR conditions, making it very flexible. For example, if a regulation states that a sailor needs to fulfill different seagoing service options to get a certification, SHACL can model these conditions clearly.

A requirement can be written to state that a sailor needs either:

• 36 months of sailing experience as a deck officer, or
• 24 months of experience with at least 12 months as a chief officer.

This ability to describe alternatives helps in capturing the nuances of regulations without losing clarity.

#The Benefits of Using SHACL

Using SHACL has proved beneficial for the NMA. It allows them to create readable models that can be understood by domain experts. This was not the case with OWL, which requires a deeper understanding of formal logic and set theory.

With SHACL, domain experts can easily see the requirements and make necessary updates without needing extensive training in semantic technologies. This flexibility allows for better management of the regulations and ensures that they remain up to date.

#Validation Capabilities of SHACL

One of the strongest features of SHACL is its ability to validate data. By applying SHACL shapes, NMA can check if individual sailors meet the specified requirements for their certifications. If a sailor does not meet the requirements, the system can generate a report detailing what information is missing.

This ability to identify gaps in information is crucial for the NMA, as it helps ensure compliance and maintain safety standards in maritime operations.

#Comparison with Other Models

While SHACL is not the only language available for modeling regulatory requirements, it has distinct advantages. Other languages like SPARQL Inferencing Notation (SPIN) and Shape Expressions (ShEx) also aim to validate RDF data, but they do not provide the same level of user-friendliness as SHACL.

SHACL is designed specifically for constraints and is more expressive than SPIN. This expressiveness proves vital for handling the complex nature of regulatory data.

While SHEx provides a grammar for RDF data, SHACL focuses on validating whether the data meets specified constraints. This makes SHACL the preferred choice for organizations like NMA that require accurate assessments against regulatory requirements.

#Conclusion

In summary, the application of SHACL to model regulatory requirements has evident benefits. It allows for handling complex maritime regulations with clarity and ease, catering to the needs of domain experts without a technical background. By adopting SHACL, the Norwegian Maritime Authority is better equipped to ensure compliance and safety within the maritime industry.

SHACL's ability to handle closed world assumptions and complex constraints makes it an ideal choice for regulatory modeling. This shift simplifies future maintenance and promotes better understanding of the requirements, ensuring that all stakeholders can work effectively towards compliance.

The experience of the Norwegian Maritime Authority demonstrates SHACL's potential to enhance regulatory modeling practices in various public sectors. As organizations seek effective solutions for compliance management, the adoption of SHACL can pave the way for better governance and oversight.