Managing Administrative Law Cases using an Adaptable Model-driven Norm-enforcing Tool
Marten C. Steketee, Nina M. Verheijen, L. Thomas van Binsbergen · 2025
Read Managing Administrative Law Cases using an Adaptable Model-driven Norm-enforcing Tool by Marten C. Steketee, Nina M. Verheijen, L. Thomas van Binsbergen (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2510-24822.txt.
Introduction
Marten C Steketee, Nina M Verheijen, and L Thomas van Binsbergen built a tool to run state law cases with rules baked in. State staff must track many laws and plans as they handle each case, most of it by hand. Shifting law reads and odd cases breed slips and slow work, which hurts fair reach to courts.
Their tool drives the case flow from a model, and runs a rules tongue to check each step as it goes. That way, set steps are forced to be kept, and the grounds for a call can be shown to the citizen. Needs for the tool came from past field work with staff. The aim is fewer slips and more clear calls.
Catalog record
| CREATOR | Marten C. Steketee, Nina M. Verheijen, L. Thomas van Binsbergen |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 28 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2510-24822
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Managing Administrative Law Cases using an Adaptable Model-driven Norm-enforcing Tool Marten C. Steketee, Nina M. Verheijen, L. Thomas van Binsbergen Abstract Governmental organisations cope with many laws and policies when handling administrative law cases. Making sure these norms are enforced in the handling of cases is for the most part done manually. However, enforcing policies can get complicated and time consuming with ever-changing (interpretations of) laws and varying cases. This introduces errors and delays in the decision-making process and therefore limits the access to justice for citizens. A potential solution is offered by our tool in which norms are enforced using automated normative reasoning. By ensuring the procedural norms are followed and transparency can be provided about the reasoning behind a decision to citizens, the tool benefits the access to justice for citizens. In this paper we report on the implementation of a model-driven case management tool for administrative law cases, based on a set of requirements elicited during earlier research. Our tool achieves adaptability and norm enforcement by interacting with an interpreter for eFLINT, a domain-specific language for norm specification. We report on the current state of the case management tool and suggest directions for further development. [orcid=0009-0007-8297-1909, [email protected], url=https://martensteketee.nl/, ] *1 1 [orcid=0009-0005-2728-9064, [email protected], url=, ] 1 [orcid=0000-0001-8113-2221, [email protected], url=http://ltvanbinsbergen.nl/, ] keywords Model-based reasoning ,Government policies ,Model-driven ,Software tools ,Compliance ,Business processes † † copyrightyear: 2024 † † copyright: Copyright for this paper by its authors. Use permitted under Creative Commons Attribution-ShareAlike 4.0 International (CC BY-SA 4.0). † † venue: Jurix’24: AI for Access to Justice Workshop, December 11, 2024, Brno, Czechia † † address: Informatics Institute, University of Amsterdam, Science Park 900, 1098 XH Amsterdam, The Netherlands † † corresp: Corresponding author. † † These authors contributed equally. 1 Introduction Digitalisation of governmental services has increased in popularity in recent years, and aims to improve the efficiency and effectiveness of public administration. However, there is not always a direct connection between the digital systems used for these services and the relevant norms. This disconnection is clearly illustrated by administrative law cases, for which large parts of the decision-making process is done manually, while the communication and registration of cases is automated. Manually keeping track of the different norms while working on a case is difficult and prone to errors. A potential solution is provided by automated reasoning about compliance according to formalised representations of norms. In this paper we present a case management tool for supporting civil servants with decision-making. Although the focus in our research was governmental case management, the tool can also be used for other business processes by changing the eFLINT model. The norm specification language eFLINT is used to reason about the applicable norms van Binsbergen et al. 2020 . The current version of the tool is presented in this paper, along with suggestions for further development. The tool adapts to changes to the (process) model and the norms encoded in the eFLINT specification by changing the user flows of the system. In our work, we apply methods and concepts from business process modelling, model-driven engineering, and normative reasoning. This paper contributes: • an adaptable case management tool that reasons about applicable norms; • an evaluation against elicited user requirements resulting in suggestions for further development. 2 Background 2.1 Normative reasoning/eFLINT eFLINT is a domain-specific language (DSL) for specifying and reasoning with norms van Binsbergen et al. 2020 . The language is developed as a DSL to simultaneously specify norms from various sources, as well as (models of) computational processes that can be interconnected with the specified norms. Given a specification of norms, the eFLINT reasoner can be integrated into systems to check compliance of actions taken in the system. For the running example, we translated rules for the quittance of municipal taxes of Amsterdam 2024 into eFLINT code. This involves adding powers and duties for the different steps in the application process. To determine whether a client is entitled to quittance of their taxes, the client must provide the public servant with information about their financial position: current income, age, and marital status. The model uses eFLINT facts to capture the information of an application and setting thresholds for income. In the fragment below, the Var and Bool keywords are used to ensure that at most one instance of these types holds true at any given time. ⬇ Open Var [ Age ] Identified by Int Open Bool [ Pension age reached ] eFLINT/model.eflint The thresholds for capital and income are specified using Var and are immediately set to a certain value using the following statement. ⬇ Var [ income limit no retirement partner ] Identified by Int . +[ income limit no retirement partner ](1574). eFLINT/parameters.eflint The eFLINT reasoner can determine which duties are currently present, and which actions can be taken once the applicant has provided enough information. This makes sure that the user is informed about compliant actions but the user isn’t forced to act in a compliant manner. The reasoner is able to cope with incompliant behaviour and will show violations of the norms to the user. The Physical keyword marks actions which can be executed by users. These synchronise with their institutional counterparts. ⬇ Act grant - request Actor servant Recipient applicant Conditioned by [ Income ] <= [ Income limit ] Creates [ Quittance granted ]. Physical send - approval - letter Syncs with grant - request ( servant , applicant ) eFLINT/acts.eflint A duty to process the application is created as soon as a client applies for an allowance. Duties can be associated with violation conditions (one or more Boolean expressions) to determine when a duty is considered violated. In this example, no violation condition is associated with the duty. A very important aspect of the eFLINT reasoner is the transparancy the system provides. The reasoner is able to show the current state of the model and provide a step by step explanation of how this state has been reached. This enables civil servants and citizens to get a better understanding of the followed procedure and reasoning behind a decision, making it easier to judge wether filing an objection is necessary. 2.2 User requirements During earlier research, interviews were held with potential users to gain insight in their requirements for a norm-enforcing case management tool Verheijen 2024 . This research was done following a user-centred design method Martin et al. 2012 . The seven participants, working at four different organisations, were either involved with handling cases from civilians or organisations subjected to laws and policies, or were involved in the process of turning laws and policies into actions, e.g., within a software application. The participants that worked for the same organisation had different roles within their organisations. Requirements were elicited from the interview results using a thematic analysis Braun and Clarke 2006 . These were used as input for the design of the tool. User tests were performed to test the user friendliness of the tool and ensure it was fit for purpose. Following the user tests, the tool was further developed to demonstrate the ability to reuse the system for different use cases and adapt to changes in norms. This section will discuss the