Persona-as-Configuration: Generative Stakeholder Reporting for Agricultural Floods

2026-07-20

AI Summary

This study addresses the challenge of generating auditable and replayable customized reports for heterogeneous stakeholders in agricultural flood-edge detection systems, a requirement fundamentally at odds with the inherent non-determinism of large language models (LLMs). To reconcile this tension, the authors propose an architectural pattern that integrates deterministic edge inference with LLM-based generation through a “role-as-configuration” paradigm, wherein user roles are codified into versioned prompt templates. Coupled with a context-aware dashboard that synthesizes structured JSON decision logs into reliable, role-specific reports, the approach enforces separation of concerns and ensures system auditability. The design received strong endorsement from domain experts under the ISO/IEC 25010 quality model, with field evaluations by agricultural end-users planned as future work.

Abstract

Cyber-physical systems built on deterministic edge inference, such as on-vehicle flood detection for agricultural fields, produce structured decision logs that must be interpreted differently by heterogeneous stakeholders. Pairing such systems with large language models (LLMs) to generate stakeholder-specific reports introduces a tension: the generative layer is non-deterministic, while the edge plane must remain replayable and auditable. We propose an architectural pattern resting on two invariants: unidirectional consumption, in which the generative layer is a strict read-only consumer of the deterministic plane and never writes back, and persona-as-configuration, in which stakeholder adaptation is a versioned prompt-template artifact rather than runtime improvisation. We instantiate the pattern as a context-aware dashboard layer over the JSON decision logs of a previously published edge-based standing-water detection system, and analyse how the integration boundary admits standard generative-reliability mitigations as configuration- or middleware-level extension points. A structured expert review rated the pattern favourably across five ISO/IEC 25010-aligned quality dimensions, with strongest agreement on separation of concerns. End-user evaluation with agricultural stakeholders is planned for future work.

Problem

• stakeholder reporting
• deterministic edge inference
• non-deterministic generation
• auditability
• cyber-physical systems

Innovation

• persona-as-configuration
• deterministic edge inference
• generative stakeholder reporting
• unidirectional consumption
• prompt templating