Decision Intelligence Architecture at HaNonn
AI can produce Predictions, Rankings, Recommendations, and Generated Content. However, Model Output is not a Decision: using it requires Decision Context, Criteria, Evidence, Constraints, and Human Judgment.
HaNonn develops Decision Architecture and AI Decision Applications within AI-first Decision Intelligence for Customer & Commerce Decisions. This page explains how HaNonn, Intent-to-Income™, Core Architecture Components, Applications, published research, and evaluation boundaries relate without treating these layers as interchangeable.
Positioning Boundary ↓
Foundational Definitions
From Model Output to Decision Architecture
The Model-to-Decision Gap is the gap between an AI-generated Output and the Context, Criteria, Evidence, Human Judgment, and use conditions required to apply that Output to a decision.
Architecture Map
How the Architecture Layers Relate
This map separates the wider Decision Intelligence context, HaNonn as the developing organization, Intent-to-Income™ as the Reference Decision Architecture, its Core Components, and the Applications that use selected structures in context.
| Layer | Entity | Role | Evidence Boundary |
|---|---|---|---|
| Context | Decision Intelligence | The broader context for applying Data, Analytics, AI, Knowledge, and Decision Logic to decision-making | External literature supports the wider context, not the performance of HaNonn or its Applications |
| Organization | HaNonn | Develops Decision Architecture and AI Decision Applications for Customer & Commerce Decisions | Organizational positioning does not establish Application effectiveness |
| Reference Architecture | Intent-to-Income™ | Defines the Core Flow from Decision Signals to Measurable Outcomes while treating Decision Context as a cross-cutting consideration | Published as a conceptual architecture; its proposed dimensions and implementation effects still require empirical validation |
| Architecture Components | Four Core Components | Translate the Reference Architecture into journey mapping, signal interpretation, interface design, and measurement | Operational definitions and implementation logic depend on the Application and Decision Context |
| Application | AI Decision Applications | Apply selected Architecture structures and capabilities to specific decisions and domains | Performance and outcomes must be evaluated separately for each Application and Decision Context |
Evidence Does Not Automatically Transfer across Layers
Publication of the conceptual Architecture establishes a citable foundation. It does not by itself validate an implementation, Application performance, Decision Quality improvement, or business outcome.
Architecture Responsibilities
What the Architecture Coordinates
Within HaNonn, Decision Intelligence Architecture coordinates the Context, Evidence, Logic, Interface, Measurement, and Human Authority required to support a decision as one connected system.
Scope Note
These responsibilities describe how HaNonn organizes Decision Intelligence Architecture. They are not presented as a universal taxonomy for the wider Decision Intelligence field.
Reference Decision Architecture
Intent-to-Income™ as the Reference Architecture
Intent-to-Income™ is a proprietary Reference Decision Architecture designed by Kittisak Pannutiyarak and developed further within HaNonn.
It connects Decision Signals to Decision Need, Decision Support, Decision Quality, and Measurable Outcomes while using Decision Context as a cross-cutting consideration throughout the Architecture.
Decision Signals
Decision Need
Decision Support
Decision Quality
Measurable Outcomes
The Architecture provides a shared reference structure. Criteria, Evidence, Constraints, Measurement, and implementation logic must still be adapted to each Decision Context, Application, and Domain.
Conceptual and Measurement Boundary
Intent-to-Income™ Version 1.0 is a conceptual Reference Decision Architecture. Understanding, Confidence, and Readiness are proposed Decision Quality dimensions that still require operational definitions, measurement instruments, and empirical validation.
- Explore the Reference Decision Architecture
— Core Flow, Components, boundaries, and application logic - Read the Research Paper Summary
— Conceptual contribution, limitations, measurement, and future research
From Architecture to Design
Four Core Architecture Components
The Four Core Architecture Components translate the Reference Architecture into design and evaluation work. Each addresses a different architectural question and produces a different form of output.
Decision Journey Mapping
Where does the decision unfold, and where do friction or support opportunities arise?
Maps the Decision Journey, Decision Context, Signals, friction, and existing support.
A structured view of the decision process and priority support opportunities.
Decision Signal Matrix
Which observations may be relevant, and what might they indicate within the Decision Context?
Separates Observations, Interpretation, Evidence Gaps, and Proposed Decision Needs.
Traceable Signal–Context–Need relationships and Decision Support hypotheses for further validation.
Decision Interface
How should Decision Support be structured, presented, compared, and controlled?
Organizes Context, Criteria, Evidence, Alternatives, Trade-offs, UX Flow, and Next Steps.
A Decision Support Blueprint or interface structure appropriate to the Application.
Measurement Logic
What evidence is needed to evaluate the process, Decision Quality, and downstream Outcomes?
Separates Process Evidence, proposed Decision Quality indicators, and Measurable Outcomes.
A context-specific evaluation and feedback plan with explicit measurement boundaries.
Component Boundary
The Four Core Architecture Components are not additional stages in the Core Flow. They are architectural mechanisms that can be applied iteratively and adapted to each Decision Context and Application.
Application Layer
From Architecture to AI Decision Applications
An AI Decision Application is a bounded implementation designed for a defined decision and Decision Context. It may combine Data, Evidence, Decision Rules, Analytical Models, Generative AI, Interfaces, and Human Review to deliver appropriate Decision Support.
The Reference Architecture defines the relationships that must be considered. Each Application then selects and operationalizes the Components, capabilities, evidence requirements, controls, and evaluation methods appropriate to its scope.
Flow Boundary:
This Application Design Flow describes how an implementation is configured and evaluated. It is not a replacement for the Intent-to-Income™ Canonical Core Flow.
HaNonn Decision Composer™
HaNonn Decision Composer™ applies selected structures from Intent-to-Income™ to analyze product information, Product Claims, Decision Criteria, Evidence, Context, and Trade-offs, producing a Decision Support Blueprint and Product Evaluation.
Its current workflow and outputs represent the Application’s design scope, not validated evidence of performance or Decision Quality improvement.
Application ≠ Architecture ≠ Model
HaNonn Decision Composer™ is an Application, not Intent-to-Income™ itself. A Model or AI capability may operate within an Application, and its Output may serve as an input or evidence. It does not independently constitute a Decision, define the Architecture, or establish Application effectiveness. Performance and outcomes require separate evaluation.
Research & Evidence
Research and Validation Status
Intent-to-Income™ has a published conceptual foundation. Evidence for implementation, Decision Quality, and outcomes must be established separately for each Application, population, and Decision Context.
| Evidence Layer | Current Status | What It Supports | Evidence Boundary |
|---|---|---|---|
| Conceptual Foundation | Published | Provides a citable definition of the Architecture, Core Flow, Components, conceptual relationships, scope, and research agenda. | Publication does not establish construct validity, causal effects, Application performance, or business impact. |
| Application Design Evidence | Application-specific | Documents how selected Components, Decision Logic, Evidence requirements, AI capabilities, interfaces, and human controls are configured. | Design documentation can demonstrate implementation logic but does not establish effectiveness. |
| Empirical Evaluation | Not yet established | Would test whether a defined intervention affects the decision process, proposed Decision Quality indicators, or downstream outcomes. | Findings must be interpreted within the tested Application, population, methods, and Decision Context. |
Conceptual Publication ≠ Empirical Validation
The published architecture establishes terminology, structure, and research propositions. Claims about effectiveness, Decision Quality improvement, or business impact require evidence from the relevant implementation and evaluation.
External Context
Global Alignment and Practice
HaNonn’s architecture is developed independently. The sources below provide relevant external context for decision governance, AI risk management, human oversight, transparency, measurement, feedback, and user control.
Global Practice & Evidence
Independent sources relevant to Decision Intelligence Architecture and human-centered AI
- Decision Governance:
Gartner’s 2026 Data and Analytics trends identify Decision Governance as a response to the growing role of AI agents in strategic, tactical, and operational decisions, emphasizing explainability, auditability, and alignment with outcomes.
View source
- Context, Risk, Measurement, and Oversight:
The NIST AI Risk Management Framework organizes AI risk-management activity through Govern, Map, Measure, and Manage, including intended-use context, documentation, measurement, risk response, and human oversight.
View source
- Human-Centered and Trustworthy AI:
The OECD AI Principles address human-centered values, transparency and explainability, robustness, safety, and accountability throughout the AI lifecycle.
View source
- Human–AI Interaction:
Google’s People + AI Guidebook addresses user needs, success criteria, mental models, explainability, feedback, control, and the handling of AI errors in AI-enabled products.
View source
Alignment ≠ Endorsement or Validation
These sources support principles relevant to the context in which HaNonn’s architecture operates. They do not define, certify, endorse, or empirically validate HaNonn, Intent-to-Income™, its Core Components, or its Applications.
Architecture Navigation
Explore HaNonn’s Decision Intelligence Architecture
These resources address distinct layers of HaNonn’s work—from organizational positioning and Reference Architecture to published research and Application design.
The Architecture defines the relationships. Core Components translate them into design work. Applications implement selected structures in context. Evaluation determines which claims the evidence can support.