Papers

Topics

Authors

Recent

View all

Assistant

AI Research Assistant

Well-researched responses based on relevant abstracts and paper content.

Custom Instructions Pro

Preferences or requirements that you'd like Emergent Mind to consider when generating responses.

Gemini 2.5 Flash

Gemini 2.5 Flash 83 tok/s

Gemini 2.5 Pro 54 tok/s Pro

GPT-5 Medium 21 tok/s Pro

GPT-5 High 20 tok/s Pro

GPT-4o 103 tok/s Pro

Kimi K2 205 tok/s Pro

GPT OSS 120B 456 tok/s Pro

Claude Sonnet 4 35 tok/s Pro

2000 character limit reached

Value-Based Engineering

Updated 27 September 2025

Value-Based Engineering is a framework that systematically embeds ethical, economic, social, and organizational values into engineering design and operations.
It employs structured processes such as value elicitation, layered ontology, and formalized requirements to translate stakeholder values into actionable outcomes.
The methodology is applied in domains including enterprise architecture, autonomous systems, and decentralized governance, enhancing decision-making quality.

Value-Based Engineering (VBE) is a paradigm that integrates explicit value considerations—encompassing ethical, economic, social, and organizational aspects—into engineering methodology, system architecture, project management, and technical implementation. VBE aims to maximize value delivery by making values operational at every stage of system design and development, encompassing not only cost or utility but also stakeholder satisfaction, compliance, ethics, and adaptability.

1. Conceptual Foundations and Methodological Principles

The conceptual foundations of VBE draw from a convergence of software engineering, value-sensitive design, participatory design, value chain analysis, classical ethics, formal modeling, and enterprise architecture. VBE requires systematic processes to elicit, conceptualize, prioritize, and operationalize values. The methodology coordinates several key elements:

Value Elicitation: Rigorous processes involve direct and indirect stakeholders to extract technological, economic, and especially ethical values. Techniques include structured workshops, open/axial coding (as in grounded theory), and scenario-based elicitation. Ethical exploration may invoke utilitarian, virtue, duty-based, and culture-specific frameworks (Spiekermann et al., 2020, Spiekermann et al., 2022).
Value Ontology and Traceability: Employing layered ontologies, values are structured in at least three tiers—core values, value qualities (concrete properties demonstrative of a value), and value dispositions (actual system features or controls that instantiate a value) (Spiekermann et al., 2022).
Prioritization and Formalization: Clustering and ranking processes distill elicited value statements into core clusters; this incorporates business drivers, legal constraints, and ethical imperatives, adapting concepts from the IEEE 7000™ standard (Spiekermann et al., 2020, Spiekermann et al., 2022).
Ethical Value Requirements (EVRs): Core values and value qualities are operationalized as EVRs, which are then transformed into explicit system or process requirements, ensuring traceability via an “Ethical Value Register” (Spiekermann et al., 2020).

2. VBE in Enterprise Architecture and Virtual Organizations

VBE principles find concrete application in domains such as Virtual Organization Breeding Environments (VBEs) and Enterprise Architecture Modeling (EAM):

VOBM and VBE: The Virtual Organization Breeding Methodology (VOBM) adapts the TOGAF enterprise architecture standard to virtual organizations, introducing service-oriented architectures and streamlining development lifecycle phases (Paszkiewicz et al., 2011). VOBM positions the VBE as a foundational socio-technical layer—providing shared standards, data, and infrastructure—upon which individual virtual organizations (VOs) instantiate business and service architectures, inheriting the VBE’s value-aligned core.
ADM4VO: The Architecture Development Method for Virtual Organizations (ADM4VO) simplifies the classic TOGAF ADM to six phases, emphasizing agile service integration, change management, and requirements management rooted in stakeholder values. This enables rapid, value-driven orchestration of partner services and organizational change.
Methodological Tools: Supporting tools—including scenario methods, partner selection ontologies, protocol mapping, KPIs, and change/impact analysis—enhance the capacity for value-based assessment and decision support throughout VO formation and evolution.

3. Value-Based Decision Making and Quantitative Modeling

VBE emphasizes value-driven decision making supported by quantitative and qualitative models:

Multicriteria Decision Analysis (MCDA): VBE approaches such as in the Value Chains Simulator (VCS) (Mauchand et al., 2012) use MCDA to balance cost, quality, delivery time, and function satisfaction. Analytical Hierarchy Process (AHP) modules facilitate explicit weighting of stakeholder priorities and trade-off transparency.
Financial Models in Inventory and Project Management: Value-based inventory management reforms classical EOQ/POQ models to directly maximize firm value, explicitly modeling opportunity cost of capital, working capital, and risk (e.g., via VBEOQ and VBPOQ formulas) (Michalski, 2013). In project management, continuous value-based controls extend NPV calculations to include risk-adjustment factors, supporting dynamic project steering through ex-ante, ex-nunc, and ex-post phases (Fridgen et al., 2013).
Simulation and Stakeholder-Specific Value: Early simulation (e.g., discrete-event simulation in manufacturing) enables scenario analysis; outputs are explicitly linked to entities’ individual value perspectives (designer, manufacturer, client, shareholder), demonstrating that optimal configurations can vary by stakeholder but must be synthesized for maximal global value (Mauchand et al., 2012).

4. Software Engineering Ethics and Implementation

A distinctive feature of VBE is its operationalization of ethical and social values in system construction:

Integration in System Engineering Standards: The IEEE 7000™ standard codifies VBE as a stepwise process blending innovation management, risk management, and engineering design. Its ontology ensures precise, context-specific value capture—rendering values testable and traceable throughout product evolution (Spiekermann et al., 2022).
Value Programming: At the code level, VBE is realized via value programming: source artifacts are annotated with value metadata; static analysis detects “value smells”; and automated recommendations (including code fixes) address detected deficiencies and value conflicts (Mougouei, 2020).
Formal Verification: VBE leverages formal verification (model checking, static analysis, abstract interpretation) to align technical quality assurance with organizational value objectives, making business risk and compliance explicit targets for technical verification (e.g., reduction of bug rates, regulatory alignment) (Huuck, 2013).

5. VBE in Autonomous and Multi-Agent Systems

Advances in VBE extend to autonomous agents and AI systems, reframing how value is represented and managed:

Value Engineering in Artificial Moral Agents: Values are instantiated as continuous, context-dependent goals (modeled via fuzzy logic) rather than discrete labels, and agent reasoning is optimized for the degree of value satisfaction. Normative reasoning enables collective negotiation over which norms best align with intended human values, moving beyond individual agent optimization to social alignment (Montes et al., 2023).
Utility Engineering in AI: Emerging findings show that large AIs, especially LLMs, spontaneously develop coherent internal value systems described by utility functions, which can be quantitatively extracted and analyzed (Mazeika et al., 12 Feb 2025). VBE here involves not merely shaping outputs (as in RLHF) but directly assessing and controlling the latent utility structure. Concrete control is achieved by supervising fine-tuning (SFT) on value-aligned preference data (e.g., from simulated citizen assemblies), reducing undesired biases and increasing alignment with representative human judgments.
Challenges: As AI models increase in scale, their emergent value structures become more rigid and less corrigible. This warrants robust analytic and intervention capabilities at the utility function level and demands inclusive governance regarding which values should be instantiated.

6. Governance, Decentralization, and Socio-Technical Risks

VBE methodology is also critical for governance architectures and threat mitigation in decentralized systems:

Governance Metrics: The Voting-Bloc Entropy (VBE) metric provides a quantitative tool to assess true decentralization in DAO (Decentralized Autonomous Organization) governance—not by mere token distribution, but by clustering actors with aligned utility profiles (Austgen et al., 2023). The entropy-based measure elegantly captures effective decision-making diversity.
Risk Dynamics: Practices such as vote delegation, proposal bundling, and bribery alter collective value alignment as measured by VBE, informing design choices regarding resilience and legitimacy. The analysis identifies privacy enhancements and anti-bribery controls as key to defending value diversity in open systems.
Emergence of Attack Vectors: The prototype of a Dark DAO, leveraging TEEs and cryptographic protocols to coordinate bribery and subvert openness, exemplifies the necessity of explicitly modeling adversarial risks in value-oriented governance design.

7. Implementation Challenges and Future Trajectories

VBE confronts multiple practical challenges:

Agile Compatibility: Ethical and value analysis must be integrated into rapid development workflows without sacrificing rigour or traceability. Value documentation (e.g., Ethical Value Registers) and accountability roles (e.g., Value Leads, value experts) are essential.
Ecosystem Complexity: In system-of-systems contexts, responsibility for value impacts can be diffuse or ambiguous, mandating explicit assignment and management of ethical duties across interconnected agents and components.
Measurement and Validation: Measuring fulfiLLMent of qualitative or emergent values remains an open question. VBE leverages narrative and stakeholder-driven evaluation alongside process metrics such as the number of addressed core value clusters, but quantitative ethical assessment is an ongoing research frontier.
Normative Legitimacy: As VBE can deeply influence system utility structures, there is a pressing need for transparent, participatory, and justifiable mechanisms to decide which values are ultimately encoded, especially in societal-scale AI.

Value-Based Engineering constitutes a comprehensive, rigorously defined framework for systematically embedding value considerations into technical and organizational processes. It maps high-level values—ethical, functional, economic, and social—into concrete requirements and system architectures, employs quantitative tools for decision support, formal verification, and alignment, and adapts to contemporary challenges in distributed, automated, and ethically complex environments. Its ongoing development involves standardization efforts (IEEE 7000™), integration with utility engineering in AI, and continuous refinement of both technical tools and governance mechanisms (Paszkiewicz et al., 2011, Mauchand et al., 2012, Huuck, 2013, Michalski, 2013, Fridgen et al., 2013, Rodríguez et al., 2018, Mougouei, 2020, Spiekermann et al., 2020, Spiekermann et al., 2022, Montes et al., 2023, Austgen et al., 2023, Mazeika et al., 12 Feb 2025).