Jul 2026· 2026 6th International Conference on Electrical, Computer and Energy Technologies (ICECET)· pp. 1-8· 0 citations· 12 references
Abstract
The increasing availability of Artificial Intelligence (AI) tools has generated significant interest within the project management community; however, structured guidance tailored specifically to Project Managers remains limited. This paper proposes a standardized AI-enabled prompt architecture aligned with the PMBOK® 8E performance domains and processes. The framework consists of a Master Prompt, Task-Level Executable Prompts, and Refine Prompts designed to create bounded, context-aware interactions between Project Managers and AI systems. The proposed architecture embeds PMBOK-aligned terminology and follows the Inputs–Tools–Outputs (ITTO) logic. The AI system functions as an analytical and generative tool within this structure, processing structured and unstructured inputs—including expert judgment—and producing standardized outputs for managerial review and refinement. The architecture is designed to be generalized and extensible across all forty project management processes. This study does not present empirical performance metrics; rather, it introduces a structured conceptual framework intended for practical application and future validation. Practitioners are encouraged to implement the architecture in real project environments to evaluate measurable improvements and contribute to further academic development in AI-enabled project management.
Artificial intelligence (AI) has become an essential component of organizational strategy; however, the knowledge generated during AI projects, including design decisions, model behavior, development processes, and practitioner insights, is often insufficiently documented. This study aimed to examine how knowledge documentation is currently practiced in organizations implementing AI. Using a systematic literature review approach, this study analyzed twenty peer-reviewed studies published between 2020 and 2026. The articles were collected from five major academic databases, namely Scopus, ScienceDirect, ACM Digital Library, IEEE Xplore, and Emerald Insight, and were selected following the PRISMA 2020 guidelines. Based on five research questions structured using the PICOC framework, several key findings emerged. First, knowledge management in AI contexts has shifted from static repositories toward more dynamic and AI-supported systems, although this transition introduces challenges such as model drift, inconsistent documentation practices, and difficulties in capturing tacit knowledge. Second, the literature presents various knowledge documentation methods and frameworks, indicating that the field remains in development without a universally accepted standard. Third, structured documentation has been shown to positively influence organizational learning, knowledge reuse, and the overall effectiveness of AI initiatives. Fourth, despite these benefits, significant gaps remain, particularly regarding the absence of standardized AI/machine learning (ML) documentation practices and the limited integration of documentation throughout the AI lifecycle. In response to these challenges, this study proposed the Knowledge Documentation Framework for AI Initiatives (KDF-AI), consisting of twelve components organized into five phases and supported by different maturity levels. Overall, this review highlighted the increasing importance of knowledge documentation as a core capability in AI-driven organizations and provided a foundation for future research and practical implementation.
Finannisa Zhafira, Fitria Handayani, Dana Indra Sensuse et al.· Jurnal Impresi Indonesia· 0 citations
A research framework for analysing the added value of generative AI in logistics organizations, with a focus on “difficult to automate” tasks and processes is presented.
Gerald Schneikart, Walter Mayrhofer· Engineering review· 0 citations
Despite the growing organizational reliance on artificial intelligence (AI) systems, knowledge documentation (KD) practices in AI initiatives remain largely ad hoc, unstandardized, and disconnected from project lifecycle management. This study addressed this gap by proposing and validating the Knowledge Documentation Framework for AI Initiatives (KDF-AI), a twelve-component, five-phase, maturity-tiered framework synthesized through a literature review of peer-reviewed studies on KD practices in organizations implementing AI. Using Design Science Research (DSR) as the methodological paradigm, the study completed two iterative cycles: a literature-based synthesis that produced KDF-AI v1 and an expert evaluation cycle that resulted in the refined KDF-AI v2. Expert content validation was conducted with three domain validators representing academic, governance, and AI practitioner perspectives using a mixed-method approach that combined quantitative Content Validity Index (CVI) assessment with deductive thematic analysis of semi-structured interviews. The results showed that 57 of 66 items (86.4%) achieved universal inter-rater agreement, producing S-CVI/UA = 0.864 and S-CVI/Ave = 0.955, both exceeding the recommended threshold of 0.80. The nine items that did not meet the threshold consistently reflected issues of clarity rather than relevance, indicating strong conceptual acceptance of the framework while highlighting the need for more operationally specific articulation in several Advanced-tier components. Nine targeted revisions resulted in the development of KDF-AI v2. The study contributed: (1) a validated lifecycle-integrated KD framework for AI initiatives; (2) a taxonomy of ten systematically identified gaps in current AI KD practices; and (3) a methodological demonstration of mixed-method CVI validation for framework development in information systems research.
Fitria Handayani, Finannisa Zhafira, D. Sensuse et al.· Jurnal Impresi Indonesia· 0 citations
Integrating AI agents into Software Engineering (SE) raises an important challenge: how can we specify and realize AI agents that work effectively alongside humans in hybrid SE teams? Determining the right granularity and separation of concerns for such agents is non-trivial. Coarse-grained agents may introduce unmanageable complexity, whereas micro-agents may create severe coordination overhead. Moreover, existing multi-agent SE frameworks typically rely on predefined role structures and do not account for project-specific characteristics or process adaptations. We address this by combining object-centric, imperative, and declarative process mining. Using event logs extracted from software repositories, our approach discovers project-specific agent roles using a predefined SE role vocabulary grounded in repository behavior and generates matching agent specifications and implementations. As proof-of-concept, we applied our approach to a well-established open-source project. We performed functional tests and an exploratory user study to determine how well the generated AI agent specifications are aligned with human expectations.
Saimir Bala, Fabiana Fournier, Lior Limonad et al.· 0 citations
Abstract The development and potential of large language models created incentives for organizations to adopt and integrate this technology in their internal processes. When the technological layer of organizations is in an acceleration phase there is a strong need to manage this change. Currently, generative artificial intelligence brings a vast area of opportunities, but this can create an “analysis paralysis” when deciding what initiatives to pursue. Different frameworks exist but they are targeting software solutions or specific domains. This paper follows an exploratory, framework-development qualitative design. An evaluation framework to assess initiatives based on generative artificial intelligence technology is developed by reusing existing decision models and enhancing them with criteria based on a simulated focus group using large language model personas. The interim findings were analysed using directed qualitative content analysis (deductive coding to a baseline codebook derived from literature), complemented by qualitative methods (free listing of evaluation criteria, anchored rating-scale elicitation, priority ranking, mapping to baseline criteria). A derived seven-criteria framework with definitions and performance scales resulted. This allows organizations to maximize chances of choosing the highest-reward initiatives and invest resources efficiently when needed to select, implement and integrate generative artificial intelligence solutions. The intention of this paper is to propose a list of seven evaluation dimensions — domain effectiveness and impact, regulatory compliance and ethical risk, user adoption, business value, technical feasibility, data availability and quality, and competitive differentiation that can be used to characterize applications based on this technology. [ms1][ms2]
Marius Sava, G. Militaru· Proceedings of the Internati...· 0 citations
Rapid advances in Information Technology (IT) and Artificial Intelligence (AI) have resulted in increasingly complex socio‑technical systems, placing new demands on human–automation collaboration. These demands are particularly acute in VUCA (Volatile, Uncertain, Complex, Ambiguous) environments, where failures can have immediate and severe consequences. Effective human‑automation teams must therefore adapt task and control allocation dynamically, while maintaining safety, accountability, and operator understanding. Adaptive automation (AA) has been widely studied as a means to support such flexibility, often described using Levels of Automation (LoA) frameworks. However, many existing LoA frameworks do not align well with decision‑cycle models commonly used to reason about responsibility distribution in VUCA contexts. This misalignment complicates the design and reuse of AA-solutions for real‑world applications. Team Design Patterns (TDPs) offer a promising approach by capturing reusable solutions to recurring challenges in human–automation teamwork. Yet, the systematic construction of coherent TDP sets remains difficult due to a lack of structured design support. To address this, we present an integrated human‑automation teaming framework that facilitates TDP development and supports cross‑disciplinary dialogue between designers, engineers, command staff, and policy‑makers. The framework extends NASA’s eight‑level LoA framework by assigning descriptive level names, explicit human‑loop status (in/on/out‑of‑the‑loop), and visual responsibility mapping inspired by the European Defence Agency’s methods‑of‑control framework. Based on a naval Uncrewed Surface Vessel use case, we developed and evaluated three TDPs with domain experts: uniform task delegation, uniform goal delegation, and a mixed‑level pattern combining different LoA across decision-cycle stages. Expert evaluation confirmed that mixed‑level TDPs best reflect operational, regulatory, and technological realities. Overall, the framework provides a structured basis for designing flexible and context‑appropriate adaptive automation in VUCA environments.
Jelle A Van Dijk, Rosa van Tuijn, Renske Verwaal-Bootsma et al.· AHFE International· 0 citations