اخبار و اعلانات
آمار
| تعداد نشریات | 38 |
| تعداد شمارهها | 1,414 |
| تعداد مقالات | 10,149 |
| تعداد مشاهده مقاله | 11,999,220 |
| تعداد دریافت فایل اصل مقاله | 7,004,800 |
طراحی مدل سامانۀ دانشیِ همزیست مبتنی بر هوش مصنوعی مولد: از مدیریت منابع انسانی مثبتگرا تا خلق هوش جمعی راهبردی | ||
| مدیریت راهبردی دانش سازمانی | ||
| مقالات آماده انتشار، پذیرفته شده، انتشار آنلاین از تاریخ 15 تیر 1405 | ||
| نوع مقاله: مقاله پژوهشی با اصالت | ||
| شناسه دیجیتال (DOI): 10.47176/SMOK.2026.2029 | ||
| نویسندگان | ||
| تیمور مرجانی* 1؛ علی داوری2؛ علی سعداله3 | ||
| 1گروه مدیریت، دانشکده علوم انسانی، دانشگاه علم و فرهنگ، تهران، ایران | ||
| 2گروه کسب و کار جدید، دانشکده کارآفرینی، دانشکدگان مدیریت، دانشگاه تهران، ایران، تهران | ||
| 3گروه مهندسی مکانیک، دانشکده فنی و مهندسی، دانشگاه علم و فرهنگ، تهران، ایران | ||
| تاریخ دریافت: 07 خرداد 1405، تاریخ بازنگری: 02 تیر 1405، تاریخ پذیرش: 12 تیر 1405 | ||
| چکیده | ||
| هدف: هوش مصنوعی مولد فرصتهای نوینی برای مدیریت دانش و منابع انسانی ایجاد کرده است، اما الگویی که بهطور همافزا شکوفایی کارکنان و خلق دانش راهبردی را تلفیق کند، به ندرت دیده میشود. این پژوهش با هدف طراحی مدل سامانۀ دانشی همزیست مبتنی بر هوش مصنوعی مولد و تبیین روابط علّی میان مؤلفههای آن انجام شده است. روش پژوهش: پژوهش با رویکرد آمیخته اکتشافی انجام شد. در مرحلۀ کیفی، با دلفی فازی و مشارکت ۱۵ خبره، مؤلفههای اولیه استخراج گردید. در مرحلۀ کمی، با دیمتل فازی روابط علّی میان مؤلفهها تحلیل شد و سپس با نگاشت شناختی فازی و شبیهسازی عاملمحور در افق پنجساله، پویاییهای سامانه مدلسازی گردید. اعتبار محتوا با CVR و پایایی با ضریب کندال (۷۳/۰) تأیید شد. یافتهها: پنج مؤلفۀ کلیدی شناسایی شد: شفافیت الگوریتمی (علیت ۲۴/۱+)، حفظ تنوع شناختی (علیت ۷۸/۳+)، بازخورد تعامل انسان ماشین (مرکزیت ۰۵/۱۰)، یادگیری شخصیسازیشده مولد (مرکزیت ۹۸/۹) و حکمرانی اخلاقی هوش مصنوعی (مرکزیت ۳۷/۹). «حفظ تنوع شناختی» با بیشترین مقدار علیت (۷۸/۳+) به عنوان قویترین متغیر علّی شناسایی شد. شبیهسازی نشان داد سامانۀ کامل، ظرفیت خلق دانش راهبردی را ۸۷ درصد افزایش، نرخ فرسودگی شغلی را از ۳۴ به ۱۲ درصد کاهش و دقت تصمیمگیری را ۴۸ درصد بهبود میبخشد. بحث: یافتهها نشان میدهند برخلاف رویکردهای سنتی کارآمدی محور، حفظ تنوع شناختی موتور محرک همافزایی دانش و بهزیستی است. شفافیت الگوریتمی، اعتماد و بازخورد را ممکن میسازد و یادگیری شخصیشده را تحت حکمرانی اخلاقی هدایت میکند. الگوی پیشنهادی شکاف ادبیات را با ارائه روابط پویا و تجویزی پر میکند. نتیجهگیری: این پژوهش چارچوبی یکپارچه برای پیوند مدیریت منابع انسانی مثبتگرا، هوش مصنوعی مولد و هوش جمعی راهبردی ارائه میدهد. پیشنهاد میشود سازمانها با اولویتدهی به حفظ تنوع شناختی و شفافیت الگوریتمی، سامانههای دانشی همزیست را بهتدریج پیادهسازی کنند. محدودیت اصلی، تعمیمپذیری محدود به صنایع دانشبنیان ایران است و پژوهشهای آتی باید اعتبارسنجی تجربی در صنایع دیگر را دنبال کنند. | ||
| کلیدواژهها | ||
| سامانۀ دانشی همزیست؛ شبیهسازی عامل محور؛ مدیریت منابع انسانی مثبتگرا؛ هوش جمعی راهبردی؛ هوش مصنوعی مولد | ||
| عنوان مقاله [English] | ||
| Designing a Symbiotic Knowledge System Based on Generative Artificial Intelligence: From Positive Human Resource Management to Strategic Collective Intelligence | ||
| نویسندگان [English] | ||
| Taimoor Marjani1؛ Ali Davari2؛ Ali Sadollah3 | ||
| 1Department of Management, Faculty of Humanities, University of Science and Culture, Tehran, Iran | ||
| 2Department of New Business, Faculty of Entrepreneurship, College of Management, University of Tehran, Tehran, Iran | ||
| 3Department of Mechanical Engineering, Faculty of Engineering, University of Science and Culture, Tehran, Iran | ||
| چکیده [English] | ||
| Abstract Purpose: The rapid proliferation of generative artificial intelligence (GenAI) is fundamentally reshaping organizational knowledge creation, while positive human resource management (PHRM) has emerged as a paradigm prioritizing e mployee flourishing alongside performance. However, a critical theoretical gap exists: no prior study has systematically integrated GenAI, PHRM, and strategic collective intelligence into a unified, dynamic model. Existing literature focuses either on AI-driven efficiency in HR processes or on isolated well-being interventions, leaving the synergistic potential of a symbiotic knowledge system unexplored. The central problem is that organizations lack a causal framework to design and implement GenAI that simultaneously fosters knowledge creation and employee psychological capital without sacrificing cognitive diversity or ethical governance. This study aims to: (1) identify the core components of a symbiotic knowledge system based on GenAI; (2) map causal relationships among these components; and (3) simulate the dynamic co flourishing loop between employee well being and strategic knowledge creation over a five year horizon. The research answers: What are the key components and causal dynamics of a GenAI mediated symbiotic knowledge system that enhances both employee flourishing and strategic collective intelligence?. Methodology: This study employs an exploratory sequential mixed method design grounded in pragmatism, comprising three stages. Stage 1 (Qualitative – Fuzzy Delphi): A systematic literature review of 412 articles narrowed to 97 relevant studies, yielding an initial pool of 34 components. A panel of 15 experts (HRM, knowledge management, AI, positive psychology) rated components over three rounds using a five point fuzzy linguistic scale (very low to very high). Triangular fuzzy numbers were defuzzified via center of gravity. Retention criterion: defuzzified mean ≥ 0.75 and Kendall’s W significant (p < 0.05). Stage 2 (Quantitative causal modeling – Fuzzy DEMATEL): The final components were structured into a pairwise comparison matrix. The same experts evaluated direct influences using the fuzzy scale. Fuzzy DEMATEL computed D_i (influence given), R_i (influence received), D_i+R_i (centrality), and D_i−R_i (net cause/effect). Threshold = 0.5780 (mean expert weight). Stage 3 (Dynamic simulation – Fuzzy Cognitive Mapping + Agent Based Modeling): Causal weights were visualized using FCMapper. Then, an agent based model (NetLogo 6.4) simulated 100 agents (employees) with initial psychological capital scores derived from a real Iranian knowledge based IT firm (baseline survey, N=120). Four scenarios were tested: baseline (no system); only C1 (algorithmic transparency); only C2 (cognitive diversity preservation); full system (all five components). Simulation ran for 60 monthly ticks (5 years). Outcome variables: strategic knowledge creation capacity (new knowledge artifacts per quarter), employee burnout rate (emotional exhaustion proxy), and strategic decision accuracy (proportion of correct choices). Validity: content validity ratio (CVR = 0.82); inter rater reliability (Kendall’s W = 0.73, p < 0.01). The ABM was calibrated with historical data (2023–2025) from the participating firm. Results: Fuzzy Delphi reduced 34 components to five core components: C1 (Algorithmic Transparency), C2 (Cognitive Diversity Preservation), C3 (Human AI Interaction Feedback), C4 (Generative Personalized Learning), and C5 (Ethical AI Governance). Fuzzy DEMATEL revealed that C2 (Cognitive Diversity Preservation) had the highest net causal effect (D_i−R_i = 3.78), followed by C1 (1.24). C3, C4, and C5 were effect variables (negative D_i−R_i). Centrality (D_i+R_i) was highest for C3 (10.05). All causal weights were positive, indicating reinforcing relationships. The agent based simulation showed: Under the full system (Scenario 4), strategic knowledge creation capacity increased by 87% compared to baseline (from 100 to 187 index points). Employee burnout rate dropped from 34% to 12% (a 22 percentage point reduction). Strategic decision accuracy improved by 48% (from baseline 50% to 74% correct). Even the partial scenario with only C2 (cognitive diversity preservation) yielded a 34% increase in knowledge creation and a 12% reduction in burnout. Sensitivity analysis confirmed C2 as the most influential parameter (sensitivity coefficient = 0.72), followed by C1 (0.54). The co flourishing loop exhibited clear reinforcement: reduced burnout → higher psychological capital → increased engagement with GenAI → more novel knowledge outputs → further personalized learning and well being gains.. Discussion: The findings provide three major theoretical insights. First, cognitive diversity preservation (C2) as the primary causal driver challenges the efficiency dominated AI design paradigm. Unlike prior work that treats bias reduction as sufficient, our model shows that actively preserving – not merely avoiding harm to – cognitive diversity is a strategic engine for both knowledge creation and well being. Second, the simulation results demonstrate that employee flourishing and strategic knowledge creation are mutually reinforcing, contradicting the traditional productivity well being trade off assumption. This supports the emerging view that well being precedes high performance and extends it to collective intelligence. Third, algorithmic transparency (C1) enables trust, which activates feedback loops (C3) and personalized learning (C4) under ethical governance (C5). This causal chain provides a practical roadmap. The full system’s 87% increase in knowledge creation highlights synergy (whole greater than sum of parts) – a novel property not present in prior static causal models. These findings fill the identified gap by moving from descriptive AI-HRM correlations to a dynamic, prescriptive symbiotic model. Conclusion: This research makes three original contributions. Theoretically, it provides the first integrated framework linking PHRM, GenAI, and strategic collective intelligence through a symbiotic knowledge system, with cognitive diversity preservation as the core engine. Methodologically, the combination of fuzzy Delphi, fuzzy DEMATEL, FCM, and agent based simulation offers a rigorous, replicable approach for socio technical system design. Practically, the five components and causal map offer actionable guidelines for human centric AI implementation in knowledge based organizations. Limitations: Single simulation calibrated with Iranian IT data limits generalizability; five year horizon may miss long term cultural shifts; model assumes rational agent behavior. Future research: Empirical validation through longitudinal case studies across industries; cross cultural comparisons; algorithm development for C2 (diversity promoting regularization); investigation of potential downsides (algorithmic fatigue, over reliance). In conclusion, a symbiotic knowledge system based on GenAI, when designed with transparency, diversity preservation, feedback, personalization, and ethical governance, can simultaneously advance organizational knowledge creation and employee flourishing – moving beyond AI as a cost cutting tool toward augmented collective intelligence. Acknowledgments The authors sincerely thank the 15 expert panel members for their dedicated participation in the fuzzy Delphi and fuzzy DEMATEL processes. We also acknowledge the cooperation of the participating knowledge based IT firm in Iran for providing baseline employee psychological capital data used to calibrate the agent based simulation. No generative AI tools were used to produce content, data, analysis, or conclusions; AI was used solely for language polishing and reference formatting, as disclosed in compliance with journal policy. The authors have no other individuals or institutions to acknowledge beyond those listed. Funding This research received no specific grant from any funding agency in the public, commercial, or not for profit sectors. The author declare that no financial support, research contracts, grants, institutional support, economic interests, professional consulting, board membership, or institutional affiliations influenced the design, execution, analysis, or reporting of this study. The organization that provided baseline data (the knowledge based IT firm) had no role in study design, data analysis, interpretation, or the decision to publish the results. Conflicts of interest The author declare no conflicts of interest. None of the author have any financial, organizational, professional, personal, or intellectual relationships that could directly or indirectly influence the design, execution, analysis, or reporting of the research results or create the appearance of such relationships. This declaration has been prepared with the knowledge and agreement of all co authors, and the corresponding author confirms its accuracy. Author contributions This author contributed to all aspects of the research, including: Conceptualization, Methodology, Data collection, Formal analysis, Software, Validation, Investigation, Resources, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration, and Funding acquisition (if any). The author has read and approved the final manuscript and is solely responsible for the integrity and accuracy of the work. | ||
| کلیدواژهها [English] | ||
| Agent Based Modeling, Cognitive Diversity Preservation, Generative Artificial Intelligence, Positive Human Resource Management, Strategic Collective Intelligence | ||
| مراجع | ||
|
| ||
|
آمار تعداد مشاهده مقاله: 3 |
||