كنفرانس بين المللی مهندسی صنايع و سيستم ها

صفحه اصلی / یازدهمین كنفرانس بين المللی مهندسی صنايع و سيستم ها

The Impact of Individual Entrepreneurial Characteristics on Project Performance: A Machine Learning-Based Predictive Framework

نویسندگان :

Narges Ghobadi¹ Amin Jamili² Mohammad Fattah³

1- دانشگاه صنعتی امیرکبیر 2- دانشگاه تهران 3- دانشگاه علم و صنعت ایران

کلمات کلیدی :

Entrepreneurial Analytics،Individual Entrepreneurial Orientation،Machine Learning،Project Performance Prediction،Entrepreneurial Self-Efficacy،Feature Selection

چکیده :

Project performance prediction remains a critical challenge, as 20–61% of projects fail despite $48 trillion in annual global investments, highlighting a significant research gap. To address this issue, a machine learning framework is developed based on entrepreneurial characteristics. Data from 200 participants are analyzed using five algorithms (LASSO, Ridge, SVR-RBF, SVR-Sigmoid, and Random Forest) to examine the relationships between Individual Entrepreneurial Orientation, Entrepreneurial Self-Efficacy, Entrepreneurial Management Skills, and project outcomes. LASSO regression demonstrates superior predictive performance (MSE = 0.524, r = 0.781), with statistical significance confirmed through Friedman (χ² = 9.12, p < 0.05) and Nemenyi post-hoc tests. Feature selection based on AIC/BIC criteria identifies three critical predictors—Proactiveness & Autonomy, Social Self-Efficacy, and Appearance Self-Efficacy—reducing assessment complexity while maintaining prediction accuracy. The findings suggest that linear models can effectively capture entrepreneurial–performance relationships, challenging assumptions about behavioral complexity in project contexts. The resulting three-feature assessment tool enables practitioners to optimize team composition and predict success probability, providing an evidence-based solution for enhancing project performance in organizations.