# Digital twin-based intelligent risk assessment and decision support system for university student entrepreneurial projects

## TL;DR

This paper introduces a digital twin-based system to improve risk assessment and decision-making for university student startups, leading to better success rates.

## Contribution

The novel integration of digital twin technology with machine learning for entrepreneurial risk assessment in educational settings.

## Key findings

- The system achieved 94.2% prediction accuracy, outperforming traditional methods.
- It improved project success rates by 23.7% and provided early warnings 22.1 days in advance.
- User satisfaction averaged 4.47 out of 5.0, showing strong acceptance.

## Abstract

University student entrepreneurial ventures face significantly higher failure rates compared to traditional businesses, primarily due to inadequate risk assessment and decision-making challenges. This research develops an innovative digital twin-based intelligent risk assessment and decision support system specifically designed for student entrepreneurial projects. The system integrates digital twin technology with machine learning algorithms to create comprehensive virtual representations of entrepreneurial ventures, enabling real-time risk monitoring, predictive analytics, and intelligent decision recommendations. The proposed framework employs a multi-layered architecture encompassing data acquisition, digital twin modeling, risk assessment engines, and intelligent decision support modules. Experimental validation using 2,847 entrepreneurial projects demonstrates superior performance with 94.2% prediction accuracy, compared to 78.5% for traditional statistical methods and 85.7% for standard machine learning approaches. The system provides early warning capabilities with average lead times of 22.1 days and achieves 23.7% improvement in project success rates. Results indicate significant enhancements in decision-making effectiveness, risk mitigation capabilities, and overall entrepreneurial project outcomes, with user satisfaction scores averaging 4.47 out of 5.0. This research contributes to the theoretical understanding of digital twin applications in entrepreneurial contexts while providing practical solutions for improving student venture success rates through intelligent risk management and decision support.