A dynamic novel approach for bid/no-bid decision-making.

Area of Science:

• Decision Science
• Artificial Intelligence
• Construction Management

Background:

• Bid/no-bid decision-making is complex and uncertain.
• Existing models may not adequately handle intricate criteria.
• Accurate decision support is crucial in volatile construction markets.

Purpose of the Study:

• To develop an integrated decision support model for bid/no-bid decisions.
• To enhance prediction accuracy and generalization ability in tendering.
• To mitigate risks associated with bid distress.

Main Methods:

• Integration of Rough Sets (RS) for data simplification.
• Application of a General Regression Neural Network (GRNN).
• Optimization of GRNN using Niche Particle Swarm Optimization (NPSO) for parameter tuning.

Main Results:

• The proposed RS and NPSO-GRNN model demonstrated improved prediction accuracy.
• The model showed enhanced generalization ability in real-world case studies.
• The approach effectively simplifies decision samples for the neural network.

Conclusions:

• The NPSO-GRNN algorithm offers significant advantages for tendering decisions.
• The developed decision support system aids managers in making informed bid/no-bid choices.
• This methodology helps prevent financial distress in construction bidding.