Expert System-Based Cost Predictive Model For Building Works: Neural Network Approach

EXPERT SYSTEM-BASED COST PREDICTIVE MODEL FOR BUILDING WORKS: NEURAL NETWORK APPROACH

Abstract:

The construction industry is inherently complex, involving various parameters and factors that significantly impact project costs. Accurate cost estimation is crucial for successful project management, as cost overruns can lead to financial strain and project delays. Traditional cost estimation methods often lack the precision and adaptability required in today’s dynamic construction environment. This research project proposes the development of an Expert System-Based Cost Predictive Model for Building Works, utilizing the power of Neural Networks for enhanced accuracy and adaptability.

This study aims to address the challenges associated with cost prediction in construction projects by harnessing the capabilities of expert systems and neural networks. The project combines domain knowledge with artificial intelligence techniques to improve the accuracy of cost estimates, making them more reliable and adaptable to changing project conditions. By integrating a neural network-based model within an expert system framework, this research seeks to create a robust predictive tool capable of learning and improving its predictions over time.

The key objectives of this research project are as follows:

1. Data Collection and Preprocessing: Gathering historical data on construction projects, including project parameters, materials, labor, equipment costs, and other relevant factors. Data preprocessing techniques will be applied to clean and normalize the data for use in the neural network.
2. Neural Network Design: Developing a neural network architecture optimized for cost prediction in building works. This includes the selection of appropriate network layers, activation functions, and hyperparameters to maximize prediction accuracy.
3. Expert System Integration: Combining the neural network with an expert system to leverage domain-specific knowledge and heuristics for cost prediction. The expert system will assist in feature selection and refining the network’s predictions.
4. Training and Validation: Training the neural network on the historical data and validating its predictions against real-world construction projects. The model will be fine-tuned and adjusted to improve its accuracy.
5. Adaptability and Learning: Implementing mechanisms within the expert system to allow the predictive model to adapt to changes in project scope, location, and other variables over time. This adaptive capability will enable the model to provide more accurate estimates as it accumulates more data.

The expected outcomes of this research project include a cost predictive model that significantly enhances the accuracy and reliability of cost estimates for building works. This model can assist construction professionals in project planning, budgeting, and risk management, ultimately leading to improved cost control and project success. Additionally, it offers the potential for broader applications in the construction industry and beyond, contributing to the advancement of expert systems and artificial intelligence in cost prediction and project management.

The fusion of expert knowledge with advanced neural network technology offers a promising avenue for revolutionizing the field of construction cost estimation, reducing uncertainties, and improving project outcomes. This research project represents a significant step towards more efficient and data-driven construction project management.

EXPERT SYSTEM-BASED COST PREDICTIVE MODEL FOR BUILDING WORKS: NEURAL NETWORK APPROACH. GET MORE ESTATE MANAGEMENT PROJECT TOPICS AND MATERIALS

Sharing is caring!