Neural Network-Based Predictive Cost Model for Building Works

Project abandonment as a result of forecasting has stimulated in recent times the stakeholders request for project costs implication.However,traditional methods used in the past in cost determination does not often captured intervening variables that impact cost as project progresses, which results in inaccurate cost determination.Therefore paradigm should shift in favor of a system that can accommodate economic variants which usually influence price movement on construction work. It is on this premise that an expert system(neural network) was used in this book to formulate a cost prediction model for building works. Background was created for the comparison of traditional and expert application exploring their methods, concept and application.Artificial Neural Network was used to generate a model which was adjudged to be of very good predictive ability and low Mean Square Error value with different data ports and validation modules. The presentation illustrate practical application of newly emerging field of prediction in neural networking and should be especially useful to construction professionals,cost modelers, cost estimators,technical personnel and project clients at large

Autorentext

Lekan M. Amusan, PhD, MSc, BSc,O.N.D: Studied Building and Construction Management at University of Lagos Akoka,Yaba,Nigeria at Undergraduate and Postgraduate Stage respectively. PhD construction management at Covenant University,Nigeria. Chartered Member of Council of Registered Builders of Nigeria with over 15 years of on-job experience.