| Summary, etc |
The construction industry is known for its demanding pace, labor-intensive work, and inherent risks. Efficient project delivery relies on accurate excavation productivity estimation. While numerous studies have explored factors affecting excavation productivity, there's a need for more research on developing robust and practical estimation models. This study presents a mathematical model for predicting excavation productivity in building construction projects within Tuguegarao City, Cagayan, Philippines. Employing a mixed-methods approach, the research combined a literature review with primary data collection through surveys (82 respondents), site observations, and interviews with project managers, site engineers, and equipment operators. Survey data reliability was validated using Cronbach's alpha (a> 0.70) for all factors. Fifteen factors influencing excavation productivity were identified and ranked using the Relative Importance Index (RII). The top ten factors were then analyzed using multiple linear regression (MLR). The final MLR model, incorporating excavator bucket capacity, limited space, soil characteristics, technical expertise, and thorough project site investigation, explained 90.5% of the variance in actual productivity (R2 = 0.905). The model's predictive capability was further validated using a separate dataset, resulting in a Mean Absolute Percentage Error (MAPE) of 5.758%. The resulting MLR equation (E.P=17.006+20.638(x1) -3.763(x2)-3.479(x3) - 3.085(x4) -2.788(x5), where x1-x5 represent the five key factors) provides a practical tool for predicting excavation productivity, enabling improved project planning, resource allocation, and cost management in the construction industry within the study area. The model's high R-squared value and low MAPE demonstrate its strong predictive capability, despite limitations inherent in the geographical scope of the study. Further research could explore the model's generalizability to other regions and construction contexts.
Keywords: Mathematical model, estimated productivity, excavation operation, multiple linear regression, mean absolute percentage error |