Nonlinear Models for Predicting Hoisting Times of Tower Cranes

Abstract

Accuracy in estimating activity duration is one of the key prerequisites for successful construction planning. Efficient material transportation plays an important role in reducing costs and time. Time measurement and work-study techniques can provide good estimation of activity duration, but forming the databank for various conditions is expensive. The use of empirical models has been developed as an alternative to overcome the deficiency while maintaining a reasonable accuracy. In this research traditional linear regression models and nonlinear neural network models have been developed for predicting hoisting times of a tower crane. It is found that nonlinear neural network models can achieve higher accuracy. However, planners may find that the regression models, which describe the relationship between the variables in more simplistic terms, could allow them to shorten the hoisting times by manipulating the input variables. The results and the merits of the models are discussed.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 44 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.