Contemporaneous Time Series and Forecasting Methodologies for Predicting Short-Term Productivity

Abstract

Productivity has a profound impact on projects that depend on time and cost of construction operations. In addition, time and cost estimates are derived from productivity. Thus, accurate prediction of productivity is essential to effectively plan and control construction operations. Predicting productivity of ongoing operations, however, is challenging. Due to dynamic and stochastic changes in productivity over time during construction, frequent and regular forecasting of short-term productivity is critical in managing ongoing operations. The present research investigated the characteristics of series of periodic productivity that should be taken into consideration to effectively predict short-term productivity continually and proactively. Given the identified characteristics, this study reviewed a few potential statistical methodologies that can make full use of contemporaneous time series data related to production for the purpose of predicting short-term productivity by using trend analysis. The methodologies were demonstrated in this paper using an example case, through which data processing and modeling procedure for modeling contemporaneous series data were explained.

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 — 34 citations exist upstream.

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