In the ever-evolving landscape of transportation infrastructure, one challenge looms large: workforce planning. As lane miles increase and the number of construction engineers and inspectors dwindles, state transportation agencies (STAs) find themselves in a precarious position. Enter Adedolapo Ogungbire, a researcher from the University of Arkansas, who has developed a groundbreaking approach to tackle this issue using machine learning (ML).
Ogungbire, an assistant professor in the Department of Civil Engineering, has published a study in the International Journal of Transportation Science and Technology, which translates to the International Journal of Transportation Science and Engineering. The study, which used the Arkansas Department of Transportation (ARDOT) as a case study, demonstrates how ML models can revolutionize workforce forecasting in the construction industry.
The research focuses on estimating the person-hour requirements of STAs at the project level, a crucial aspect of effective planning. Ogungbire and her team developed and compared various ML regression models, including linear, tree ensembles, kernel-based, and neural network-based models. The results were striking. The random forest regression model, a tree ensemble with bagging, emerged as the top performer, recording a mean R-squared value of 0.91. Other models, such as an ensemble neural network model and linear models, also showed promising results, with R-squared values as high as 0.80 and 0.78, respectively.
“These findings underscore the capability of ML models to provide more accurate workforce demand forecasts for STAs and the construction industry,” Ogungbire stated. This enhanced accuracy in workforce planning can significantly improve resource allocation and management, a boon for any industry, but particularly for the energy sector, where infrastructure projects are often massive and complex.
The implications of this research are far-reaching. As the energy sector continues to expand and evolve, with a growing focus on renewable energy infrastructure, the need for accurate workforce planning becomes even more critical. ML models, as demonstrated by Ogungbire’s study, can provide the precision and foresight needed to manage these projects efficiently.
Moreover, this research paves the way for future developments in predictive modeling and data-driven forecasting. As ML technology continues to advance, we can expect to see even more sophisticated models that can handle increasingly complex datasets and provide more accurate predictions.
For the construction industry, this means better planning, improved resource management, and ultimately, more successful projects. For the energy sector, it means the ability to meet growing demands, adapt to changing technologies, and build a sustainable future. As Ogungbire’s research shows, the future of workforce planning is here, and it’s powered by machine learning.
