Search

Construction projects are data-rich environments. However, those data are usually captured for site-specific reasons, e.g., the filing and approval of inspection requests, with little regard to how they can be leveraged for improved project management. Typically, scheduling techniques rely on general probability estimates, which do not capture the details of the site processes causing schedule deviations. This paper illustrates how machine learning techniques can mine project data to forecast delay in the midst of the project. The proposed method uses concrete pouring requests as an example of a site data stream and implements a random forest predictive model to forecast the likelihood of acceptance for these requests. Embedded in the proposed approach is an analysis that allows for the addition of probabilistic time delays associated with the forecast of rejected requests. The methodology was tested on a real-world case study, allowing for the comparison between a project duration estimate based on critical path method (CPM) with static buffers and a project duration obtained using the proposed method. The results show a difference of 10% between the two durations. The paper shows how using data streams from a construction site with machine learning techniques can enhance project duration estimates in execution. [ABSTRACT FROM AUTHOR]

This paper brings forth from the literature a series of learning curve models and evaluates them through the lens of the construction industry. The review suggests that there is still no consensus on which model provides the best fit and predictability for construction data. As such, this paper puts forth a new model that is suitable for the modern construction industry as it accommodates for both mechanization and forgetting. The proposed model is similar to the Wright model (an exponential model of learning), but, through recursion, places more emphasis on recent data. The proposed model shows an error of less than 1% when predicting the cumulative average completion times in three out of four cases examined. [ABSTRACT FROM AUTHOR]

This paper examines the value of utilizing dynamic, revenue-maximizing, congestion pricing on a privately operated tolled route, when the only alternative is a highly congested, public, free-access route. Three methods are specified by which revenue can be maximized-through the selection of a fixed percent of users to set the toll price for, through the selection of a fixed level of service on the toll road, and through the use of a nonlinear optimization model. These methods are tested on a case study involving a highly congested corridor. In simulation, the corridor’s flow rates are observed following the posting of a toll, the toll is updated, and the new flow observed until a subsequent round of revenue maximizing is applied to the new conditions. It was concluded that in a highly congested corridor a dynamic, revenue-maximizing toll can finance the construction of new capacity without degrading social welfare. [ABSTRACT FROM AUTHOR]