PACOTECH™’s data analytics service offering consists of the capability to examine large amounts of data produced during a project’s lifecycle: Planning, Design, Construction, Operation and Maintenance. In addition, data analytics can be applied at the completion of a capital construction project or program to identify lessons learned. We produce predictive models that can help identify negative trends more quickly. Furthermore. we employ artificial Intelligence software that can uncover hidden patterns, correlations and other insights that can be used to gain a better insight into project issues.
For example, our data mining and text mining capabilities enable prediction of cost overruns and schedule delays. We have also developed models combining data available at the time of bidding, like the number of bidders, the engineer’s estimate and the anticipated project schedule with text data describing the type of project being built. Additionally, we have used a text mining technique called sentiment analysis that can be used to examine project communications (records and documents) to quickly identify emerging project issues, trends, and problems.
PACOTECH™ has also employed data analytics as a way of identifying maintenance problem areas for railroad track. We have employed LDA, a type of text mining analysis that groups words, to identify major accident themes of railroad equipment accidents and grade crossing accidents. This type of predictive analytics can be used to predict maintenance problems and to reduce life cycle costs.