Activity codes recorded by drillers are very useful for quantifying invisible lost time (ILT). However,classifying more than 100 activity codes accurately and consistently across various rig operations becomesinfeasible for human operators. We propose an auto-suggestive system that guides the drillers to the correctcodes based on memos they enter into the system. This aims to both eliminate manual classification errorsand improve memo entry. The method for extracting activity codes from memos can be broken into the following steps. The firststep consists of filtering unnecessary text and vectorizing the memos. The vectors are then re-weighted usingthe term frequency-inverse document frequency (TFIDF) statistical measure. Next, data resampling helps tocreate a uniform set of labels for the training data, because there are quite a few important activity codes thatappear infrequently with respect to others. Finally, a classifier is trained. It is shown that the finalized modelcan be used as a real-time auto-suggestive mechanism during the drillers’ data input process. Moreover, itsuse for cleaning up historical datasets is also explored.
展开▼
