Controlling Downhole Mud Loss using Predictive Data Analytics


Fluid losses during drilling lead to greater expenses from mud loss, difficulty in well control and zonal isolation, and non-productive time (NPT). In this paper, the impacts of each drilling parameter have been studied to understand the significance of each parameter. Hence, a new approach to model volume loss has been developed that is highly effective in lost circulation control in Basra's oil fields. Accurate prediction of volume loss is very important to avoid associated drilling problems such as NPT, cost, formation fracture, fluid influx, kicks, and even blowouts. To date, there is no analytical model to predict the volume loss prior to drilling as a proactive action. The models provide a prediction for mud loss ranging from partial to complete loss in the Dammam Formation. This work presents a new multiple linear regression modeling of volume loss, which makes accurate predictions (maximum error of ± 3%) for a wide range of mud losses. Lost circulation events for more than 300 wells drilled in Basra's oil fields (e.g., South Rumaila, North Rumaila, Zubair, Nahur Umar, and Luhais fields) were accurately identified through summarizing daily drilling reports (DDR), final well reports, and technical reports. Critical drilling parameters such as mud weight (MW), equivalent circulation density (ECD), yield point (Yp), plastic viscosity (PV), rate of penetration (ROP), strokes per minute (SPM), revolutions per minute (RPM), weight on bit (WOB), and bit nozzles were recorded at the time of each mud loss event, and the severity of the mud loss event, depth, and result of any mitigation attempts were also noted. To develop the multiple linear regression models, a huge body of real field data was formulated, and an extensive statistical study was performed. The results were analyzed to formulate a simplified multiple linear regression model that predicts the volume loss conveniently for the thief formation without requiring iterative calculation procedures. The approach has been developed based on analyzing actual mud loss events while drilling this formation, to develop key statistical models for ROP, ECD, and mud losses. These models are then tested with other new wells' data to check their validity and to demonstrate how the models can be used to set key drilling parameters in Basra's oil fields. Finally, this approach that was developed in this study can be used globally if the formation is naturally fractured and has the same characteristics as the Dammam Formation.


Geosciences and Geological and Petroleum Engineering

Research Center/Lab(s)

Center for Research in Energy and Environment (CREE)

Keywords and Phrases

Basra; Iraq; Lost circulation; Volume loss modeling

International Standard Serial Number (ISSN)


Document Type

Article - Journal

Document Version


File Type





© 2019 Begell House Inc., All rights reserved.

Publication Date

01 Oct 2019